Donlin Creek Project
 NI 43-101 Technical Report
Southwest Alaska, U.S.

Effective Date February 5, 2008

Prepared for:
NovaGold Resources Inc.
Vancouver, B.C.

Prepared by:
Kevin Francis, P.Geo.
NovaGold Resources Inc.
Vancouver, B.C.

C O N T E N T S
1.0	SUMMARY		1-1
	1.1	Conclusions and Recommendations	1-3
2.0	INTRODUCTION		2-1
	2.1	Qualified Person	2-1
	2.2	Previous Technical Studies	2-1
	2.3	Technical Report Sections and Required Items under NI 43-101	2-2
3.0	RELIANCE ON OTHER EXPERTS		3-1
	3.1	Mineral Tenure	3-1
	3.2	Surface Rights, Access and Permitting	3-1
	3.3	Environmental	3-1
4.0	PROPERTY DESCRIPTION AND LOCATION		4-1
	4.1	Coordinate System	4-1
	4.2	Mineral Tenure	4-1
	4.3	Agreements and Permits	4-12
	4.4	Environmental	4-13
	4.5	Permits and Process	4-16
5.0	ACCESSIBILITY, CLIMATE, LOCAL RESOURCES, INFRASTRUCTURE AND PHYSIOGRAPHY		5-1
6.0	HISTORY		6-1
	6.1	1996 Activities	6-2
	6.2	1997 Activities	6-3
	6.3	1998 Activities	6-3
	6.4	1999 Activities	6-3
	6.5	2000 Placer Dome	6-4
	6.6	2001 NovaGold	6-4
	6.7	2002 NovaGold	6-4
	6.8	2003 Placer Dome	6-6
	6.9	2004 Placer Dome	6-7
	6.10	2005 Placer Dome	6-7
	6.11	2006 NovaGold	6-7
	6.12	2006–2007 Barrick	6-9
7.0	GEOLOGICAL SETTING		7-1
	7.1	Regional Geology	7-1
	7.2	Property Geology	7-2
	7.3	Deposit Geology	7-5

TOC i

		7.3.1	Sedimentary Stratigraphy	7-6
		7.3.2	Igneous Rocks	7-7
		7.3.3	Structural Geology	7-9
8.0	DEPOSIT TYPES			8-1
9.0	MINERALIZATION			9-1
		9.1.1	Mineralization and Alteration	9-1
		9.1.2	Vein Types	9-3
		9.1.3	Alteration	9-4
		9.1.4	Minor Elements and Deleterious Materials	9-5
		9.1.5	Structural Controls on Mineralization	9-6
10.0	EXPLORATION			10-1
	10.1	2002 NovaGold		10-1
	10.2	2005 Placer Dome		10-1
	10.3	2006–2007 Barrick		10-1
11.0	DRILLING			11-1
	11.1	2002 NovaGold		11-1
	11.2	2005 Placer Dome		11-2
	11.3	2006–2007 Barrick Drilling		11-2
	11.4	Orientation of Mineralization		11-3
12.0	SAMPLING METHOD AND APPROACH			12-1
		12.1.1	Logging	12-1
		12.1.2	Sampling	12-1
13.0	SAMPLE PREPARATION, ANALYSES AND SECURITY			13-1
	13.1	Drill Hole Sample Preparation		13-1
		13.1.1	Prior to 2006	13-1
		13.1.2	2006–2007	13-1
	13.2	Sample Analysis		13-2
	13.3	Donlin Security and Sample Transport		13-2
		13.3.1	Project Site	13-2
		13.3.2	Anchorage Security and Sample Transportation	13-3
	13.4	Assay Quality Assurance and Quality Control (QA/QC)		13-3
		13.4.1	1995–2002 QA/QC Protocol	13-3
		13.4.2	Standard Reference Material and Blank Material in 2005-2006	13-4
		13.4.3	QA/QC 2005 Results	13-5
		13.4.4	QA/QC 2006 and 2007 Results	13-5
	13.5	Specific Gravity Determinations		13-6
14.0	DATA VERIFICATION			14-1
	14.1	Prior to 2005 Campaign		14-1

TOC ii

	14.2	2005 Campaign	14-1
	14.3	2006 and 2007 Drilling Campaigns	14-1
15.0	ADJACENT PROPERTIES		15-1
16.0	MINERAL PROCESSING AND METALLURGICAL TESTING		16-1
	16.1	Mineralogy	16-1
	16.2	Direct Leach / CIL	16-1
	16.3	Crushing / Grinding	16-9
	16.4	Flotation	16-9
	16.5	Pressure Oxidation (POX)	16-10
	16.6	CIL / Gold Recovery	16-11
	16.7	Environmental Considerations	16-11
	16.8	Conceptual Process Plant Design	16-12
17.0	MINERAL RESOURCE AND MINERAL RESERVE ESTIMATES		17-1
	17.1	Introduction	17-1
	17.2	Estimation Approach	17-1
	17.3	Geologic Model	17-1
	17.4	Capping	17-3
	17.5	Specific Gravity Data and Statistics	17-11
	17.6	Assay Statistics	17-12
	17.7	Compositing	17-12
	17.8	Block Model Construction	17-13
	17.9	Definition of Mineralized Envelopes	17-14
		17.9.1  Gold	17-14
		17.9.2  Sulphur	17-14
		17.9.3  Shell Construction	17-15
	17.10 Estimation Parameters		17-16
		17.10.1 Gold Grade Estimation	17-16
		17.10.2 Sulphur Grade Estimation	17-26
		17.10.3 Grade Estimation for Other Elements	17-26
		17.10.4 CO2 , Ca, and Mg Assays	17-28
	17.11 Classification of Waste Rock Management Categories		17-28
	17.12 Variography		17-30
	17.13 Model Validation		17-30
	17.14 Resource Classification and Summary		17-38
		17.14.1 Resource Classification	17-38
		17.14.2 Resource Tabulation	17-38
18.0	OTHER RELEVANT DATA AND INFORMATION		18-1

TOC iii

19.0	ADDITIONAL REQUIREMENTS FOR TECHNICAL REPORTS ON DEVELOPMENT PROPERTIES AND PRODUCTION PROPERTIES	19-1
20.0	INTERPRETATION AND CONCLUSIONS	20-1
21.0	RECOMMENDATIONS	21-1
22.0	REFERENCES	22-1
	22.1  Glossary	22-1
	22.2  Abbreviations and Units of Measure	22-4
23.0	DATE AND SIGNATURE PAGE	23-1

F I G U R E S
Figure 4-1:	Location Map	4-9
Figure 4-2:	Lease Block Map	4-10
Figure 4-3:	Project Location Map	4-11
Figure 7-1:	Regional Geology of Donlin Creek Area	7-1
Figure 7-2:	Main Trend Geology (Piekenbrock and Petsel 2003)	7-4
Figure 7-3:	Aeromagnetic Image showing Interpreted Faults and Intrusive Rocks (source NovaGold)	7-5
Figure 7-4:	General Geology of the Resource Area Showing Intrusive Rock Units and Faults (100 m level projected to surface) (source Barrick)	7-6
Figure 7-5:	Lewis Area Section, Looking Northeasterly Showing Intrusive Rocks, Drill Holes (source Barrick)	7-10
Figure 7-6:	ACMA Area Section, Looking Southeasterly Showing Intrusive Rocks, Drill Holes (source Barrick)	7-11
Figure 9-1:	ACMA Gold Distribution, Looking Southeasterly Showing Intrusive Rocks (colours) and Gold Grade Shells (>1 g/t stipple, >3 g/t pattern) (source Barrick)	9-2
Figure 9-2:	Lewis Gold Distribution Looking Northeasterly Showing Intrusive Rocks (colours) and Gold Grade Shells (>1 g/t stipple, >3 g/t pattern) (source Barrick)	9-2
Figure 9-3:	Major Mineralized Corridors – Donlin Creek Resource Area (Piekenbrock and Petsel, 2003)	9-8
Figure 16-1:	Variability of Flotation Gold Recovery Results for Oxidation Samples	16-10
Figure 17-1:	Raw Au Assays – Cumulative Frequency Distribution of All Rock Types	17-5
Figure 17-2:	Raw Au Assays – Greywacke Cumulative Frequency Distribution Plot	17-5
Figure 17-3:	Raw Au Assays – Shale and Argillite Cumulative Frequency Distribution Plot	17-6
Figure 17-4:	Raw Au Assays – Mafic Dyke Cumulative Frequency Distribution Plot	17-6
Figure 17-5:	Raw Au Assays – Rhyodacite Aphanitic Porphyry Cumulative Frequency Distribution Plot	17-7

TOC iv

Figure 17-6:	Raw Au Assays – Rhyodacite Fine-Grained Porphyry Cumulative Frequency Distribution Plot	17-8
Figure 17-7:	Raw Au Assays – Rhyodacite Porphyry Cumulative Frequency Distribution Plot	17-8
Figure 17-8:	Raw Au Assays – Rhyodacite Coarse-Grained Blue Porphyry Cumulative Frequency Distribution Plot	17-9
Figure 17-9:	Raw Au Assays – Rhyodacite Lath-Rich Porphyry Cumulative Frequency Distribution Plot	17-9
Figure 17-10:	Bench Section Depicting Estimated Measured and Indicated Au Blocks Plotted Against Drill Holes – 0 masl – 50 m Section Width (source Barrick)	17-24
Figure 17-11:	Bench Section Depicting Estimated Measured and Indicated Au Blocks Plotted Against Drill Holes – 100 masl – 50 m Section Width (source Barrick)	17-25
Figure 17-12:	0.25 g/t Au Indicator Variogram for 6 m Composites	17-30
Figure 17-13:	Trend Plot of RCK02_RDA Gold Grades ID3 (au_use) versus NN (pau)	17-34
Figure 17-14:	Correlogram of 6 m Composite Gold Grade for RCK02_RDA Intrusive	17-35
Figure 17-15:	Grade – Tonnage Curve for Gold Estimate for RCK02_RDA Intrusive	17-36

T A B L E S
Table 1-1:	Donlin Creek Project Mineral Resource Summary(1)(2)(3)(4) Effective Date February 5, 2008	1-5
Table 2-1:	Contents Page Headings in Relation to NI 43-101 Prescribed Items—Contents	2-3
Table 4-1:	Donlin Creek Land Leased from Calista	4-2
Table 4-2:	Donlin Creek LLC Mineral Claims	4-2
Table 4-3:	Claims located on Limestone Deposit	4-7
Table 4-4:	Prospecting Sites	4-8
Table 4-5:	Federal Agency Permit and Authorizations	4-17
Table 6-1:	Work History Summary	6-1
Table 6-2:	Mineral Resources - MRDI - January 24, 2002	6-6
Table 6-3:	Mineral Resources - NovaGold - January 19, 2006	6-8
Table 7-1:	Donlin Creek Stratigraphy	7-7
Table 7-2:	Donlin Creek Intrusive Rocks	7-7
Table 9-1:	Results of Stereonet and Scatter Plots of Veins by Resource Domain	9-7
Table 11-1:	Core Holes Drilled in 2006	11-3
Table 16-1:	Summary of Metallurgical / Process Reports on Donlin Creek	16-2
Table 16-2:	Estimated Bond Ball Work Index (BWI)	16-9
Table 16-3:	Summary of Final Flotation Recovery by Geological Domain (without oxide)	16-10
Table 17-1:	Three-Dimensional Solids Used in the Modelling Process	17-1
Table 17-2:	Rock Codes in the Drill Database	17-4

TOC v

Table 17-3:	Donlin Capping	17-10
Table 17-4:	Coefficient of Variance for Au Greater than 0.1 g/t, by Rock Type	17-11
Table 17-5:	Specific Gravity Values by Rock Type	17-12
Table 17-6:	Specific Gravity Values by Grouped Rock Type	17-12
Table 17-7:	Composite Database – Fields Used to Generate Estimates	17-13
Table 17-8:	Block Model Origin and Extent	17-14
Table 17-9:	Discriminator Model Search Parameters	17-15
Table 17-10:	Discriminator Model Sample Constraints	17-16
Table 17-11:	Au Estimation Parameters Internal to Intrusive Indicator	17-18
Table 17-12:	Au Estimation Parameters External to Intrusive Indicator	17-19
Table 17-13:	Au Estimation Parameters Internal to the Shale Indicator	17-20
Table 17-14:	Au Estimation Parameters External to the Shale Indicator	17-21
Table 17-15:	Au Estimation Parameters Internal to the Greywacke Indicator	17-22
Table 17-16:	Au Estimation Parameters External to the Greywacke Indicator	17-23
Table 17-17:	Sulphur Regression Formulae	17-26
Table 17-18:	Arsenic Regression Formulae	17-27
Table 17-19:	Mercury Regression Formulae	17-27
Table 17-20:	Antimony Regression Formulae	17-28
Table 17-21:	ARD Categories	17-29
Table 17-22:	Recommended Waste Rock Management Categories	17-29
Table 17-23:	Comparison of Estimated Gold Grade (au_use) to Nearest Neighbour (pau) Model	17-33
Table 17-24:	Comparison of Estimated Sulphur Grade (su_use) to Nearest Neighbour (psu_use) Model	17-33
Table 17-25:	Donlin Creek Resource Classification	17-38
Table 17-26:	Donlin Creek Project Mineral Resource Summary(1)(2)(3)(4) Effective Date February 5, 2008	17-39

A P P E N D I C E S
Appendix A	Core Logging Manual
Appendix B	6 m Composited Assays Exceeding 4 g/t
Appendix C	2006 – 2007 QA/QC Report
Appendix D	Block Model Swath Plots by Rock Type
Appendix E	Change of Support Curves by Rock Type

TOC vi

1.0 SUMMARY

NovaGold Resources Inc. (NovaGold) has prepared an updated Technical Report that meets the guidelines as outlined by Canadian National Instrument 43-101 for its Donlin Creek property, an early Tertiary age gold-arsenic-antimony-mercury hydrothermal system located in southwest Alaska. NovaGold and Barrick Gold Corporation (Barrick) have 50/50 ownership in the property through their joint ownership of a limited liability company, the Donlin Creek LLC, which will oversee all aspects of project development. Kevin Francis, Manager of Resources for NovaGold and a qualified person as defined by NI 43-101, has reviewed sampling procedures, examined quality assurance/quality control (QA/QC) practices/results, and reviewed Barrick’s estimate of Mineral Resources which includes drilling information through October 3, 2007. In addition to those reviews, the author performed a site visit in July 2006 for the purpose of observing drilling and sampling procedures and to examine drill core and geologic logging practices and, as at the effective date of the report, confirms that there has been no material change in the scientific and technical information about the property since that site visit.

The Donlin Creek project is located near the of the Kuskokwim River about 25 km north of the village of Crooked Creek and approximately 70 km northeast of Aniak, a regional hub. The latitude and longitude of the deposit is approximately 62^o 01’ N and 158^o 12’ W.

The gold-bearing deposits have been combined into two resource areas: ACMA (containing ACMA proper, 400 Zone, Aurora and Akivik deposits) and Lewis (comprising North and South Lewis, Vortex, Rochelieu and Queen deposits). The project is serviced by commercial air services out of Anchorage and Aniak and by a 25 km long winter road from the town of Crooked Creek.

The property consists of 109 km² (42 mi², 10,900 ha) of privately owned Native Alaskan land. Calista Corporation (Calista), a regional Native corporation, owns the subsurface rights, and The Kuskokwim Corporation (TKC), a village corporation, owns the surface rights. Placer Dome acquired a 20-year lease from Calista effective May 1, 1995. Annual property payments are US$200,000 through the end of feasibility and increasing to US$500,000 per annum once a feasibility study is completed. Calista holds a retained net royalty of 1.5% until payback of capital, increasing to 4.5% thereafter. Placer Dome exercised its back-in right and assumed management of the continued development of the Donlin Creek project in 2003. Barrick acquired Placer Dome in February 2006. Barrick and NovaGold formed a joint venture to develop the property pursuant to a joint venture agreement dated November 13, 2002. The NovaGold and Barrick joint venture was converted to a limited liability company, the Donlin Creek LLC, with NovaGold and Barrick each holding a 50/50 interest therein.

Page 1-1

Upon submission of a feasibility study, Calista retains a 90-day back-in right to participate in the project at a level of 5% to 15% by committing to contribute its share of capital. Their share would be divided pro rata from Barrick and NovaGold.

The Donlin Creek project geology consists of flysch sequence sedimentary units of the Cretaceous Kuskokwim Group intruded by Late Cretaceous to early Tertiary felsic intrusive rocks. The sediments consist of interbedded greywacke, shale and siltstone. Greywacke is dominant (Lewis resource area), but shale-rich areas also occur (ACMA resource area). The overall bedding strikes NW and dips 10° to 50° southwest. The intrusive units consist of porphyritic rhyodacite and rhyolite and lesser mafic dykes and sills. Sills are common in the ACMA and southern Lewis areas, whereas dykes dominate in the North Lewis area. The dykes and sills range from a few meters to more than 60 m in width.

Mineralization is best developed in the felsic intrusive rocks, with lesser mineralization in sediments (principally in the greywacke units). It is structurally and lithologically controlled along NNE-trending extensional fault/fracture zones and best developed where those zones intersect favourable host lithologies such as the felsic intrusive dykes and sills and greywacke. Gold mineralization is associated with quartz, carbonate and sulphide (pyrite, arsenopyrite and stibnite) vein and veinlet networks (dominant) as well as disseminated in favourable host rocks typically adjacent to veins (subdominant). The gold occurs primarily in the lattice structure of arsenopyrite. Realgar, native arsenic and stibnite can be found generally associated with the higher-grade gold mineralization.

Two sets of similar protocols were used for the samples that formed the basis of the Lewis and ACMA mineral resource model. Prior to 2002, most of the samples from Placer Dome’s work were processed in their own laboratory. NovaGold’s samples were processed by Bondar-Clegg (now ALS Chemex), a commercial laboratory. The 2006 and 2007 samples were assayed at ALS Chemex. The results can be evaluated together because the Standard Reference Material (SRM), the blank material and the duplicate protocol were the same. The performance of each SRM was within acceptable limits and showed that the overall assay process was in control for the work done. Good reproducibility of the gold values is demonstrated. The blank sample program worked well and demonstrated negligible contamination in the assay process.

The database used to estimate the mineral resources consists of samples and geological information from 1,578 drill holes and trenches. Since the last technical report in January 2006, Barrick has drilled 359 holes totalling 135,362 m (444,101 ft) of geologically logged and assayed core within the resource area. Assay data transfer to the resource database was validated from electronic assay certificates.

Page 1-2

	Metallurgical testwork, under the direction of Barrick, appears to have been completed in sufficient detail to support a feasibility level study.
	Gold is mainly carried by arsenopyrite.
	Variation is observed in processing behaviour between intrusives and sediments, but less so between the geographical sources. Concentration by flotation is efficient, being 91 to 97% for intrusives and 82 to 89% for sediments. Generally, direct cyanidation yields less than 10% gold recovery; whereas oxidation of the sulphides prior to cyanidation yields recoveries exceeding 90% for intrusives and 78 to 89% for the sediments.
	Accordingly, process testing has been directed towards development of the following conceptual flowsheet:
	•	concentration by flotation using nitrogen
	•	high pressure oxidation in an autoclave
	•	carbon-in-leach (CIL) cyanidation of the concentrate
	•	carbon strip and regeneration circuits
	•	gold electrowinning, and
	•	refining and production of doré bars
	This processing concept incorporates proven commercial unit operations. No issues have been identified to date that might lead to economic performance of this sequence that would be substantially different from similar processes in commercial operation today.
	Presently there is a 160-person exploration camp on site. The exploration drilling for 2008 has commenced and, as of the date of this report, two core rigs are operating. Exploration is focussed on step out drilling in the East ACMA area testing for additional fault/fracture zones in the favourable intrusive rock type.
	There is no history of significant development or hard rock production. Minor placer gold production has taken place within the project area.
1.1	Conclusions and Recommendations
	Donlin Creek is a large-scale gold deposit with a long history of successful exploration. Presently the deposit is open in all directions and is still the subject of an extensive exploration drilling program. The resource estimate prepared by Barrick has been completed using generally-accepted methodology.

Page 1-3

The mineral resource estimates for the Donlin Creek project were calculated by Barrick. The estimates were made from 3-dimensional (3D) block models utilizing Vulcan™ mine planning software. Industry-accepted methods were used to create interpolation domains based on mineralized geology, and grade estimation based on inverse distance to a power cubed (ID3). Acceptable mineralized envelopes were defined through probability-assisted domaining. This method limited the waste intervals of the intrusive units at ACMA from diluting the grades in the mineralized regions and honoured the significant contribution of greywacke-hosted mineralization together with mineralized felsic intrusive units at Lewis. Extremely high gold assays were capped prior to compositing into 6 m lengths.

Reasonableness of grade interpolation was reviewed by visual inspection of sections and plans displaying block model grades, drill hole composites and geology. Good agreement was observed. Global and local bias checks in block models, using nearest-neighbour estimated values versus the ID3 gold grade estimate, found no evidence of bias. Change of support measurements relative to a 6 m by 6 m by 6 m selective mining unit (SMU) indicate the grade estimate is not smooth enough to be used for mine planning without prior adjustment.

The logic for mineral resource classification of ACMA and Lewis was consistent with the CIM definitions referred to in National Instrument 43-101 (NI 43-101). The indicated mineral resource category is supported by the present drilling grids over the ACMA and Lewis deposits (nominal 25 m to 35 m). The measured mineral resource category is supported only in blocks pierced by exploration drill holes. Inferred mineralization is limited to a reasonable expectation of mining by a conceptual open pit shell using a metal price of US$650 per ounce of gold and recent estimates of mining, geotechnical and metallurgical parameters.

The mineralization of the Donlin Creek project effective date as of February 5, 2008, is classified as measured, indicated and inferred mineral resources. The classified mineral resources are shown in Table 1-1. The mineral resource has been constrained within a conceptual pit based on US$650 per ounce of gold and using recent estimates of mining, geotechnical and metallurgical parameters. Barrick’s Technical Services Evaluations Group estimated a variable net smelter return (NSR) cut-off grade based on recent estimates of mining costs, processing costs (dependent upon sulphur content), selling costs and royalties, rather than gold grade alone. The NSR cut-off equates to approximately 0.8 g/t gold at the average estimated sulphur grade of 1.12% .

The mineral resource estimates for Donlin Creek project show an increase in resources over NovaGold’s September 2006 mineral resource estimates. This increase is the result of additional drilling completed during 2006 and 2007.

Page 1-4

Table 1-1:	Donlin Creek Project Mineral Resource Summary(1)(2)(3)(4)
	Effective Date February 5, 2008

	Tonnes	Au	Contained Au
	(M)	(g/t)	(Million oz)
US$0.01 NSR/t Cut-off
Measured Mineral Resource	4.3	2.73	0.38
Indicated Mineral Resource	367.4	2.46	29.0
Measured + Indicated Mineral Resources	371.7	2.46	29.38
Inferred Mineral Resource	46.5	2.31	3.46

⁽¹⁾ Mineral resources that are not mineral reserves do not have demonstrated economic viability
⁽²⁾ Rounding differences may occur.
⁽³⁾ Resources are constrained within a Lerchs-Grossman (LG) open-pit shell using the long-term metal price assumption of US$650/oz of gold. Assumptions for the LG shell included pit slopes variable by sector and pit area: mining cost is variable with depth, averaging US$1.57/t mined; process cost is calculated as the percent sulfur grade x US$2.09 + US$10.91; general and administrative costs, gold selling cost and sustaining capital cost are reflected on a per tonne basis. Average sulphur content is 1.12% . Based on metallurgical testing, gold recovery is assumed to be 89.5% . Blocks with a cost margin of US$0.01/t or higher above the variable cut-off were reported.
⁽⁴⁾ Waste blocks within the open-pit shell surrounded by blocks above cut-off are included in resource estimate. Blocks above cut-off within the open-pit shell surrounded by blocks of waste are excluded from resource estimate.

NovaGold recommends that Donlin Creek LLC continue to advance the project toward a feasibility study.

Page 1-5

2.0	INTRODUCTION
	NovaGold Resources Inc. (NovaGold) is completing this Technical Report to update the resource estimate for the Donlin Creek project in Alaska. Barrick has provided NovaGold with a new resource estimate containing drilling data through October 2007. Resource estimates were previously reported in Technical Reports dated February and March 2002 (Juras, 2002 and Juras and Hodgson, 2002), January 2006 (Dodd, 2006) and September 2006 (Dodd et al., 2006). Kevin Francis, P.Geo., an employee of NovaGold, serves as the non-independent Qualified Person (QP) responsible for the preparation of this Technical Report as defined in National Instrument 43-101, Standards of Disclosure for Mineral Projects, and in compliance with Form 43-101F1 (the Technical Report). A non-independent qualified person is permitted under NI 43- 101 because the Donlin Creek project is a joint venture between a producer (Barrick) and a non-producer (NovaGold).
	NovaGold relied on scientific and technical information prepared by or under the supervision of a qualified person of Barrick relating to the preparation of resource estimates in this Technical Report.
	The work entailed review of pertinent geological data in sufficient detail to prepare the Technical Report.
2.1	Qualified Person
	Kevin Francis, Manager of Resources and an employee of NovaGold Resources Inc., served as the QP responsible for the preparation of the report as defined in National Instrument 43-101, Standards of Disclosure for Mineral Projects, and in compliance with Form 43-101F1 (the Technical Report).
	The QP visited the Donlin Creek project on July 10, 2006 during which time additional background data were reviewed and, as at the effective date of the report, confirms that there has been no material change in the scientific and technical information about the property since that site visit.
2.2	Previous Technical Studies
	Technical reports completed for the project and on file at www.sedar.com include:
	Juras, S., 2002: Technical Report, Donlin Creek Project, Alaska, unpublished NI43- 101F1 Technical Report to NovaGold Resources Inc. by MRDI, effective date January 24, 2002 .

Page 2-1

	Juras, S. and Hodgson, S., 2002: Technical Report, Preliminary Assessment, Donlin Creek Project, Alaska, unpublished NI43-101F1 Technical Report to NovaGold Resources Inc. by MRDI, report date March 2002.
	Dodd, S., 2006: Donlin Creek Project 43-101 Technical Report, unpublished NI43- 101F1 Technical Report to NovaGold Resources Inc. by NovaGold Resources Inc., effective date January 19, 2006
	Dodd, S., Francis, K. and Doerksen, G., 2006: Preliminary Assessment Donlin Creek Gold Project Alaska, USA, unpublished NI43-101Fi Technical Report to NovaGold Resources Inc. by SRK Consulting (US), Inc., effective date September 20, 2006
2.3	Technical Report Sections and Required Items under NI 43-101
	Table 2-1 relates the sections as shown in the contents page of this report to the Prescribed Items Contents Page of NI 43-101. The main differences are that Item 25 “Additional Requirements for Technical Reports on Development Properties and Production Properties” is incorporated into the main body of the report, immediately following Item 18, “Other Relevant Data and Information”, and that all illustrations (Item 26, “Illustrations”) are included in the body of the report immediately following the text citation of the appropriate illustration.

Page 2-2

Table 2-1: Contents Page Headings in Relation to NI 43-101 Prescribed Items—Contents

NI 43-101	NI 43-101 Heading	Report	Report Section Heading
Item Number		Section
		Number
Item 1	Title Page		Cover page of report
Item 2	Table of Contents		Table of contents
Item 3	Summary	Section 1	Summary
Item 4	Introduction	Section 2	Introduction
Item 5	Reliance on Other Experts	Section 3	Reliance on Other Experts
Item 6	Property Description and Location	Section 4	Property Description and Location
Item 7	Accessibility, Climate, Local Resources,	Section 5	Accessibility, Climate, Local
	Infrastructure and Physiography		Resources, Infrastructure and
			Physiography
Item 8	History	Section 6	History
Item 9	Geological Setting	Section 7	Geological Setting
Item 10	Deposit Types	Section 8	Deposit Types
Item 11	Mineralization	Section 9	Mineralization
Item 12	Exploration	Section 10	Exploration
Item 13	Drilling	Section 11	Drilling
Item 14	Sampling Method and Approach	Section 12	Sampling Method and Approach
Item 15	Sample Preparation, Analyses and Security	Section 13	Sample Preparation, Analyses and
			Security
Item 16	Data Verification	Section 14	Data Verification
Item 17	Adjacent Properties	Section 15	Adjacent Properties
Item 18:	Mineral Processing and Metallurgical Testing	Section 16	Mineral Processing and
			Metallurgical Testing
Item 19	Mineral Resource and Mineral Reserve	Section 17	Mineral Resource and Mineral
	Estimates		Reserve Estimates
Item 20	Other Relevant Data and Information	Section 18	Other Relevant Data and
			Information
Item 21	Interpretation and Conclusions	Section 20	Interpretation and Conclusions
Item 22	Recommendations	Section 21	Recommendations
Item 23	References	Section 22	References
Item 24	Date and Signature Page	Section 23	Date and Signature Page
Item 25	Additional Requirements for Technical Reports	Section 19	Additional Requirements for
	on Development Properties and Production		Technical Reports on Development
	Properties		Properties and Production
			Properties
Item 26	Illustrations		Incorporated in report under
			appropriate section number,
			immediately after first citation in text

Page 2-3

3.0	RELIANCE ON OTHER EXPERTS
	Placer Dome assumed project management in 2003, which was subsequently transferred to Barrick as a result of their takeover of Placer Dome in February 2006. NovaGold relies upon Barrick to have collected and analysed drill hole samples, directed metallurgical testing and constructed the resource estimate using accepted industry practice by, or under the supervision of, qualified persons.
3.1	Mineral Tenure
	The QP has not reviewed the mineral tenure, nor independently verified the legal status or ownership of the project area or underlying property agreements. The QP has relied upon Barrick for this information.
3.2	Surface Rights, Access and Permitting
	The QP has relied upon Barrick for information regarding Surface Rights, Road Access and Permits, including the status of the granting of surface rights for land designated for mining, milling, dumps and tailings impoundments.
3.3	Environmental
	The QP has relied relied upon Barrick to provide the environmental status for the project.

Page 3-1

4.0	PROPERTY DESCRIPTION AND LOCATION
	Donlin Creek is located in southwest Alaska in the United States of America, approximately 70 km (44 mi) northeast of Aniak, a regional hub (see Figure 4-1). The property consists of 109 km2 (42 mi2, 10,858 hectares) of privately owned Native Alaskan land. Calista Corporation (Calista), a regional Native corporation, owns the subsurface rights, and The Kuskokwim Corporation (TKC), a village corporation, owns the surface rights. The resource areas are within T. 23 N., R. 49. W. (see Figure 4-2), Seward Meridian, Kuskokwim and Mt. McKinley Recording Districts, Crook Creek Mining District, Iditarod A-5 USGS 1:63,360 topography map. These areas consist of the ACMA and 400 Zone, Aurora and Akivik prospects (grouped as ACMA) and the Lewis, South Lewis, Vortex, Rochelieu and Queen prospects (grouped as Lewis) (see Figure 4-3).
4.1	Coordinate System
	The Donlin Creek project uses Universal Transverse Mercator (UTM) Zone 4 (metres). The map datum is NAD83.
4.2	Mineral Tenure
	The land status of the Donlin Creek area is shown in Figure 4-2. Most of the rights (surface and subsurface) are governed by conditions defined by the Alaska Native Claims Settlement Act (ANCSA). Section 12(a) of ANCSA entitled each village corporation to select surface estate land from an area proximal to the village in an amount established by its population size. Calista receives conveyance of the subsurface when the surface estate in those lands is conveyed to the village corporation. Section 12(b) of ANCSA allocated a smaller entitlement to the regional corporations with the requirement they reallocate it to their villages as they choose. Calista receives subsurface estate when its villages receive 12(b) lands. Calista reallocated its 12(b) entitlement in 1999 according to a formula based on original village corporation enrolments.
	The Donlin Creek exploration and mining lease currently includes a total of 42 contiguous sections leased from Calista (Table 4-1), which holds the subsurface (mineral) estate for Native-owned lands in the region. The leased land is believed to contain 10,858 hectares (26,830 acres). Title to all of these sections has been conveyed to Calista by the Federal Government. Calista owns the surface estate on 9 of these 42 sections. A separate Surface Use Agreement with TKC, which owns the surface estate of the remaining 33 sections, grants non-exclusive surface use rights to the Donlin Creek LLC. All of these sections have now been conveyed to Calista/TKC

Page 4-1

by the Federal Government. Figure 4-2 shows the lease block. Figure 4-3 shows the resource area and prospect names.

Table 4-1: Donlin Creek Land Leased from Calista

Township (Twp)	Range (Rng)	Sections (Sec)
22 North	48 West	5 & 6
22 North	49 West	1, 2 , 3, 10 and 11
23 North	48 West	5, 6, 7, 8, 16, 17, 18, 19, 20, 21, 28, 29, 30, 31, 32, 33
23 North	49 West	1, 10, 11, 12, 14, 15, 21, 22, 23, 24, 25, 26, 27, 28, 33, 34, 35, 36

In addition to the leased land, Donlin Creek LLC holds 176 unpatented mineral claims (Table 4-2) comprising 8,968 hectares (22,160 acres) primarily surrounding the leased land in the Kuskokwim and Mt. McKinley recording districts. Of these, 32 claims are tentatively approved (T.A.) for conveyance from the Federal to State government subject to official surveying. The remaining 144 claims are state-selected (S.S.). These claims have not been legally surveyed. All claims are either 16.2 hectares (40 acres), 32.4 hectares (80 acres) or 64.8 hectares (160 acres) in size.

Table 4-2: Donlin Creek LLC Mineral Claims

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
578768	DNC # 1	T.A.	08/14/07	16.2	24N	47W	31
578769	DNC # 2	T.A.	08/14/07	16.2	24N	47W	31
578770	DNC # 3	T.A.	07/10/07	16.2	24N	48W	36
578771	DNC # 4	T.A.	07/10/07	16.2	24N	48W	36
578772	DNC # 5	T.A.	07/10/07	16.2	24N	48W	36
578773	DNC # 6	T.A.	07/10/07	16.2	24N	48W	36
578774	DNC # 7	T.A.	07/10/07	16.2	24N	48W	35
578775	DNC # 8	T.A.	07/10/07	16.2	24N	48W	35
578776	DNC # 9	T.A.	07/10/07	16.2	24N	48W	35
578777	DNC # 10	T.A.	07/10/07	16.2	24N	48W	35
578778	DNC # 11	T.A.	07/10/07	16.2	24N	48W	36
578779	DNC # 12	T.A.	07/10/07	16.2	24N	48W	36
578780	DNC # 13	T.A.	07/10/07	16.2	24N	48W	36
578781	DNC # 14	T.A.	07/10/07	16.2	24N	48W	36
578782	DNC # 15	T.A.	08/14/07	16.2	24N	47W	31
578783	DNC # 16	T.A.	08/14/07	16.2	24N	47W	31
578784	DNC # 17	T.A.	08/14/07	16.2	24N	47W	31
578785	DNC # 18	T.A.	08/14/07	16.2	24N	47W	31
578786	DNC # 19	T.A.	07/10/07	16.2	24N	48W	36
578787	DNC # 20	T.A.	07/10/07	16.2	24N	48W	36
578788	DNC # 21	T.A.	07/10/07	16.2	24N	48W	36

Page 4-2

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
578789	DNC # 22	T.A.	07/10/07	16.2	24N	48W	36
578790	DNC # 23	T.A.	07/10/07	16.2	24N	48W	35
578791	DNC # 24	T.A.	07/10/07	16.2	24N	48W	35
578792	DNC # 25	T.A.	07/10/07	16.2	24N	48W	35
578793	DNC # 26	T.A.	07/10/07	16.2	24N	48W	35
578794	DNC # 27	T.A.	07/10/07	16.2	24N	48W	36
578795	DNC # 28	T.A.	07/10/07	16.2	24N	48W	36
578796	DNC # 29	T.A.	07/10/07	16.2	24N	48W	36
578797	DNC # 30	T.A.	07/10/07	16.2	24N	48W	36
578798	DNC # 31	T.A.	08/14/07	16.2	24N	47W	31
578799	DNC # 32	T.A.	08/14/07	16.2	24N	47W	31
578800	DNC # 33	S.S.	09/27/96	16.2	23N	48W	1
578801	DNC # 34	S.S.	09/27/96	16.2	23N	48W	1
578802	DNC # 35	S.S.	09/27/96	16.2	23N	48W	2
578803	DNC # 36	S.S.	09/27/96	16.2	23N	48W	2
578804	DNC # 37	S.S.	09/27/96	16.2	23N	48W	2
578805	DNC # 38	S.S.	09/27/96	16.2	23N	48W	2
578806	DNC # 39	S.S.	09/27/96	16.2	23N	48W	2
578807	DNC # 40	S.S.	09/27/96	16.2	23N	48W	2
578808	DNC # 41	S.S.	09/27/96	16.2	23N	48W	1
578809	DNC # 42	S.S.	09/27/96	16.2	23N	48W	1
578810	DNC # 43	S.S.	09/27/96	16.2	23N	48W	2
578811	DNC # 44	S.S.	09/27/96	16.2	23N	48W	2
578812	DNC # 45	S.S.	09/27/96	16.2	23N	48W	2
578813	DNC # 46	S.S.	09/27/96	16.2	23N	48W	2
578814	DNC # 47	S.S.	09/27/96	16.2	23N	48W	2
578815	DNC # 48	S.S.	09/27/96	16.2	23N	48W	2
578816	DNC # 49	S.S.	09/27/96	16.2	23N	48W	11
578817	DNC # 50	S.S.	09/27/96	16.2	23N	48W	11
644952	GROUSE 1	S.S.	05/04/04	64.8	23N	50W	35
644952	GROUSE 1	S.S.	05/04/04	64.8	23N	50W	35
644953	GROUSE 2	S.S.	05/04/04	64.8	23N	50W	35
644954	GROUSE 3	S.S.	05/04/04	64.8	23N	50W	36
644955	GROUSE 4	S.S.	05/04/04	64.8	23N	50W	36
644956	GROUSE 5	S.S.	05/04/04	64.8	23N	49W	31
644957	GROUSE 6	S.S.	05/04/04	64.8	23N	49W	31
644958	GROUSE 7	S.S.	05/04/04	64.8	23N	49W	32
644959	GROUSE 8	S.S.	05/03/04	64.8	23N	49W	32
644960	GROUSE 9	S.S.	05/04/04	64.8	23N	50W	35
644960	GROUSE 9	S.S.	05/04/04	64.8	23N	50W	35

Page 4-3

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
644961	GROUSE 10	S.S.	05/04/04	64.8	23N	50W	35
644962	GROUSE 11	S.S.	05/04/04	64.8	23N	50W	36
644963	GROUSE 12	S.S.	05/04/04	64.8	23N	50W	36
644964	GROUSE 13	S.S.	05/04/04	64.8	23N	49W	31
644965	GROUSE 14	S.S.	05/04/04	64.8	23N	49W	31
644966	GROUSE 15	S.S.	05/04/04	64.8	23N	49W	32
644967	GROUSE 16	S.S.	05/03/04	64.8	23N	49W	32
644968	GROUSE 17	S.S.	05/04/04	64.8	23N	50W	26
644968	GROUSE 17	S.S.	05/04/04	64.8	23N	50W	26
644969	GROUSE 18	S.S.	05/04/04	64.8	23N	50W	26
644970	GROUSE 19	S.S.	05/04/04	64.8	23N	50W	25
644971	GROUSE 20	S.S.	05/04/04	64.8	23N	50W	25
644972	GROUSE 21	S.S.	05/04/04	64.8	23N	49W	30
644973	GROUSE 22	S.S.	05/04/04	64.8	23N	49W	30
644974	GROUSE 23	S.S.	05/04/04	64.8	23N	49W	29
644975	GROUSE 24	S.S.	05/03/04	64.8	23N	49W	29
644976	GROUSE 25	S.S.	05/04/04	64.8	23N	50W	26
644976	GROUSE 25	S.S.	05/04/04	64.8	23N	50W	26
644977	GROUSE 26	S.S.	05/04/04	64.8	23N	50W	26
644978	GROUSE 27	S.S.	05/04/04	64.8	23N	50W	25
644979	GROUSE 28	S.S.	05/04/04	64.8	23N	50W	25
644980	GROUSE 29	S.S.	05/04/04	64.8	23N	49W	30
644981	GROUSE 30	S.S.	05/04/04	64.8	23N	49W	30
644982	GROUSE 31	S.S.	05/04/04	64.8	23N	49W	29
644983	GROUSE 32	S.S.	05/03/04	64.8	23N	49W	29
644984	GROUSE 33	S.S.	05/04/04	64.8	23N	50W	23
644984	GROUSE 33	S.S.	05/04/04	64.8	23N	50W	23
644985	GROUSE 34	S.S.	05/04/04	64.8	23N	50W	23
644985	GROUSE 34	S.S.	05/04/04	64.8	23N	50W	23
644986	GROUSE 35	S.S.	05/04/04	64.8	23N	50W	24
644987	GROUSE 36	S.S.	05/04/04	64.8	23N	50W	24
644988	GROUSE 37	S.S.	05/04/04	64.8	23N	49W	19
644989	GROUSE 38	S.S.	05/04/04	64.8	23N	49W	19
644990	GROUSE 39	S.S.	05/04/04	64.8	23N	49W	20
644991	GROUSE 40	S.S.	05/03/04	64.8	23N	49W	20
644992	GROUSE 41	S.S.	05/04/04	64.8	23N	50W	23
644993	GROUSE 42	S.S.	05/04/04	64.8	23N	50W	23

Page 4-4

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
644993	GROUSE 42	S.S.	05/04/04	64.8	23N	50W	23
644994	GROUSE 43	S.S.	05/04/04	64.8	23N	50W	24
644994	GROUSE 43	S.S.	05/04/04	64.8	23N	50W	24
644995	GROUSE 44	S.S.	05/04/04	64.8	23N	50W	24
644996	GROUSE 45	S.S.	05/04/04	64.8	23N	49W	19
644997	GROUSE 46	S.S.	05/04/04	64.8	23N	49W	19
644998	GROUSE 47	S.S.	05/04/04	64.8	23N	49W	20
644999	GROUSE 48	S.S.	05/03/04	64.8	23N	49W	20
645000	GROUSE 49	S.S.	05/04/04	64.8	23N	50W	14
645001	GROUSE 50	S.S.	05/04/04	64.8	23N	50W	14
645002	GROUSE 51	S.S.	05/04/04	64.8	23N	50W	13
645002	GROUSE 51	S.S.	05/04/04	64.8	23N	50W	13
645003	GROUSE 52	S.S.	05/04/04	64.8	23N	50W	13
645003	GROUSE 52	S.S.	05/04/04	64.8	23N	50W	13
645004	GROUSE 53	S.S.	05/04/04	64.8	23N	49W	18
645004	GROUSE 53	S.S.	05/04/04	64.8	23N	49W	18
645005	GROUSE 54	S.S.	05/04/04	64.8	23N	49W	18
645006	GROUSE 55	S.S.	05/04/04	64.8	23N	49W	17
645007	GROUSE 56	S.S.	05/03/04	64.8	23N	49W	17
645008	GROUSE 57	S.S.	05/03/04	64.8	23N	49W	16
645009	GROUSE 58	S.S.	05/03/04	64.8	23N	49W	16
645010	GROUSE 59	S.S.	05/04/04	64.8	23N	50W	14
645011	GROUSE 60	S.S.	05/04/04	64.8	23N	50W	14
645012	GROUSE 61	S.S.	05/04/04	64.8	23N	50W	13
645013	GROUSE 62	S.S.	05/04/04	64.8	23N	50W	13
645014	GROUSE 63	S.S.	05/04/04	64.8	23N	49W	18
645014	GROUSE 63	S.S.	05/04/04	64.8	23N	49W	18
645015	GROUSE 64	S.S.	05/04/04	64.8	23N	49W	18
645016	GROUSE 65	S.S.	05/04/04	64.8	23N	49W	17
645017	GROUSE 66	S.S.	05/03/04	64.8	23N	49W	17
645018	GROUSE 67	S.S.	05/03/04	64.8	23N	49W	16
645019	GROUSE 68	S.S.	05/03/04	64.8	23N	49W	16
645020	GROUSE 69	S.S.	05/04/04	64.8	23N	50W	11
645021	GROUSE 70	S.S.	05/04/04	64.8	23N	50W	11
645022	GROUSE 71	S.S.	05/04/04	64.8	23N	50W	12
645023	GROUSE 72	S.S.	05/04/04	64.8	23N	50W	12
645024	GROUSE 73	S.S.	05/04/04	64.8	23N	49W	7

Page 4-5

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
645024	GROUSE 73	S.S.	05/04/04	64.8	23N	49W	7
645025	GROUSE 74	S.S.	05/04/04	64.8	23N	49W	7
645025	GROUSE 74	S.S.	05/04/04	64.8	23N	49W	7
645026	GROUSE 75	S.S.	05/04/04	64.8	23N	49W	8
645027	GROUSE 76	S.S.	05/03/04	64.8	23N	49W	8
645028	GROUSE 77	S.S.	05/03/04	64.8	23N	49W	9
645029	GROUSE 78	S.S.	05/03/04	64.8	23N	49W	9
645030	GROUSE 79	S.S.	05/04/04	64.8	23N	50W	11
645031	GROUSE 80	S.S.	05/04/04	64.8	23N	50W	11
645032	GROUSE 81	S.S.	05/04/04	64.8	23N	50W	12
645033	GROUSE 82	S.S.	05/04/04	64.8	23N	50W	12
645034	GROUSE 83	S.S.	05/04/04	64.8	23N	49W	7
645035	GROUSE 84	S.S.	05/04/04	64.8	23N	49W	7
645035	GROUSE 84	S.S.	05/04/04	64.8	23N	49W	7
645036	GROUSE 85	S.S.	05/04/04	64.8	23N	49W	8
645036	GROUSE 85	S.S.	05/04/04	64.8	23N	49W	8
645037	GROUSE 86	S.S.	05/03/04	64.8	23N	49W	8
645038	GROUSE 87	S.S.	05/03/04	64.8	23N	49W	9
645039	GROUSE 88	S.S.	05/03/04	64.8	23N	49W	9
645040	GROUSE 89	S.S.	05/04/04	64.8	23N	50W	2
645041	GROUSE 90	S.S.	05/04/04	64.8	23N	50W	2
645042	GROUSE 91	S.S.	05/04/04	64.8	23N	50W	1
645043	GROUSE 92	S.S.	05/04/04	64.8	23N	50W	1
645044	GROUSE 93	S.S.	05/04/04	64.8	23N	49W	6
645045	GROUSE 94	S.S.	05/04/04	64.8	23N	49W	6
645046	GROUSE 95	S.S.	05/04/04	64.8	23N	49W	5
645046	GROUSE 95	S.S.	05/04/04	64.8	23N	49W	5
645047	GROUSE 96	S.S.	05/03/04	64.8	23N	49W	5
645048	GROUSE 97	S.S.	05/03/04	64.8	23N	49W	4
645049	GROUSE 98	S.S.	05/03/04	64.8	23N	49W	4
645050	GROUSE 99	S.S.	05/04/04	64.8	23N	50W	2
645051	GROUSE 100	S.S.	05/04/04	64.8	23N	50W	2
645052	GROUSE 101	S.S.	05/04/04	64.8	23N	50W	1
645053	GROUSE 102	S.S.	05/04/04	64.8	23N	50W	1
645054	GROUSE 103	S.S.	05/04/04	64.8	23N	49W	6
645055	GROUSE 104	S.S.	05/04/04	64.8	23N	49W	6
645056	GROUSE 105	S.S.	05/04/04	64.8	23N	49W	5

Page 4-6

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
645056	GROUSE 105	S.S.	05/04/04	64.8	23N	49W	5
645057	GROUSE 106	S.S.	05/03/04	64.8	23N	49W	5
645058	GROUSE 107	S.S.	05/03/04	64.8	23N	49W	4
645059	GROUSE 108	S.S.	05/03/04	64.8	23N	49W	4

Donlin Creek LLC also holds 28 claims comprising 1,813 hectares (4,480 acres) on a limestone resource in the vicinity of the Donlin Creek project (Table 4-3). The Donlin Creek LLC will need to demonstrate that the limestone is of high quality in order to secure the rights to these claims. These claims have not been legally surveyed.

Table 4-3: Claims located on Limestone Deposit

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	SEC.
641011	TUMS 1	T.A.	04/30/03	64.8	15N	45W	10
641012	TUMS 2	T.A.	04/30/03	64.8	15N	45W	10
641013	TUMS 3	T.A.	04/30/03	64.8	15N	45W	16
641014	TUMS 4	T.A.	04/30/03	64.8	15N	45W	16
641015	TUMS 5	T.A.	04/30/03	64.8	15N	45W	15
641016	TUMS 6	T.A.	04/30/03	64.8	15N	45W	16
641017	TUMS 7	T.A.	04/30/03	64.8	15N	45W	16
641018	TUMS 8	T.A.	04/30/03	64.8	15N	45W	15
641019	TUMS 9	T.A.	04/30/03	64.8	15N	45W	20
641020	TUMS 10	T.A.	04/30/03	64.8	15N	45W	21
641021	TUMS 11	T.A.	04/30/03	64.8	15N	45W	21
641022	TUMS 12	T.A.	04/30/03	64.8	15N	45W	20
641023	TUMS 13	T.A.	04/30/03	64.8	15N	45W	21
641024	TUMS 14	T.A.	04/30/03	64.8	15N	45W	21
641025	TUMS 15	T.A.	04/30/03	64.8	15N	44W	33
641026	TUMS 16	T.A.	04/30/03	64.8	15N	44W	33
641027	TUMS 17	T.A.	04/30/03	64.8	14N	44W	5
641028	TUMS 18	T.A.	04/30/03	64.8	14N	44W	4
641029	TUMS 19	T.A.	04/30/03	64.8	14N	44W	4
641030	TUMS 20	T.A.	04/30/03	64.8	14N	44W	5
641031	TUMS 21	T.A.	04/30/03	64.8	14N	44W	4
641032	TUMS 22	T.A.	04/30/03	64.8	14N	44W	4
641033	TUMS 23	T.A.	04/30/03	64.8	14N	44W	8
641034	TUMS 24	T.A.	04/30/03	64.8	14N	44W	9
641035	TUMS 25	T.A.	04/30/03	64.8	14N	44W	9
641036	TUMS 26	T.A.	04/30/03	64.8	14N	44W	8

Page 4-7

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	SEC.
641037	TUMS 27	T.A.	04/30/03	64.8	14N	44W	9
641038	TUMS 28	T.A.	04/30/03	64.8	14N	44W	9

Additionally, Donlin Creek LLC has three prospecting sites comprising 97 hectares (240 acres) (Table 4-4). These claims have not been legally surveyed.

Table 4-4: Prospecting Sites

			Located/	Size in
ADL #	Claim Name	Status	T.A.'ed	Hectares	Twp	Rng	Sec
582920	PDD 9	S.S.	10/06/96	32.4	23N	48W	2
582923	PDD 12	S.S.	10/06/96	32.4	23N	48W	2
582925	PDD 15	S.S.	10/06/96	32.4	23N	48W	11

Page 4-8

Figure 4-1: Location Map

Page 4-9

Figure 4-2: Lease Block Map

Page 4-10

Figure 4-3: Project Location Map

Page 4-11

4.3 Agreements and Permits

Placer Dome acquired a 20-year lease from Calista effective May 1, 1995. The lease agreement contains a provision that extends the lease period beyond 20 years as long as mining or processing operations continue in good faith or good faith efforts are being made to place a mine on the property into production (as discussed below). On November 13, 2002, NovaGold Resources Alaska, Inc., a wholly-owned subsidiary of NovaGold Resources Inc., earned a 70% interest in the project by expending US$10 million on exploration and development of the project. Once the financial commitment was fulfilled, Placer Dome had 90 days to decide on one of three options: a) to remain at 30% interest and participate as a minority partner; b) to convert to a 5% Net Profits Interest (NPI); or c) to exercise a back-in right to re-acquire a majority interest in the project (70%) by expending three times the amount expended by NovaGold at the time the back-in is exercised, completing a feasibility study, and making a decision to construct a mine at a production rate of not less than 600,000 ounces of gold per year within a five-year period from the exercise back-in. On February 11, 2003, Placer Dome exercised its back-in right and assumed management of the continued development of the Donlin Creek project. In January 2006, Barrick acquired Placer Dome and assumed Placer Dome’s joint venture responsibilities with regard to Donlin Creek.

On December 1, 2007, NovaGold entered into a limited liability company agreement with Barrick that provided for the conversion of the Donlin Creek project into a new limited liability company, the Donlin Creek LLC, which is jointly owned by the Company and Barrick on a 50/50 basis. As part of the Donlin Creek LLC, NovaGold has agreed to reimburse Barrick over time for approximately US$63.5 million, representing 50 percent of Barrick’s approximately US$127 million expenditures at the Donlin Creek project from April 1, 2006 to November 30, 2007. NovaGold’s reimbursement will be made following the effective date of the agreement, by the Company paying the next approximately US$12.7 million of Barrick’s share of project development costs, and the remaining approximately US$50.8 million will be paid out of future mine production cash flow. These amounts were agreed to subject to adjustment upon audit of the US$127 million expenditure. After the Company’s initial contribution, all funding will be shared by both parties on a 50/50 basis. Upon submission of a feasibility study, Calista retains a 90 day back-in right to participate in the project at a level of 5% to 15% by committing to contribute its share of capital. Their share would be divided pro rata from Barrick and NovaGold.

An advance minimum royalty (“AMR”) on the Donlin Creek property of US$200,000 is payable by the joint venture to Calista annually until a feasibility study is completed, after which the AMR will increase to US$500,000 per year. Upon commencement of production, a net smelter return royalty on production equal to the greater of 1.5% of

Page 4-12

	the revenues from valuable minerals production and US$500,000 is payable to Calista, until the earlier of the expiry of five years or the payback of all pre-production expenses incurred by Barrick and NovaGold. Thereafter, the annual net smelter return royalty on production will be increased to the greater of 4.5% of the revenues from valuable minerals production and US$500,000.
	Lyman Resources has existing placer mining leases covering approximately four square miles within the Donlin lease area. The Lyman family also has title to approximately 13 acres of surface estate within the Snow Gulch area. This lease area lies immediately to the north of the current open pit shell outline but should not result in any significant conflicts with the pit shell or envisioned infrastructure layout. The Calista Exploration and Lode Mining Lease grants priority to extraction of the lode resource in the event of a conflict of use between lode and placer mining operations, provided that a two-year notice period is provided to Lyman Resources. Negotiations regarding the future of the Lyman holdings are ongoing.
	Barrick has maintained all of the necessary permits for exploration and camp facilities. These permits are active at the Alaska Department of Natural Resources (hard rock exploration, temporary water use), the Corp of Engineers (individual 404 and nationwide 26), Alaska State Department of Conservation (wastewater, drinking water, food handling), the Alaska Department of Fish and Game (title 16 – fish), the Environmental Protection Agency (NPDES) and the Federal Aviation Administration (airport).
4.4	Environmental
	The environmental baseline study program was initiated in 1996 and has run continuously at varying levels of activity, except for a hiatus in 2001 and the first half of 2002. Initial work focused on studies needed to support the exploration program, largely independent of the ultimate project design. These studies included surface water quality in the general area, meteorology, aquatic studies in the main drainages, wetlands delineation in the vicinity of the known resource and some waste rock characterization.
	In 2003, as project concepts began to be defined, the baseline program was expanded to include studies such as ambient air monitoring, terrestrial wildlife and avian surveys, groundwater monitoring, detailed aquatic studies, cultural site surveys, detailed waste rock characterization and additional wetlands delineation in the areas of the facilities and supporting infrastructure. Evaluation of initial baseline studies and design concepts was then used as a basis for the expansion of initial studies, the initiation of new studies and the discontinuation of older studies. As the project continued to evolve, and based on feedback from regulatory and public consultation, additional

Page 4-13

studies were initiated to address issues of concern such as mercury and the impact of barge traffic on subsistence fishing and river erosion.

The three primary reasons for collecting baseline data are to inform the design process, to determine environmental controls to mitigate the impacts of exploration activities and future development on the area, and to characterize the project environment in anticipation of compliance with the National Environmental Policy Act (NEPA) and permitting. The environmental baseline data provide a reference point for environmental assessments and facilitate early detection of potential changes that may occur during mine development and operation.

Permits issued by federal agencies constitute “federal actions.” Any major federal action requires review under NEPA. All elements of a project and their cumulative effects are considered and evaluated in a NEPA review. In addition, alternatives to the proposed action are evaluated and potential mitigation measures are identified. For Donlin Creek, NEPA will require the preparation of an environmental impact statement (EIS). Typically, under NEPA the federal agency with the predominant permit is designated the lead agency. The lead agency for this project has not yet been selected.

Over the nearly 12 years since exploration and environmental baseline data collection began, considerable effort has been spent developing support for the project by fostering local relationships, developing a strong local workforce, educating stakeholders about the project and mining in general and providing stakeholders with regular project updates and site visits. This activity has enabled the Donlin Creek LLC to better understand and address the perspectives and concerns of the project stakeholders and has resulted in broad public support for the project, especially in the upriver region surrounding the immediate project area. This support has taken the form of resolutions from tribal councils and organizations, participation by individuals and tribal groups in various project-related forums and permissions granted to conduct environmental baseline studies on tribal lands.

Donlin Creek will require a considerable number of permits and authorizations from both federal and state agencies. Much of the groundwork to support a successful permitting effort is done prior to the submission of permit applications, so that issues can be identified and resolved, supporting baseline data can be acquired, and regulators and stakeholders can become familiar with the proposed project.

To support successful application for the more than 60 permits, this project will likely require extensive baseline environmental information, supporting scientific analysis, and detailed engineering design. Donlin Creek LLC and predecessors have invested significant money, resources, and time acquiring this information over the last 5 years, and in some cases the last 12 years. Designing in line with baseline data in advance

Page 4-14

of filing permit applications has resulted in a project that affords due consideration to all environmental concerns and is designed to mitigate potential impacts on the environment wherever practicable.

The comprehensive permitting process for Donlin Creek can be divided into three clear categories, all of which are important to the successful establishment of a future mining operation:

1.	Exploration stage permitting – required to obtain approval for exploration drilling, environmental baseline studies, and feasibility engineering studies.
2.	Pre-application phase – conducted in parallel with feasibility engineering studies. This stage includes the collection of environmental baseline data and interaction with stakeholders and regulators to facilitate the development of a project that can be successfully permitted.
3.	The NEPA process and formal permit applications – formal agency review and analysis of the project, resulting in the issuance or denial of permits.

As a result of comprehensive interaction with regulators and routine informal interaction with individual agencies during exploration permitting, the project is now well positioned to trigger the NEPA review and move forward with permit applications for construction, operations, and closure. Regulators who will be administering this review now have a solid understanding of the project and confidence in the manner in which the supporting baseline data have been collected and evaluated.

Permit review timelines are controlled by the requirements of the federal NEPA review and State requirements for meaningful public and agency participation to determine if the project is in the State’s best interest. Having engaged in comprehensive dialogue with stakeholders and regulatory agencies, and by moving forward with a well-defined project description, the Donlin Creek LLC has positioned itself well for an optimal permit review timeline.

Upon completion of the NEPA review, a positive Record of Decision (ROD), and final issuance of permits and authorizations, the Environmental Management System (EMS), consisting of a number of management and maintenance plans for the Donlin Creek gold project, will be fully implemented. The comprehensive permit review process will determine the precise number of management plans required to address all aspects of the project to ensure compliance with environmental design and permit criteria. Each plan will describe the appropriate environmental engineering standard (e.g., secondary containment for petroleum products, process solutions, and reagents) and the applicable operations requirements, maintenance protocols, and response actions.

Page 4-15

4.5	Permits and Process
	The Donlin Creek project will require multiple State and Federal permits, approvals, licenses, and authorization from Federal, State, Local and Tribal governments. Table 4-1 provides a brief summary of the permits required for the Donlin Creek project.

Page 4-16

Table 4-5: Federal Agency Permit and Authorizations

Agency	Authorization
Federal
Bureau of Land Management (BLM)	Surface Estate Lease (facilities managed lands)
	Land Use Permit (activities on BLM managed lands)
	Access Right-of-Way (BLM managed lands)
Environmental Protection Agency (EPA)	CWA Section 402 NPDES Permit (discharges to waters of the
	U.S.)
	Spill Prevention Containment and Contingency (SPCC) Plan
	Storm Water Pollution Prevention Plan – Construction and
	Operations
U.S. Army Corps of Engineers (USACE)	CWA Section 404 Permit (wetlands dredge and fill)
	River and Harbors Act (RHA) Section 10 (structures in
	navigable waters)
	RHA Section 9 (dams and dikes in navigable waters –
	interstate commerce)
U.S. Coast Guard	RHA Section 9 Construction Permit (bridge across navigable
	waters)
	Marine Protection, Research, and Sanctuaries Act compliance
	(ocean dumping requires a permit)
Bureau of Alcohol, Tobacco, and Firearms	License to Transport Explosives
	Permit and License for Use of Explosives
Federal Aviation Administration	Notice of Landing Area Proposal (existing airstrip)
	Notice of Controlled Firing Area for Blasting
U.S. Department of Transportation	Hazardous Materials Registration
National Marine Fisheries Service	Marine Mammal Protection Act authorization (IHA/LOA)
U.S. Fish and Wildlife Service	Section 7 of the Endangered Species Act, Consultation
	requiring a Biological Assessment or Biological Opinion
State
Office of Project Management and Permitting	Alaska Coastal Management Program Consistency Applicability
	Determination
Division of Mining, Land, and Water	Plan of Operations
	Reclamation Plan Approval
	Mining License
	Land Use Permits and Leases
	Right-of-Ways, Easements, Material Sales, etc.
	Certificate of Approval to Construct a Dam
	Certificate of Approval to Operate a Dam
	Temporary Water Use Permit
	Water Rights Permit/Certificate to Appropriate Water
	Tidelands Permit
Office of History and Archaeology/State Historic	Section 106 Historical and Cultural Resources Protection Act
Preservation Office	clearance
Office of Habitat Management and Permitting	Fish Habitat Permit
	Culvert/Bridge Installation Permit

Page 4-17

Agency	Authorization
Division of Water	Section 401 Water Quality Certification (CWA 404 permit)
	Section 401 Water Quality Certification (CWA 402 permit)
	Wastewater Disposal Permits
	Non-Domestic Wastewater Disposal Permit
	Storm Water Discharge Pollution Prevention Plan
	Domestic Wastewater Disposal Permit
	Approval to Construct and Operate a Public Water Supply
	System
Division of Environmental Health	Solid Waste Disposal Permits
	Food Sanitation Permit
Division of Air Quality	Air Quality Construction Permit (first 12 months)
	Air Quality PSD Title V Operating Permit (after 12 months)
	Air Quality permit to Open Burn

Each Federal and State permit will have compliance stipulations that require scrutiny and negotiation that can typically be resolved within 60 days of the ROD.

Project delays could occur as public opposition, inefficiencies in regulator review or project changes made by the owner.

Present environmental liabilities are believed to be limited to the exploration camp.

Page 4-18

5.0	ACCESSIBILITY, CLIMATE, LOCAL RESOURCES, INFRASTRUCTURE AND PHYSIOGRAPHY
	The Donlin Creek property is located in southwest Alaska, approximately 19 km (12 mi) north of the village of Crooked Creek on the Kuskokwim River (see Figure 4.1). The Kuskokwim River is a regional transportation route and is serviced by commercial barge lines. A 25 km (15 mi) long winter road, designated as an Alaska State Highway route and transportation corridor, accesses the property from the barge site at the village of Crooked Creek (Figure 4.2). The project has an all-season camp with facilities to house up to 160 people and an adjacent 1,500 m (5,000 ft) long airstrip capable of handling aircraft as large as C-130 Hercules (19,050 kg or 42,000 lb capacity), allowing efficient shipment of personnel, large equipment and supplies. The project is directly serviced by commercial air services out of both Anchorage, 450 km (280 mi) to the east, and Aniak, 80 km (50 mi) to the west.
	The area has a relatively dry interior continental climate with typically less than 50 mm (20") total annual precipitation. Summer temperatures are relatively warm and may reach nearly 30°C (83°F). Minimum temperatures may fall to well below -20°C (0°F) during the winter months.
	The project area is one of low topographic relief on the western flank of the Kuskokwim Mountains. Elevations range from 150 m to 640 m (500 ft to 2,100 ft). Ridges are well rounded and easily accessible by all-terrain vehicle. Hillsides are forested with black spruce, tamarack, alder, birch and larch. Soft muskeg and discontinuous permafrost are common in poorly drained areas at lower elevations.
	The Maximum Design Earthquake for the “High” hazard potential water and tailings dams is characterized by a peak horizontal ground acceleration of 0.4g from a Magnitude 7.8 event. For the design of plant site buildings and other structures, the seismic design provisions of the 2006 International Building Code result in a mapped short-period spectral response acceleration, Ss, of 0.5g, and a mapped 1s period spectral response acceleration, S1, of 0.14g. Site Class B conditions are anticipated for the plant site, pending future confirmation.
	The project is currently isolated from power and other public infrastructure. The exploration camp has a capacity of 160 persons. Power is provided by diesel generators. In regard to mining operations, sufficient space is available to site the various facilities, including personnel housing, stockpiles, tailing storage facility, waste rock storage facilities and processing plants. Ample water supply is available from surface and subsurface sources. Additional diesel generators or alternative power sources would need to be developed for mining operations.

Page 5-1

6.0	HISTORY
	Prior to 2003, operators have undertaken significant work on the property. Table 6-1 summarizes the work history at Donlin Creek.

Table 6-1: Work History Summary

Year	Company	Work Performed	Results
1909 to 1956	Various	Gold discovered on Donlin Creek in	Total placer gold production of
	prospectors and	1909. Placer mining by hand,	approximately 30,000 oz.
	placer miners	underground and hydraulic
		methods.
1970s to	Robert Lyman	Resumed sluice mining in Donlin	800 oz Au recovered in the first year of
1996	and heirs	area and placer mined Snow	operation in Snow Gulch.
		Gulch.
1974, 1975	Resource	Regional mineral potential	Soil, rock and vein samples return
	Associates of	evaluation for Calista Corporation.	anomalous gold values. Trench rock
	Alaska (RAA)	Soil grid and 3 bulldozer trenches	sample results range from 2 ppm Au to
		dug in Snow area.	20 ppm Au.
1984 to 1987	Calista	Minor work. Various mining
	Corporation	company geologists visit property.
1986	Lyman	Placer drilling finds abundant gray,	Initial discovery of Far Side (Carolyn)
	Resources	sulphide-rich clay near Quartz	prospect.
		Gulch.
1987	Calista	Rock sampling of ridge tops and	Anomalous gold values from auger
	Corporation	auger drill sampling of Far Side	holes; best result = 9.7 ppm Au.
		prospect.
1988, 1989	Western Gold	Airborne geophysics, geologic	Initial work identified eight prospects
	Exploration and	mapping and soil sampling over	with encouraging geology ± Au values
	Mining Co.	most of project area. Total of	(Snow, Dome, Quartz, Carolyn, Queen,
	(WestGold)	13,525 m of D-9 Cat trenching at all	Upper Lewis, Lower Lewis and
		prospects. Over 15,000 soil, rock	Rochelieu). Drilling at most of these
		chip and auger samples collected.	prospects led to identification of the
		947 m of AX core drilling, 404 m	Lewis areas as having the best bulk-
		(239 holes) of auger drilling and	minable potential. Calculated gold
		10,423 m of RC drilling (125 holes).	resource of 3 M tons at average grade of
		First metallurgical tests and	2.50 ppm (218,908 oz) at 1 ppm cut-off.
		petrographic work.	WestGold dissolved by early 1990.
1993	Teck Exploration	1,400 m of D-9 Cat trenching and	Identified new mineralized areas and
	Ltd.	two 500 m soil lines in Lewis area.	expanded property resource estimate to
		Petrographic, fluid inclusion and	3.9 M t at average grade of 3.15 g Au/t
		metallurgical work.	(393,000 oz Au).

Page 6-1

Year	Company	Work Performed	Results
1995 to 2000	Placer Dome	87,383 m of core, 11,909 m of RC	Discovery of American Creek Magnetic
		drilling and 8,493 m of trenching.	Anomaly (ACMA) when testing an
		Environmental work.	aeromag anomaly. Numerous mineral
			resource calculations.
2001, 2002	NovaGold	39,092 m of core, 11,589 m of RC	43-101 Preliminary Assessment
		drilling, 89.5 m of geotechnical	Measured and indicated: 117.4 million
		drilling and 268 m of water	tonnes at 2.91 g/t (1.5 g/t cut-off),
		monitoring holes. Updated	Inferred: 142.4 million tonnes at 3.1 g/t
		resource estimate.	(1.5 g/t cut-off)
2003	Placer Dome	25,448 m of core and 5,979 m of	Infill drilled throughout the resource
through		RC drilling	area. Discovered a calcium carbonate
2005			resource.
2006	Barrick		Infill drilled throughout the resource
through			area. Completion of DC7a resource
2007			estimation model.

6.1	1996 Activities
	Major activities included:
		building a 75-person Weatherhaven tent camp
		constructing a 1,500 m (5,000 ft) airstrip on American Ridge
		constructing more than 4 km (2.5 mi) of new road between camp and mineral prospects
		drilling a total of 34,995 m (144 holes, both core and reverse circulation)
		assaying more than 21,000 drill, rock and soil samples
		excavating more than 2,500 m of trenches for sampling and mapping purposes in southeast Lewis area.
	Most core drilling was on Lewis and Queen ridges, but eight core holes were drilled on the Dome, Far Side (formerly Carolyn) and Snow prospects. Seven RC drill holes were located at the southern end of an aeromagnetic anomaly southwest and west of Lewis Ridge. Four water wells were drilled for camp and drilling purposes.
	Metallurgical studies were conducted on both sedimentary- and igneous-hosted mill feed from the Lewis area.

Page 6-2

6.2	1997 Activities
	The goal of the 1997 exploration program was to develop a structural/geologic model of the Lewis/Queen area that would assist in determining mineralization controls. The following tasks were completed during 1997:
		8,129 m of reverse circulation (RC) drilling in 52 holes concentrated in wetlands and environmentally sensitive areas
		15,771 m of HQ core drilled in 67 holes across the property
		4,222 m of trenches excavated and a detailed geologic and mineralization map completed in the Lewis area
		Air photos taken of the Donlin Creek project area
		25 line km of max-min (electromagnetic) geophysical survey completed in the ACMA, 400 and southern Lewis areas
		1,800 line km of aeromagnetic survey completed at 50 m line spacing and 50 m elevation over the property
		more than 600 soil samples collected in the ACMA and 400 areas
		2,100 m of 1996 and 1997 trenches reclaimed in the Lewis area
		continuation of baseline environmental studies.
6.3	1998 Activities
	The main tasks completed in 1998 include:
		24,131 m of HQ core drilled in 96 holes, mainly in the Lewis, Queen and ACMA areas (ACMA discovered when testing a magnetic anomaly)
		1,904 m of trenching and mapping in the Lewis/Vortex areas and 150 m of trenching and mapping in the ACMA area (includes re-trenching and re-mapping of older trenches)
		Air photos taken of the Donlin Creek project area from the airstrip to Dome at 1:20,000 scale
		geological reconnaissance within the Donlin Creek property boundary
		ongoing reclamation of trenches throughout the property
		continuation of baseline environmental studies.
6.4	1999 Activities
	Two programs were completed during 1999: an exploration drilling program focused in the ACMA/400 area, and a property-wide exploration program to locate other higher- grade prospects. Results were:

Page 6-3

		9,189 total m of core drilled in 33 holes
		646 soil samples and 92 rock samples collected
		17.7 km of IP and resistivity lines completed
		2,237 m of trenching and mapping (Dome, Queen, Far Side and Vortex)
		property-wide 1:10,000 geological mapping
		ongoing reclamation of trenches throughout the property (900 m reclaimed in 1999)
		continued baseline environmental studies.
6.5	2000 Placer Dome
	Work during 2000 included an IP/resistivity survey and a drill program to test IP/resistivity anomalies coincident with soil geochemistry anomalies generated in the Dome-Quartz area. Results included:
		41.6 km of IP/resistivity lines
		1,403 m of core drilled in 7 holes from the Dome and Quartz areas
		completion of a supplemental resource economic study
		continued baseline environmental studies.
6.6	2001 NovaGold
	NovaGold began field work on the project in 2001 after finalization of a joint venture agreement with Placer Dome. Work in 2001 included the following:
		7,403 m of HQ core drilled in 42 holes from the ACMA area
		822 m of trenching in the Lewis area
6.7	2002 NovaGold
	NovaGold continued work on the property in 2002 focusing on expanding both the ACMA resource and defining mineralization and new resources in adjacent prospect areas (Aurora, 400, Akivik as well as Vortex). Work in 2002 included the following:
		39,092 m of HQ core in 194 holes from the ACMA, Aurora, 400, Akivik and Vortex areas
		89.5 m of HQ core in 2 geotechnical holes from Anaconda Creek
		11,589 m of exploration RC drilling and sampling in 147 holes from the ACMA, Akivik, Aurora and Nuno areas
		268 m of RC drilling in 5 water monitoring wells
		resource estimation and preliminary assessment (AMEC, 2002a and 2002b)

Page 6-4

	Measured and indicated: 104.1 million tonnes at 3.00 g/t (1.5 g/t cut-off)
	Inferred: 129.1 million tonnes at 3.11 g/t (1.5 g/t cut-off)
	contracted an updated economic study
	continued baseline environmental studies.
Stephen Juras, P.Geo., of MRDI Canada, a division of AMEC E&C Services Limited (MRDI), estimated the mineral resources for the project effective January 24, 2002, and issued a report in February 2002 entitled “Technical Report, Donlin Creek Project, Alaska”. The report is filed on SEDAR under NovaGold Resources Inc.
MRDI reports the estimates were made from 3D block models utilizing commercial mine planning software (MineSight®). Mineralized domains were interpolated based on mineralized geology and grade estimation based on ordinary kriging. Mineralized envelopes were defined through PACK on gold thresholds of 0.7 g/t Au and 0.5 g/t Au for the ACMA and Lewis zones, respectively. Gold values, in the 2 m composites, were capped at 30 g/t Au and 20 g/t Au for the ACMA and Lewis zones, respectively.
MRDI reported mineral resources at gold prices ranging from US$250 to US$350 per ounce of Au and corresponding cut-off grades of 3.5 to 1.5 g/t Au respectively. The capped mineral resource estimate at US$350 per ounce Au is summarized in Table 6- 2. MRDI reported that the mineral resource classification was consistent with CIM definitions referenced in NI43-101. The resource estimate is reliable but is irrelevant given the material amount of drilling completed since 2002.

Page 6-5

Table 6-2: Mineral Resources - MRDI - January 24, 2002

Classification	Tonnes		Grade	Cont. Au
	(t 000's	)	(g/t Au)	(oz 000's	)
Measured
ACMA	2,262		4.48	326
Lewis	4,331		2.76	384
Total Measured	6,593		3.35	710
Indicated
ACMA	32,327		3.54	3,678
Lewis	65,203		2.70	5,652
Total Indicated	97,530		2.98	9,330
Measured + Indicated
ACMA	34,589		3.60	4,004
Lewis	69,534		2.70	6,036
Total Measured +
Indicated	104,123		3.00	10,040
Inferred
ACMA	48,852		3.53	5,550
Lewis	80,291		2.86	7,371
Total Inferred	129,144		3.11	12,921

Notes: 1.	Mineral resources estimated at gold price of US$350/oz
2.	Mineral resources estimated at cut-off grade of 1.5 g/t Au
3.	Columns may not total exactly due to rounding

6.8	2003 Placer Dome
	Placer Dome elected to return as operator in 2003 as per the joint venture agreement. Work in 2003 included the following:
		updated the resource estimation based on NovaGold’s 2002 and previous drill programs (AMEC, 2003)
			Measured and indicated: 117.4 million tonnes at 2.9 g/t (1.5 g/t cutoff)
			Inferred: 142.4 million tonnes at 3.1 g/t (1.5 g/t cutoff)
		calcareous sandstone investigations
		economic studies

Page 6-6

6.9	2004 Placer Dome
	Placer Dome focussed on environmental and geotechnical studies in 2004. Work included the following:
		2,335 m of RC drilling and sampling in 17 condemnation holes
		852 m of HQ core in 3 geotechnical holes
		geologic mapping and sampling for carbonate-rich material
		continued environmental baseline studies
6.10	2005 Placer Dome
	Placer Dome focussed on resource conversion, geotechnical investigation and environmental studies in 2005. Work included the following:
		24,596 m of HQ core (resource infill, geotechnical, condemnation) in 90 holes from the ACMA, Akivik, 400 Vortex, Lewis and Far East areas
		3,644 m of RC drilling and sampling in 30 condemnation, water well and calcium carbonate exploration holes
		154 m in 28 auger holes for geotechnical purposes
		22 test pits for geotechnical purposes
		continued environmental baseline studies
		continued economic studies
6.11	2006 NovaGold
	NovaGold reviewed and validated a mineral resource model constructed by Placer Dome with an effective date of January 19, 2006. Stanton Dodd authored a report entitled “Donlin Creek Project, 43-101 Technical Report”, dated January 20, 2006. The report is available on SEDAR under NovaGold Resources Inc.
	The resource estimates were made from 3D block models using proprietary Placer Dome mine modelling software (OP). Table 6-3 summarizes the resource estimate. The resource estimate is reliable but is irrelevant given the material amount of drilling completed since January 2006.

Page 6-7

Table 6-3: Mineral Resources - NovaGold - January 19, 2006

Classification		Tonnes		Grade	Cont. Au
		(t 000's	)	(g/t Au)	(oz 000's	)
Measured		16,100		2.84	1,469
Indicated		151,100		2.75	13,360
Measured + Indicated		167,200		2.76	14,829
Inferred		156,000		2.72	13,643
Notes: 1.	Mineral resources estimated at gold price of US$450/oz
2.	Mineral resources estimated at cut-off grade of 1.2 g/t Au
3.	Columns may not total exactly due to rounding

A September 2006 preliminary economic assessment (“PEA”) prepared for NovaGold by SRK Consulting (US), Inc. confirmed the economics of a conventional open-pit mining operation at a production rate of 60,000 t/d with the potential to produce on average 1.4 million ounces of gold per year over the estimated 22-year life of the project in the report titled “Preliminary Assessment Donlin Creek Gold Project Alaska, USA” dated September 20, 2006. The report is available on SEDAR under NovaGold Resources Inc.

Costs, appropriate with this level of the study, were estimated and formed the foundation of the economic analysis of the project on a 100% basis. The study was prepared based on a technical and economic review by a team of consultants who are specialists in the fields of mineral exploration, mineral resource estimation and classification, open-pit mining, mineral processing and mineral economics. The study was completed under the direction of Gordon Doerksen, P.E., an independent Qualified Person as defined by NI 43-101. The resource estimate with an effective date of January 19, 2006 was the basis of the PEA.

SRK also completed a sensitivity analysis to determine the economic effects of changes to the capital and operating costs and the gold price, to determine the economic potential of the Donlin Creek project. In the first 7 years, the study projects average annual production of approximately 1.885 million ounces of gold at an average cash cost of US$223/oz of gold. The project would generate an average annual pre-tax cash flow of approximately US$482 million for the first 7 years using a long-term gold price of US$500/oz, resulting in rapid payback of all mine capital in less than 5 years.

SRK’s analysis indicated that using a gold price of US$500/oz, Donlin Creek could generate a pre-tax rate of return of 12.1% and a net present value (NPV) at a 5%

Page 6-8

	discount rate (“NPV5%”) of US$1,001 million, resulting in a capital cost payback period of less than 5 years. A sensitivity analysis on the project shows that the NPV is most sensitive to changes in the gold price, followed by changes to operating costs and capital costs. For example, a gold price of US$550/oz increases the NPV5% to US$1,453 million, and a gold price of US$450/oz decreases the NPV5% to US$554 million.
	This financial analysis includes capital costs to construct a powerline connecting the Donlin Creek project site to the existing Anchorage/Fairbanks power grid. The study was preliminary in nature, and included inferred mineral resources that are considered too speculative geologically to have the economic considerations applied that would allow them to be categorized as mineral reserves, and there is no certainty that the conclusions reached in this PEA will be realized. The PEA is reliable but is not current given the material amount of drilling and technical studies completed since January 2006 and possible capital cost price escalation.
6.12	2006–2007 Barrick
	Barrick focussed on resource conversion, geotechnical investigation, metallurgical and environmental studies from 2006 through 2007. Work included the following:
		135,362 m (444,101 ft) of primarily HQ core (resource infill, geotechnical, metallurgical, condemnation) in 359 holes in Lewis and ACMA.
		continued environmental baseline studies
		water geochemistry
		peat exploration
		wind power generation studies
		metallurgical studies

Page 6-9

7.0	GEOLOGICAL SETTING
7.1	Regional Geology
	The Kuskokwim region of southwestern Alaska is predominately underlain by rocks of the Upper Cretaceous Kuskokwim group (Figure 7-1). These include coarse- through fine-grained clastic rocks that reach an estimated thickness of 12 km (7.5 miles). Minor basin margin andesitic tuff and flows are also present near the top of the sequence and may represent an initiation of volcanism that later culminated in widespread Late Cretaceous and early Tertiary igneous activity. These basin margin volcanic rocks also suggest that deep penetrating structures controlled basin subsidence.

Figure 7-1: Regional Geology of Donlin Creek Area

Kuskokwim Group sediments filled a northeast-trending strike-slip basin that subsided between a series of amalgamated terranes including Mesozoic marine volcanic rocks, Paleozoic clastic and carbonate rocks, and Proterozoic metamorphic rocks. Kuskokwim Group rocks generally do not display penetrative metamorphic fabric, but they are locally folded.

Page 7-1

	Igneous activity was coeval with Late Cretaceous sedimentation in the Kuskokwim basin and continued into the early Tertiary. Intermediate composition volcano-plutonic complexes intrude and overlie Kuskokwim Group rocks throughout the region. The igneous rocks are predominantly tuffs, flows, and composite comagmatic monzonite to granodiorite plutons. Volcanic and plutonic rocks range in age from 76 to 63 Ma and 71 to 66 Ma, respectively. Kuskokwim sedimentary rocks are often extensively hornfelsed near plutons. Volumetrically minor Late Cretaceous intermediate to mafic intrusive bodies are also common and often associated with mercury and antimony occurrences. Felsic to intermediate hypabyssal granite to granodiorite porphyry dikes, sills, and plugs are also widely distributed and often associated with placer and lode gold occurrences (e.g., Donlin Creek). Many dikes were emplaced within or near northeast-trending extensional zones. Contacts between porphyry igneous rocks and Kuskokwim sedimentary rocks are generally sharp and do not display hornfelsed margins. Age dates range from 70 to 65 Ma, but a genetic association with the volcano-plutonic complexes is uncertain.
	The Donlin Creek area lies between two regional, northeast-trending, right lateral faults: the Denali-Farewell fault system to the south, and the Iditarod-Nixon Fork fault system to the north. The region contains numerous northeast to east-northeast- and northwest to west-northwest-trending lineaments that probably represent steeply dipping strike-slip faults. Fault movement in the Donlin Creek region appears to be right lateral on northeast structures and left lateral on northwest structures. Folding in the region probably occurred soon after sedimentation, since folds are truncated by the volcano-plutonic complexes. East-trending open folds are prominent east of the Donlin Creek area, but appear truncated to the west by the Donlin Creek fault, a splay of the Iditarod-Nixon Fork fault.
7.2	Property Geology
	Simplified property scale igneous geology is shown in Figure 7-2. Undivided Kuskokwim Group sedimentary rocks (uncoloured) and felsic intrusive rocks associated with the 70 to 65 Ma igneous event are the main rock units. Greywacke is dominant in the northern part of the resource area (Lewis, Queen, Rochelieu, Akivik), while shale-rich units are common in the southern part of the resource area (South Lewis, ACMA). Sedimentary bedding strikes northwest and dips moderately to the southwest. Overall, sedimentary structure in the northern resource area is monoclinal, while sedimentary rocks in the southern resource display open easterly-trending folds.
	The earliest intrusive rocks at Donlin Creek are 74 to 72 Ma intermediate to mafic dykes and sills. They are not abundant, but occur widely throughout the property as generally thin and discontinuous bodies. The later and much more voluminous 70 to 65 Ma felsic dykes and sills vary from a few metres to 60 m (200 ft) wide and intrude

Page 7-2

the sedimentary rocks along a 8 km long by 3 km wide (5 mile x 2 mile), northeast-trending corridor. Sills are common in the southern resource area (shale-dominant), while dikes dominate in the north (greywacke-dominant). The felsic dykes and sills have similar mineralogy and generally display a porphyry texture indicative of relatively shallow emplacement. Although these rocks belong to the regionally important granite porphyry igneous event, geologists working on the property classify them into five textural varieties of rhyodacite. Rhyodacite is a term normally reserved for volcanic to sub-volcanic rock types, but it is also used informally for igneous rocks emplaced at a shallow depth. These units are chemically similar, temporally and spatially related, and probably reflect textural variations of related intrusive events. Differences include phenocryst size and abundance, groundmass texture, and overall colour.

In chronological order, the oldest documented structures are easterly-trending open folds and southerly-directed thrusts or low to moderate north-dipping reverse faults. The ca. 70 Ma north-northeast-striking rhyodacite porphyry dikes and west-northwest-striking sills were emplaced in multiple pulses and post-date the minor mafic dyke and sill event. Cross-cutting high- and low-angle northeast- and northwest-striking faults developed after the thrusts, and may have been active during, as well as after, igneous and hydrothermal activity. These structural trends are clearly evident in the surface geology and as topographical and aeromagnetic linears. Finally, gold-bearing north-northeast-striking extensional fractures formed and cut igneous rocks and faults. The fractures are best developed in the relatively competent igneous rocks and coarser greywacke-dominant sedimentary sequences.

Page 7-3

Figure 7-2: Main Trend Geology (Piekenbrock and Petsel 2003)

Page 7-4

As shown in Figure 7-3, the geophysical (magnetic) expression of the hydrothermal system is a pronounced northeast-trending aeromagnetic low (blue shades) related to the low magnetic susceptibility of the intrusive rocks, fracture-controlled, magnetite-destructive hydrothermal alteration, and possible weak thermal metamorphism of the enclosing sedimentary rocks.

Figure 7-3:	Aeromagnetic Image showing Interpreted Faults and Intrusive Rocks (source
	NovaGold)

7.3	Deposit Geology
	General geology of the resource area is shown in Figure 7-4. The 100 m (328 ft) level of the geological model is projected to the topographic surface.

Page 7-5

Figure 7-4:	General Geology of the Resource Area Showing Intrusive Rock Units and Faults (100
	m level projected to surface) (source Barrick)

7.3.1	Sedimentary Stratigraphy
	Preliminary stratigraphy for sedimentary rocks in the immediate deposit area is shown in Table 7-1.
	The stratigraphy in the deposit area consists of multiple turbidite sequences and is therefore very complex. Transition zones of rhythmically interbedded shales and lithic sandstones are common. Marker beds are not yet recognized, so absolute stratigraphic breaks are difficult to identify.

Page 7-6

Table 7-1: Donlin Creek Stratigraphy

Assigned		Apparent Thickness	Apparent Thickness
Nomenclature	Principal Rock Type	(ft)	(m)
Upper Greywacke	greywacke	328+	100+
Upper Siltstone	siltstone/shale	164	50
Main Greywacke	greywacke	262	80
Main Shale	shale/argillite	up to 459 (with sills)	up to 140 (with sills)
Basal Greywacke	greywacke	656	200+

Thicknesses presented for each unit in Table 7-1 are from the southern resource area also known as the American Creek magnetic anomaly, or ACMA area. In general, the Main Shale appears to thin to the west, whereas the Upper Siltstone appears to thicken in the same direction. The northern part of the resource area is mostly greywacke, while the southern area is shale rich. The coarse-grained greywacke contains abundant metamorphic lithic fragments and locally abundant igneous and sedimentary clasts. Shale-rich sedimentary rocks contain minor syngenetic pyrite, minor coaly plant debris to thin coal seams, and rare thin 10 cm (0 to 4") volcanic ashfall beds. The ash beds are restricted to low-energy shale and argillite intervals.

7.3.2 Igneous Rocks

The mafic dykes and sills and the five varieties of rhyodacite recognized in the Donlin Creek deposit are listed from oldest to youngest in Table 7-2, and are described below.

Table 7-2: Donlin Creek Intrusive Rocks

Name	Code	Age
Mafic Dykes/Sills	MD	oldest
Fine-Grained Porphyry	RDF	-
Crowded Porphyry	RDX	-
Lath-Rich Porphyry	RDXL	-
Aphanitic Porphyry	RDA	-
Blue Porphyry	RDXB	youngest

MD – Mafic Dykes

The earliest intrusive rocks at Donlin Creek are a series of intermediate to mafic dykes and sills (MD). These dykes and sills are thin 1 to 3 m (3 to 10 ft) and are normally characterized by intense pervasive carbonate and bright green clay + fuchsite (?) alteration. The mafic rocks are compositionally variable, typically porphyritic, and have been compared to lamprophyres. In the transition area between Akivik and ACMA, an

Page 7-7

area of extremely abundant mafic sills occurs within the Lower Greywacke immediately below the Main Shale. The mafic sills locally host high-grade gold.

RDF – Fine-Grained Porphyry

The RDF dykes are the earliest rhyodacite intrusions recognized at Donlin. They are typically fine-grained, felsic porphyries with distinctive small feldspar phenocrysts set in a grey fine-grained matrix. RDF intrusives occur as dykes 5 to 10 m (16.5 to 32.8 ft) wide and appear to fill the north-northeast extension fracture zones and the east-striking compressional (e.g., Lo fault) faults.

RDX – Crowded Porphyry

The RDX rocks are volumetrically the most significant intrusive phase on the property. The unit is characterized by a homogenous crowded porphyry texture and sharp intrusive contacts with little (<2" or <5 cm) to no chill margins. The unit occurs as two 50 to 100 m (164 to 328 ft) wide dike zones in the eastern edge of the north to north-northeast Lewis/South Lewis mineralized trend. RDX also occurs as sills throughout the southern portion of the property as the lowest part of the stratigraphy. The sills begin as sub-horizontal units in the South Lewis area and follow the syncline/anticline structure as they dip from sub-horizontal to near-overturned at depth in the ACMA area.

RDXL – Lath-Rich Porphyry

The RDXL unit is a rhyodacite phase characterized by large elongate plagioclase laths in a population of smaller K-spar phenocrysts. It has significant coarser grained biotite and seems to be more texturally enhanced by alteration than the other units. RDXL occurs as two important dykes in the Akivik area that strike south into the centre of the ACMA deposit. In the Akivik and ACMA areas, RDXL occurs as a significant sill immediately below the RDX sill package. The RDXL sill continues to the west, but pinches out to the east. Limited RDXL dykes can be seen within the main RDX dyke packages in the main Lewis area, but these dykes are volumetrically insignificant.

RDA – Aphanitic Porphyry

The RDA unit is a rhyodacite rock with a salt-and-pepper texture of fine biotite phenocrysts and variable quartz and K-spar phenocrysts set in an aphanitic matrix. It has distinctive flow-banded margins. Numerous (up to eight) RDA dikes strike south from the Vortex/Rochelieu area into the East ACMA/ACMA area. The dykes are typically found west of the Vortex fault, but can be found between the Lo and Vortex and below the Lo fault. An extensive sill package of RDA is located immediately above the RDX sills in the ACMA area. In west ACMA, the RDA sills are buttressed

Page 7-8

	against, and locally cross-cut, RDX sills. Another package of RDA sills is found south of the AC fault, in the Aurora domain.
	RDXB – Blue Porphyry
	The final intrusive event at Donlin Creek is represented by unit RDXB, or Blue Porphyry. This unit is coarsely porphyritic with large blocky feldspars set in a graphite and sulphide-rich matrix that gives the unit a distinctively dark appearance. This dark colour occasionally looks like an alteration product. It is often more intense near faults or appears to have pooled behind fluid barriers such as fractures and veinlets. Alteration “fronts” occasionally cut across feldspar phenocrysts. Darker colour near contacts with carbonaceous rocks suggests remobilization of carbon. The RDXB unit locally hosts important high-grade disseminated gold in addition to the more typical sheeted veins. RDXB occurs as two major dykes, the Lewis Blue Porphyry dyke and the Vortex Blue Porphyry dyke. Extensive, though thin, RDXB sills occur in the highest part of the section in South Lewis and ACMA areas. The RDXB sills occur as both distinct sills and co-mingled with RDA in the core of ACMA and in the Aurora domain.
7.3.3	Structural Geology
	The sedimentary rocks in the northern part of the resource area are monoclinal with average dips to the southwest of about 10° to 50°. Southeasterly-plunging open folds are evident in the southern resource area near the Lewis–ACMA transition (note intrusive rock map, Figure 7-4). Recent drilling shows that the south limb of the prominent syncline is vertical to overturned toward American Creek. Structural data from deep oriented core holes show an abrupt change from vertical to nearly horizontal bedding attitudes at depth. This vertical fold limb and the deep nearly flat bedding may suggest drag on a major low-angle fault at depth beneath ACMA.
	North-dipping low-angle reverse or thrust faults (?), including the Lo Fault and similar structures (Figure 7-4), are believed to have formed along stratigraphic competency breaks during north-northeast directed compression. There is some suggestion that the Lo Fault and the related upward transition from more massive greywacke to shale may have influenced the change from dykes in the Lewis area (Figure 7-5) to sills in the ACMA area (Figure 7-6). Relative displacements and an oblique sense of movement on later northwest and northeast high-angle faults are portrayed in Figure 7-4. Rob’s Fault, southeast of the known resource area, is very poorly understood, except where intersected in a few widely spaced shallow drill holes. None of these faults hosts significant mineralized material, but they may have locally enhanced ground preparation and influenced the circulation of gold-bearing hydrothermal fluids (fluid barriers). However, drilling in 2006 intersected gold-bearing, sulphide-rich tectonic and possible hydrothermal breccias in structures subparallel and near the Vortex fault. Similar mineralized material was also found in the Rochelieu deposit

Page 7-9

area, but the structural control remains uncertain pending additional drilling. These sulphide bodies indicate that both high- and low-angle structures may have been active during the gold mineralization. Finally, gold-bearing, north-northeast-striking, steeply southeast-dipping fracture zones cut across all faults and rock types in the deposit.

Figure 7-5:	Lewis Area Section, Looking Northeasterly Showing Intrusive Rocks, Drill Holes
	(source Barrick)

Page 7-10

Figure 7-6:	ACMA Area Section, Looking Southeasterly Showing Intrusive Rocks, Drill Holes
	(source Barrick)

Page 7-11

8.0	DEPOSIT TYPES
	Two distinct styles of gold-rich mineralization (Dome-Duqum style and ACMA-Lewis style) occur within the Donlin Creek trend. The Dome-Duqum mineralization, an early, high-temperature porphyry style, is characterized by fracture-controlled stockwork, laminated quartz-only veins containing varying proportions of copper, zinc, bismuth, silver, tellurium, selenium and local native gold mineralization. Silicification is locally associated with the veins. This style of mineralization occurs in the northern part of the property (Dome and Duqum prospects) (Figure 7.2). Contact metamorphism (hornfelsing) of the sedimentary rocks adjacent to host intrusives is common in areas containing this style of mineralization. Cross-cutting relationships were established in trench mapping during the 1999 field season that indicate the relative older age of the Dome mineralization.
	The ACMA-Lewis style of mineralization, a later low-temperature, low-sulfidation epithermal system, constitutes the main mineralizing system within the Donlin Creek property. This is the sole style of mineralization within the current resource area. The ACMA-Lewis style consists of sheeted quartz, quartz-carbonate and sulphide only veins characterized by a gold-arsenic-antimony-mercury geochemical signature. The bulk of the gold occurs in the lattice structure of arsenopyrite. Stibnite, realgar and native arsenic are commonly observed associated with zones of higher-grade gold mineralization but do not appear to host any significant gold mineralization compared to arsenopyrite. Disseminated gold-bearing arsenopyrite can also be found typically adjacent to veins and vein zones. Mineralization is best developed in all intrusive rocks and, to a much lesser extent, sediments (mainly greywacke). Sedimentary units in areas of ACMA-Lewis mineralization typically show no contact metasomatic effects.

Page 8-1

9.0	MINERALIZATION
9.1.1	Mineralization and Alteration
	Disseminated gold-bearing sulphides occur in the rhyodacite dikes and, to a lesser extent, in adjacent sediments. Structurally controlled mineralized veins formed within north-northeast-trending extensional fracture zones. These mineralized fracture zones are strongly developed where they intersect competent rock types such as felsic dykes and sills or massive greywacke. Quartz-carbonate-sulphide (pyrite, stibnite, and arsenopyrite) veins are the primary mineralized features, but gold also occurs in thin, discontinuous vein and fracture fillings. Veins seldom exceed 1 cm (0.4") wide; vein density can range up to 5 to 10 per metre; and vein zones vary from 2 to 35 m (6.5 to 115 ft) wide. Individual vein zones generally display limited lateral and vertical continuity. However, swarms of many anastomosing vein zones form larger mineralized corridors that display extensive lateral and depth continuity.
	The ACMA–Lewis style of mineralization is consistent with a low-temperature, low- sulphidation, epithermal gold model involving a strongly reduced, CO2-rich, weakly acidic, bisulphide-complexed, gold-bearing fluid. The deposit(s) is characterized by a gold-arsenic-antimony-mercury geochemical signature, sheeted quartz ± carbonate and sulphide veins, and disseminated sulphides. Common minerals observed in mineralized zones include pyrite, marcasite, arsenopyrite, stibnite, realgar, and native arsenic. Pyrite is the most common mineral and appears to be the earliest sulphide phase. It is ubiquitous in the rhyodacite and occurs as disseminated grains and micro- fracture fillings.
	Disseminated pyrite in the sedimentary rocks occurs as fine to coarse grains (up to 5 mm across) preferentially concentrated near dyke/sill contacts or as syngenetic pyrite. Relative abundance of pyrite is not an indicator of gold grade.
	Broad selvages of disseminated gold-bearing arsenopyrite and pyrite are found adjacent to veins and vein zones. Arsenopyrite commonly replaces pyrite and typically occurs as fine to very fine grains disseminated in intrusive rocks and as coarser aggregates in fractures and quartz-carbonate veins. In practice, fine-grained arsenopyrite can be difficult to distinguish from ubiquitous disseminated graphite.

Page 9-1

Figure 9-1:	ACMA Gold Distribution, Looking Southeasterly Showing Intrusive Rocks (colours)
	and Gold Grade Shells (>1 g/t stipple, >3 g/t pattern) (source Barrick)

Figure 9-2:	Lewis Gold Distribution Looking Northeasterly Showing Intrusive Rocks (colours) and
	Gold Grade Shells (>1 g/t stipple, >3 g/t pattern) (source Barrick)

Page 9-2

	Native arsenic occurs as dark grey, granular massive to botryoidal grains that often fill vugs in quartz-carbonate ± sulphide veins and other open spaces in breccias or fractures. Realgar and orpiment occur in late, quartz-sulphide veins. Stibnite commonly occurs as disseminated grains and masses within carbonate veins and occasionally as interlocking needles in open spaces within quartz-carbonate veins and on fracture surfaces. Other accessory sulphides and sulfosalts observed in the deposit include marcasite, pyrrhotite, chalcopyrite, chalcocite, covellite, tennantite, tetrahedrite, galena, sphalerite, and boulangerite. Pyrrhotite, stibnite, and boulangerite are paragentically late and appear to postdate most deformation while chalcopyrite, tennantite-tetrahedrite, pyrite, and arsenopyrite are both pre- and post-deformation.
	Very rare native gold particles (1 to 20 µm) have been observed in process mineralogy studies of ACMA–Lewis area material, but most of the gold occurs in the crystal structure of arsenopyrite and, to a lesser extent, in pyrite. Fine-grained arsenopyrite (<20 µm diameter) contains 5 to 10 times more gold than coarse-grained arsenopyrite. Gold seen in polished sections occurs as 1 to 3 µm blebs with no clear paragenetic relationship to other minerals. Stibnite, realgar, and native arsenic are often associated with higher gold grades but contain very minor gold compared to arsenopyrite.
9.1.2	Vein Types
	Multiple vein types apparently formed from a single hydrothermal fluid. Therefore, vein mineral assemblages show a continuum from pyrite through arsenopyrite, native arsenic, realgar, and orpiment, rather than discreet paragenetic stages. Stibnite is ubiquitous in all vein types but seems to increase in later vein stages. Gold grade and vein quartz generally increase from vein types V1 through V3 and then markedly decrease in V4, a carbonate-dominant vein type. Observed relationships in the vein assemblages are consistent with decreasing temperature, decreasing pH, and increasing fS2 or fO2 as a function of boiling or simple oxidation (fluid mixing).
	The vein assemblages from earliest to latest are described below.
	V1: The earliest veins are thin, discontinuous sulphide (>50%) veins with pyrite and trace arsenopyrite, little or no quartz (<30%) or carbonate (<50%), and a broad disseminated selvage of pyrite. They commonly contain minor amounts of ankerite in the veins, as well as ankerite with adjacent disseminated pyrite. The disseminated ankerite is pervasive in the rock matrix and also occurs as distinct overgrowths or rims around plagioclase phenocrysts. Such veins are typically low grade and show broad pervasive selvages of poorly crystalline illite alteration.
	V2: Thin, discontinuous quartz (>30%) sulphide veins contain variable pyrite and arsenopyrite and may have broad, often pervasive selvages of fine-grained needle-like

Page 9-3

	arsenopyrite. In many instances, a broad pyrite aureole surrounds the arsenopyrite selvage. Open-space vuggy textures are common, and trace amounts of stibnite are found in some veins. These veins show moderate grade, strong illite alteration, and variable but overall decreased ankerite content.
	V3a: The highest grade veins are thicker, more continuous, open-space quartz veins with pyrite, arsenopyrite, native arsenic, and variable amounts of stibnite. These veins commonly show broad arsenopyrite-rich selvages with little to no ankerite. In some instances, minor amounts of calcite are present in the veins.
	V3b: These are thick, continuous quartz veins with open-space textures and complex mineralogy, including pyrite, arsenopyrite, stibnite, native arsenic, realgar, orpiment, and trace sphalerite in intensely illite altered material. Realgar sometimes fills adjacent feldspar sites, creating a “pumpkin patch” texture in the wall rock. Gold grades are commonly very high.
	V4: The latest vein phase consists of broadly oriented carbonate-quartz (>50% and <50%, respectively) vein sets. This set has the lowest gold grade, is common in the sedimentary rocks, and seems to form a halo around mineralized zones. The carbonate halo is zoned outward from ankerite to ferroan dolomite to calcite.
9.1.3	Alteration
	Alteration mineral relationships appear to record decreasing pH due to boiling, falling temperature, liberated SiO2, and the effective removal of Fe, Ca, and Na from the system. The following proximal to distal silicate alteration zones and carbonate and graphite alteration products are associated with the ACMA-Lewis hydrothermal system. Silica is largely restricted to veins and is not an important wall rock alteration product.
	Illite Zone
	Intrusive rocks within the mineralized corridors typically display pervasive ammonia- illite alteration of the feldspars. Intense and more crystalline illite often occurs with higher-grade material. Minor amounts of higher temperature dickite are found in some of the strong illite zones. Minor smectite with locally very high grade gold is also occasionally found in strong illite zones.
	Illite–Kaolinite Zone
	Broad haloes of admixed illite and kaolinite surround the illite-rich mineralized corridors and are generally very low-grade or barren.

Page 9-4

	Smectite Zone
	Minor smectite as beidellite with minor ankerite and zeoloite in texturally destructive zones appears to occur distally within the system and probably records retrograde alteration during cooling and fluid mixing.
	Carbonate
	Carbonate veins appear to form an ankerite halo, often in the sedimentary package surrounding the mineralized intrusive phases. Ankerite is consistent with a reduced CO2-rich vapour phase evolved during boiling and forming a distal oxidized ankerite/kaolinite halo. The ankerite would serve as a trap for calcium liberated from illite alteration of plagioclase and for iron liberated from illite alteration of biotite and the pyrite-to-arsenopyrite phase transition.
	Graphite
	Very fine-grained graphite is frequently found in igneous units in open spaces or high- porosity areas, often with coarse illite or as isolated, shred-like fragments and grains. Graphite is consistent with an early proximal alteration product prior to the onset of boiling.
	Silica
	Significant pervasive wall rock silicification does not exist in the ACMA–Lewis gold deposits. However, vein relationships show an increase in quartz content from early sulphide-dominant veins to late silica-dominant veins. This probably reflects a simple thermal gradient culminating in the distal, massive quartz/stibnite veins seen around the margins of the district and shown to have cooler fluid inclusion formation temperatures. Silica is made available through the majority of hydration reactions involved in alteration of the original potassium feldspar, plagioclase, and biotite. Kaolinite and smectite reactions consume available silica; therefore, no significant silica veins are present with these assemblages.
9.1.4	Minor Elements and Deleterious Materials
	The most abundant minor elements associated with gold-bearing material are iron (Fe), arsenic (As), antimony (Sb), and sulphur (S). They are contained primarily in the mineral suite associated paragenetically or spatially with hydrothermal deposition of gold, including pyrite (FeS2), arsenopyrite (FeAsS), orpiment (As2S3), realgar (AsS), and native arsenic (As), and stibnite (Sb2S3). Minor hydrothermal pyrrhotite (Fe1-xS) and marcasite (FeS2), and syngenetic or sedimentary pyrite also account for some of the Fe and S.

Page 9-5

	Much less abundant elements such as copper (Cu), lead (Pb), and zinc (Zn) are contained in relatively rare or accessory hydrothermal mineral species observed in the deposit, including chalcopyrite (CuFeS2), chalcocite (Cu2S), covellite (CuS), tennantite (Cu12As4S13), tetrahedrite (Cu12Sb4S13), galena (PbS), sphalerite (ZnS), and boulangerite (Pb5Sb4S11). Small amounts of silver (Ag) in the deposit are most likely accommodated within the crystal structures of tetrahedrite and galena, and to a lesser extent in some of the other sulphides. Very minor Ni in the secondary sulphide mineral millerite (NiS), and minor Co in various secondary minerals have been observed in sedimentary rocks. The Ni and Co probably have a sedimentary origin.
	Three other elements that that have particular processing significance are mercury (Hg), chlorine (Cl), and fluorine (F). Graphitic carbon and carbonate minerals also negatively affect the metallurgical process.
	Most of the Hg is contained in pyrite followed by marcasite and stibnite. Very low level Hg was also detected in arsenopyrite and associated with realgar. Primary Hg minerals such as cinnabar (HgS) are absent or exceedingly rare. Mercury can also be accommodated in the crystal structures of other minerals such as tetrahedrite and sphalerite, but it is not known whether these accessory minerals contain Hg at Donlin. Native gold in the general Donlin Creek area is also known to have relatively high concentrations of Hg compared to other gold occurrences in the region.
	Process mineralogy studies show that muscovite (KAl2(AlSi3O10)(OH)2) and apatite (Ca5(PO4)3(F,Cl,OH)) are the principle sources of Cl and F and that the relatively more abundant muscovite accounts for most of the Cl and F. Muscovite is normally a rock forming mineral but it can also form during hydrothermal alteration along with structurally similar alteration products (illite) known to be associated with gold-bearing rocks in the deposit. Apatite is commonly found as an accessory mineral in intrusive and sedimentary rocks and as a hydrothermal alteration or vein mineral.
	Graphitic carbon (C) is relatively abundant in the sediments and variably disseminated in the intrusive rocks as a possible alteration product. Carbonate minerals occur as both pervasive, fine-grained hydrothermal alteration products, often intergrown with fine disseminated sulphide, and also in carbonate and quartz-carbonate ± sulphide veins. They include ankerite (CaFe(CO3)2), dolomite (CaMg(CO3)2), calcite (CaCO3), and very minor siderite (Fe(CO3)).
9.1.5	Structural Controls on Mineralization
	A detailed report on the geology and interpretation of the Donlin Creek gold deposit was completed by Piekenbrock and Petsel (2003) and forms the basis of the following discussion.

Page 9-6

Mineralization was structurally controlled along north-northeast-trending fault/fracture zones and mineralized zones are best developed where they intersect favourable rock types such as felsic intrusive dykes and sills and greywacke.

The orientation of the mineralized zones is consistently sub-parallel to main δ1 axis (N24E) of the D2 compressive structural regime. Veins in the ACMA-Lewis resource transition through a continuum of changing mineralogy and increasing grade while maintaining a generally consistent north-northeast-strike and southeast-dip. The last phase of veining consists of a more dispersedly oriented carbonate-quartz vein set (V4) and has the lowest gold grade of all the vein types.

Structural orientations were collected from drill core using the clay impression method and recorded in right-hand rule conventions. At the time of the 2003 study, a total of 12,799 oriented data points were available for various structures in both trenches and drill holes. Table 9-1 summarizes the results of stereonet and scatter plots of veins by resource domain. The data were limited to drill hole data within the resource area.

Table 9-1: Results of Stereonet and Scatter Plots of Veins by Resource Domain

Domain	Number of Data Points	Bulls Eye on Stereonet
All Veins	4,496	N 024/68
ACMA	904	N 022/74
Vortex Lower	33	N 009/57
Lewis	1,244	N 017/57
Vortex Upper	157	N 013/70
Akivik	706	N 030/70
Tortured	755	N 024/65
Wedge	86	N 021/69
400	453	N 039/73
AC Block	158	N 032/65

Vein orientation measurements within each domain do not deviate significantly from the dominant overall vein orientation of N 024/68. As such, this formed the basis for selection of the search ellipse orientation for grade estimation during the 2006 resource estimation.

Page 9-7

Figure 9-3: Major Mineralized Corridors – Donlin Creek Resource Area (Piekenbrock and Petsel, 2003)

Page 9-8

10.0	EXPLORATION
	Three major exploration/drill programs were conducted within the resource area at Donlin Creek between 2002 and 2007: one by NovaGold (2002), the other by Placer Dome (2005) and the third by Barrick (2006 and 2007). Each program is discussed below.
10.1	2002 NovaGold
	NovaGold exploration work in 2002 consisted of diamond core drilling totalling 39,092 m (128,255 ft) and RC drilling totalling 11,589 m (38,022 ft) (see Figure 4-3). Drilling was concentrated in the western part of the resource area centered on ACMA. The purpose of the program was to extend known mineralized zones and potentially define new zones. An RC drill was used primarily to explore for zones of new mineralization with follow-up core drilling to better define the zones. The program was successful in expanding mineralization beyond the ACMA and 400 areas resulting in the discovery of the Aurora and Akivik zones. The initial step-out drilling was done on a rough 100 m by 100 m grid (328 ft by 328 ft) with a subsequent 50 m by 50 m drill spacing (164 ft by 164 ft) dependent on initial results. The majority of core holes were drilled to the north and inclined at 60° from horizontal to better intersect both sill contacts and vein zones.
	NovaGold also re-logged select pre-2001 core holes in order to develop an updated lithology and structure model. These data aided in planning the exploration drilling.
10.2	2005 Placer Dome
	The Placer Dome exploration work in 2005 consisted of diamond core drilling totalling 24,596 m (80,696 ft) and RC drilling totalling 3,644 m (11,955 ft). The majority of the core drilling was done in the Lewis and ACMA areas in order to reduce drill hole spacing in select areas containing inferred resource. A rough 80 m by 60 m grid (262 ft by 197 ft) for the new in-fill holes was used. This spacing proved optimal using the existing drill holes. Core holes were drilled westerly with an average inclination of 60° from horizontal to best intersect vein zones.
10.3	2006–2007 Barrick
	Primary exploration activities were core drilling contracted to Boart Longyear of Salt Lake City, Utah. The drilling grid has been reduced to 25 m to 35 m centers (82 ft to 115 ft). The exploration results compared favourably to the pre-existing geologic and gold grade estimation model.

Page 10-1

11.0	DRILLING
11.1	2002 NovaGold
	NovaGold completed drilling in ACMA and adjacent mineralized areas of the Donlin Creek project in 2002 using two types of drills. Core drilling totalled 39,092 m (128,255 ft) in 194 drill holes and RC drilling totalled 11,589 m (38,022 ft) in 147 holes.
	Core holes ranged in length from 17 m to 572 m (56 ft to 1,877 ft), averaging 201.5 m (661 ft). Drilling was done by wireline method using H-size equipment (HQ). Four core drill rigs were used. Drilling was well supervised, the sites were clean and safe, and work was efficiently conducted.
	Holes were primarily drilled at a declination of between 60° and 70°. Down-hole surveys were taken about every 30 m (98 ft) using a reflex camera.
	The RC holes ranged in length from 30.5 m to 140 m (100 ft to 459 ft), averaging 79 m (259 ft). One RC drill rig was used. The RC drilling was well supervised, the sites were clean and safe, and work was efficiently conducted.
	Drill hole collars were located respective to a property grid. Proposed drill hole collars were located using a Garmin GPS. Final and completed collars were surveyed with an Ashtech GPS utilizing post-processing software for ±0.1 m (±0.3 ft) accuracy. Coordinates were given in the UTM coordinate system.
	Standard logging and sampling conventions were used to capture information from the drill core and, where applicable, RC chips. The core was logged in detail using paper forms with the resulting data entered into the main database (Access© database) either by the logging geologist or a technician. Five types of data were captured in separate tables: Lithology, Mineralization, Alteration (visual), Structural and Geotechnical. Remarks were also captured. Lithology was recorded in a 2 to 4 letter alpha code. The Mineral table captured visual percent veining (by type) and sulphide (pyrite, arsenopyrite, stibnite and realgar). Specific alteration features including FeOx and carbonate alteration were also captured using a qualitative scale. Structural data consist of type of structure, measurements relative to core axis and oriented core measurements, if applicable. The Geotechnical table records percent recovery and RQD for the entire hole, and fracture intensity where warranted. The protocols and coding are similar to those used by Placer Dome during its drilling campaigns.
	In the fall of 2001, a preliminary study of alteration variability was undertaken on hand samples using a PIMA short-wave infrared spectrometer (SWIR). Based on those results, a PIMA was again utilized in early 2002 to ascertain alteration mineralogy in relation to detailed logging observations, as well as assay and geochemical results.

Page 11-1

	That study successfully demonstrated that SWIR spectrometry was efficient in defining alteration assemblages controlling the distribution of gold grade.
	A more serviceable, high throughput ASD SWIR spectrometer was subsequently used in 2002 in order to collect alteration data for the entire Donlin Creek resource area. Virtually all core holes within the ACMA, Aurora, 400 and Akivik areas, including core from previous drill campaigns, were analyzed using the spectrometer during the 2002 field season. A significant portion of drill core from the South Lewis and Vortex areas was also completed.
	Drill core was well handled and maintained. Data collection was competently done and found to be consistent from hole to hole and between different loggers. Core recovery in the intrusive units, both where mineralized and unmineralized, was excellent, usually mid 90s to 100%. Recovery in the shale dominant sediments was more variable, ranging from 80s to high 90s. Overall, the 2002 drill program and data capture were conducted in a competent manner.
11.2	2005 Placer Dome
	The 2005 Placer Dome drilling program at Donlin Creek utilized both core and RC drills. Core drilling totalled 24,596 m in 90 drill holes whereas RC drilling totalled 3,644 m in 30 holes.
	Core drilling focussed on in-fill drilling primarily in the ACMA and Lewis areas. Core holes ranged in length from 79 m to 544 m, averaging 273 m. Drilling was done by wireline method using H-size equipment (HQ). Three core drill rigs were used. Most holes were drilled at a declination of between 50° and 60°.
	The RC holes ranged in length from 102 m to 201 m, averaging 121.5 m. As in 2002, both drill programs were well supervised, the sites were clean and safe, and work was efficiently done.
	Down-hole and collar survey methods, logging conventions and data entry procedures were the same as used in 2002. Alteration data were again collected using an ASD SWIR spectrometer. Core recovery was excellent. Overall, the 2005 drill program and data capture were conducted in a competent manner.
11.3	2006–2007 Barrick Drilling
	2006
	In 2006 the project team drilled 92,804 m (304,475 ft) of core with eight LF-70 drill rigs in 327 drill holes. Of that, 235 holes totalling 84,800 m (278,215 ft) of core were

Page 11-2

focused on converting inferred resource to measured and indicated resource. However, significant drilling was also devoted to a broad range of pre-feasibility and feasibility objectives, including pit slope stability, metallurgy, waste rock studies, facilities condemnation and engineering, and calcium carbonate resource bulk sampling, delineation, and exploration. Drilling is summarized in Table 11-1. Barrick continued the same logging procedures and down-hole and collar survey methods as used in the past. However, ACE core orientation tools were utilized for oriented holes and the data were entered into an acQuire database. Core recovery in both mineralized and unmineralized rocks was consistently excellent and generally exceeded 90% in intrusive rocks and 80 to >95% in sedimentary rocks. True widths of mineralization are difficult to determine but drill holes were oriented to reflect true width intercepts limited only by the equipment and geologic knowledge.

Table 11-1: Core Holes Drilled in 2006

		Feet	Metres
Objective	Number	(ft)	(m)
Deep ACMA exploration	1	2,896	883
Infill resource conversion	223	259,871	79,229
Near-pit exploration (Akivik)	2	1,469	448
Waste dump condemnation	1	2,414	736
Waste (ABA)	8	10,411	3,174
Geotechnical	38	13,051	3,979
Port road geotechnical	8	768	234
Carbonate bulk sample (PQ)	14	1,036	316
Carbonate resource definition	20	6,891	2,101
Carbonate exploration	2	836	255
Metallurgical studies (PQ)	10	4,753	1,449
Total Core Holes	327	304,397	92,804

	2007
	Through October 3, 2007, 124 exploration holes comprising 50,562 m of core (165,886 ft) were completed. Core recovery in both mineralized and unmineralized rocks was consistently excellent and generally exceeded 90% in intrusive rocks and 80 to >95% in sedimentary rocks. True widths of mineralization are difficult to determine but drill holes were oriented to reflect true width intercepts limited only by the equipment and geologic knowledge.
11.4	Orientation of Mineralization
	Figure 9-3 illustrates the orientation of the generally steeply dipping mineralized corridors. Gold mineralization is primarily found at the intersection of the structural corridors with intrusive rock types.

Page 11-3

12.0	SAMPLING METHOD AND APPROACH
12.1.1	Logging
	Standard logging conventions adopted by Barrick were used to capture geologic data from both core and RC chips. The chips were logged on paper forms and the data entered into an electronic database. Core logging data were captured in five tables: lithology, mineralization, alteration (visual), structural, and geotechnical (percent recovery and rock quality designation, or “RQD”). The logging manual is found in Appendix A.
	In the fall of 2001, a preliminary study of alteration assemblages in intrusive rocks was conducted on hand specimens using a PIMA short-wave infrared spectrometer (SWIR). That study successfully demonstrated that alteration assemblages could be efficiently identified with SWIR spectrometry. A more serviceable, high-throughput analytic spectral device (ASD) SWIR was used in 2002 to collect alteration data for the entire Donlin Creek resource area. Virtually all core holes from the ACMA and Akivik areas, as well as a significant portion from south Lewis, were analyzed during the 2002 field season. All intrusive rock intervals drilled through 2007 were measured with the ASD SWIR.
12.1.2	Sampling
	The sampling protocol for the 2006 and 2007 programs was similar to that followed by Placer Dome and NovaGold in previous campaigns.
	Holes are sampled from the top of bedrock to the end of the hole. Overburden, excluding the organic layer, may also be sampled if abnormally thick and composed of abundant rock clasts. Geologists mark 2 m (6.6 ft) sample intervals through mineralized rock and variable width (up to 9.15 m or 30 ft) buffers of non-mineralized material adjacent to mineralized zones. Weakly mineralized to barren sedimentary rock can be marked on 3 m (9.8 ft) intervals at the discretion of the geologist. This sample interval approach is reliable and appropriate, and is used throughout the mining industry for this style of gold deposit. Sample intervals are broken at rock type contacts. An aluminum tag inscribed with the sample number is stapled to the core box with a same-numbered paper tag at each sample break. A sampling cutting list is generated that also specifies the insertion points for control samples.
	The core is then digitally photographed and split in half with an electric rock saw that uses water-cooled diamond saw blades. Core cutters orient the core in the saw to ensure a representative split. One-half of the core is returned to the core box for storage at site, and the other half is bagged for sample processing. In December and January (2007), a total of approximately 12,000 m (39,360 ft) of whole core was

Page 12-1

shipped to an off-site logging and core splitting facility in Anchorage. This facility was managed by Alaska Earth Science (AES) and staffed with both AES and Barrick personnel to ensure that logging, sampling, core splitting and sample shipment procedures were identical to those used at the Donlin site facility.

Page 12-2

13.0	SAMPLE PREPARATION, ANALYSES AND SECURITY
13.1	Drill Hole Sample Preparation
13.1.1	Prior to 2006
	Sample preparation, quality assurance and quality control for assays prior to 2002 were analyzed by AMEC and reported in their March 2002 Technical Report (Juras and Hodgson, 2002). Samples drilled in 2002 were analyzed by AMEC and reported in the first half of 2003. AMEC determined the assays and the database were suitable for resource estimation. No sample collection occurred within the mineral resource areas between 2003 and the beginning of 2005.
13.1.2	2006–2007
	Most of the core samples were crushed at the Donlin camp facility. Samples of core split in Anchorage were shipped to an ALS Chemex preparation lab for crushing, pulverizing and assaying.
	The Donlin camp preparation lab is housed in a heated steel building. A partition separates the core cutting area from the crushing area. The crushing side of the operation is equipped with a dust control ventilation system to minimize contamination. The camp sample preparation procedure consists of the following steps:
		The entire bagged sample is dried in an oven heated to between 85°C and 90°C for 12 hours.
		The sample is put into trays for processing through a jaw crusher. The sample tag stays with the sample.
		Blank samples (one of three QA/QC control samples) are inserted into the sample stream.
		The sample is crushed until the end product passes 70% minus 2 mm (10 mesh). Sieve analyses are performed periodically to check crush quality, and the crusher jaws are adjusted as necessary. Blank material is periodically used to clean the crushers, and operators are instructed to increase the cleaning frequency when unusually sulphide-rich material is processed.
		Sample is then passed through a riffle splitter four to six times to obtain a nominal 250 g (9 oz) split. This subsample is put into a numbered pulp bag, and the remainder, or coarse reject, is put back into the original sample bag. The splitter and sample pans are cleaned with compressed air.

Page 13-1

		Two additional control samples—standard reference material (SRM) and a duplicate—are inserted as specified on the cutting list prepared by the geologist. Control samples including SRM, duplicates and blank samples are included in every batch of 20. The blank is prepared by processing a sample from a bin of gravel-size crushed rock by passing it through the jaw crusher and riffle-splitting it to ~200 g (7 oz). When a duplicate is required, the crushed core sample is passed through the riffle splitter once, and each half is split repeatedly to obtain a ~200 g (7 oz) sample.
13.2	Sample Analysis
	Final sample preparation and chemical analysis for gold, sulphur and trace element suites were completed at ALS Chemex in Vancouver, an ISO9001:2000 accredited laboratory. The preparation consists of the following:
		The splits of crushed core were reduced to rock flour or “pulp” in a ring-and-puck grinding mill (to better than 85% passing minus 75 µm or 200 mesh).
		A 1 oz (30 g) sub-sample of the pulp was fire assayed primarily using a fire assay- atomic absorption spectroscopy (AAS) method. Prior to 2007, the primary analytical method was Au-AA23 with detection range of 0.005 to 10 g/t gold. In 2007 analytical methods switched primarily to Au-AA25 with a detection range of 0.01 to 100 g/t gold to take advantage of the higher top detection range.
		Samples that assayed >10 g Au/t were re-assayed with a fire assay-gravimetric method (NovaGold, 2002) or from 2005 through 2006 by “ore grade” fire assay- AAS method (Au-AA25) with a detection range of 0.01 to 100 g/t gold. In 2007, one sample exceeding 100 g/t was assayed via gravimetric method Au-GRA21 with a detection range of 0.05 to 1000 g/t gold.
		Significant drill hole composites exceeding 4 g/t gold received through October 3, 2007 (the assay cut-off date for the DC7a block model) are tabulated in Appendix B.
13.3	Donlin Security and Sample Transport
13.3.1	Project Site
	Core samples are transported from the field and are brought to the yard adjacent to the geology office and logging tents at the end of each drill shift. Core storage is secure because Donlin is a remote camp and access is strictly controlled. Unauthorized camp personnel have generally been excluded from the core cutting and sample preparation building, but strict access procedures were initiated following an

Page 13-2

	audit by Barrick of the sample preparation lab in mid-2006 (Heberlein, 2006). Assay splits of prepared core, along with the control samples, are packed in a shipping bag, secured with a numbered security seal, and sealed in boxes for shipment. The coarse rejects and remaining split core are returned to a storage yard south of the airstrip for long-term storage. The sample shipment procedure is as follows:
		Boxed assay splits are flown from the Donlin camp to Aniak airport via Vanderpool Flying Service.
		Samples are shipped from Aniak via Frontier Flying Service to the ALS Chemex lab facility in Fairbanks, Alaska. All sample shipments are accompanied by a Frontier Flying Service waybill. This allows each sample to be tracked from camp to ALS Chemex.
		The samples are logged into the ALS Chemex data system in Fairbanks before shipment to the ALS Chemex Vancouver (or other ALS Chemex facility), where they are pulverized and assayed. The Fairbanks lab returns a custody form that reports on the condition of security seals.
13.3.2	Anchorage Security and Sample Transportation
	The Anchorage logging and splitting facility was housed in a secure, dedicated, warehouse/office facility. Visitor access to the facility was strictly controlled by AES, the facility manager. Outside visitation for tours or purposes other than daily delivery or pick-up required advance approval by the Donlin Project Manager. Whole core shipped from camp to the facility was transported on Lynden Air Cargo. Lynden waybills and Barrick custody forms were used to track samples from camp to Lynden’s Anchorage airport facility and from there by Lynden trucks to the Anchorage logging facility. Split core samples shipped from camp to the ALS Chemex Fairbanks lab followed similar protocol. Bagged split core samples were tied into shipping bags and loaded into palletized supersacks. Supersacks were closed with numbered security seals and shipped on Lynden trucks to ALS Chemex in Fairbanks. Waybills aided tracking within the Lynden transport system, and ALS Chemex reported on the condition of security seals in the same manner as shipments from camp.
13.4	Assay Quality Assurance and Quality Control (QA/QC)
13.4.1	1995–2002 QA/QC Protocol
	Beginning with Placer Dome in 1995 and continuing with NovaGold through 2002, a systematic and comprehensive program of QA/QC has been employed for rock sampling and definition drilling programs at Donlin Creek. The QA/QC protocols include the random and blind insertion of the following:

Page 13-3

		standard reference materials (SRMs) to monitor the accuracy of lab results
		coarse reject duplicates to monitor analytical precision
		blank control samples to monitor contamination during sample preparation and analysis.
	From 1996 to 2002, SRMs were inserted at an average rate of one per 24 samples. Over the same period, coarse reject duplicates were inserted at an average rate of one per 24 samples, and blanks were inserted at an average rate of one per 25 samples. Almost all samples associated with SRM and blank control samples that returned values beyond acceptable tolerance limits were re-assayed until the control sample results were either acceptable or validated by duplication.
	Based on the results obtained from the comprehensive QA/QC programs at Donlin Creek, AMEC concluded in 2003 that the quality of the Donlin Creek drill database was sufficient for resource estimation. A comprehensive discussion of Donlin Creek sample QA/QC procedures and results can be found in the report, “Status Update for the Donlin Creek Pre-feasibility Study” (AMEC, 2003).
	A slightly modified QA/QC protocol was implemented in 2005. Three QA/QC samples—one blank, one coarse reject duplicate, and one SRM—were randomly inserted into every block of 20 sample numbers. Thus, in every block of 20 sample numbers there were 17 drill hole samples and three QA/QC control samples.
	The 20 sample blocks met two criteria:
		The blocks must begin with a sample number in which the penultimate digit is an EVEN number, and end with a sample number in which the penultimate digit is an EVEN number.
		The blocks must begin with a sample number that ends in 1.
	Example: XXXXevennumber1 - XXXXevennumber0, such as XXX021-XXX040, XXX041-XXX060, etc.
13.4.2	Standard Reference Material and Blank Material in 2005-2006
	Leftover SRM material (Std-C and Std-D) from the 2002 campaign was used at the beginning of the 2005 season. When these SRMs were depleted, additional reference material was purchased from Analytical Solutions (OREAS 6Pb and OREAS 7Pb) and CDN Laboratories (CDN-GS-3). After the 2005 season, two additional standards (Std-G and Std-H) were created from Donlin Creek coarse reject material. These two new standards, in addition to CDN-GS-3, were used during the 2006 season. The new

Page 13-4

	standards created from coarse reject material were certified by Barrick’s chief geochemist after industry accepted round robin assay and statistical analyses.
	Blank material consisting of washed river gravel was purchased from Anchorage Sand and Gravel for the 2005 season and the beginning of the 2006 season. In early to mid-2006, the blank material was changed to granite chips purchased from Lowe’s in Anchorage.
13.4.3	QA/QC 2005 Results
	Lab performance was checked by continuously monitoring control sample assays. Assay batches containing a control sample assay that exceeded two standard deviations from the accepted value were immediately re-assayed, and the original assays were replaced with assays that satisfied the QA/QC standard. An exception to this protocol was made for batches containing negligible gold values.
	All control samples functioned well. SRM monitoring kept assay accuracy within acceptable limits. Duplicate sample data indicated no bias in sample preparation and analytical precision. Blank assay data demonstrated negligible contamination in the sample preparation and assay processes.
13.4.4	QA/QC 2006 and 2007 Results
	The QA/QC protocol was identical to that used in 2005, except that the frequency of control samples was increased from 4% to 6%. No serious systematic preparation or analytical lab problems were detected.
	Blanks
	Two types of blank material were used, a product from Anchorage Sand and Gravel for the beginning of the 2006 season followed by granite chips purchased from Lowe’s for later in 2006 and 2007. Each blank type was analyzed for gold by two different assay protocols, Au-AA23 and Au-AA25. A threshold of three times the gold detection limit was used by NovaGold for failure analysis. Dividing the samples into two time frames, a total of 5 samples (0.1%) failed before September 2006 and 163 samples (4.6%) of blanks failed after September 2006. NovaGold recommends follow up to determine the suitability of the granite chips as blank material or whether a change in sample preparation has been the cause of the change in failure rate. The failures were all much less than the proposed range of gold cut-off grades and, in NovaGold’s opinion, the failures are not material.

Page 13-5

	Duplicates
	A total of 3,885 duplicate pairs obtained from coarse reject material were taken at Donlin Creek during 2006 and 2007. Successful or “passing” duplicates were identified by calculating the Absolute Relative Difference (ARD):
	ARD = [A-B] / 0.5 * (A + B)
	A criterion of ARD must be less than 0.2 for 90% of the pairs used by NovaGold. The percent passing was an acceptable 93%.
	Standards
	A total of 3,573 results from 15 different standard reference materials (SRM) were provided by Barrick. Barrick set acceptable limits for all standards at the best value +/- 2  standard deviations. Failed standards were to be re-assayed until the results were acceptable or validated by duplication. NovaGold was not provided with the information necessary to recalculate best values so our results are indicative.
	A total of 123 standards exceeded the range, and 58 (47%) were flagged with a comment regarding the failure. These comments commonly referred to a standard swap, or that the result failed by a calculated percentage but was “within acceptable limits”. The definition of acceptable limits should become a defined term in Donlin Creek LLC procedure manuals. Donlin Creek LLC is aware of the remaining uncommented 65 failed standards and will be addressing the issue shortly. NovaGold’s opinion is that the standards have been properly monitored and decisions regarding re-assay of failures have been reasonable.
	The details of the QA/QC analysis are included in Appendix C.
13.5	Specific Gravity Determinations
	Historically, only two specific gravity (SG) values were used in tonnage calculations: 2.65  g/cm3 for the intrusive units and 2.71 g/cm3 for the sediment units. Additional SG measurements were collected in 2006 to provide better coverage of deposit rock units and geographic subregions. Statistical evaluation of these new SG values shows that they are quite similar to the historical intrusive rock and sedimentary rock SG values. Therefore, the historical values were used for this mineral resource estimation.
	The following methodology is used to determine SG:
		Samples of whole core approximately 5 to 10 cm (2 to 4") in length are first weighed dry and then weighed in water. This is accomplished by removing the dry weighing tray assembly from the scale, placing the sample in the wire basket,

Page 13-6

hooking the basket to the scale, and submerging the sample in a five-gallon bucket of water. A small sample bag containing rock chips placed in a small basket midway along the wire assembly acts as a tare weight to compensate for removing the weighing tray. This makes it easier to return the scale to zero when changing from dry to wet measurements.

The formula for SG calculation is: Weight in Air/(Weight in Air – Weight in Water). The specific gravities are computed in acQuire once the weights are entered into the database.

Measurements are collected for all rock types at a minimum frequency of one sample from all logged rock type intervals and one sample every 15 to 20 m (49.2 to 65.6 ft) in the larger rock units. Mineralized rock takes precedence over unmineralized rock in a given rock type interval, but sufficient measurements of unmineralized material are also collected to document potential variability.

Page 13-7

14.0	DATA VERIFICATION
14.1	Prior to 2005 Campaign
	As a test of data integrity, the data used to estimate the January 2002 Donlin Creek mineral resources reported in the February and March 2002 Technical Reports (Juras, 2002, and Juras and Hodgson, 2002) were checked several ways. AMEC concluded that the assay and survey database used for the Donlin Creek mineral resource estimation was sufficiently free of error to be adequate for resource estimation.
14.2	2005 Campaign
	NovaGold conducted a 100% check of 2005 drill hole Au assays within the resource area against electronic assay certificates. An error rate of less than 1.5% was uncovered. NovaGold also checked the collar and down-hole survey data. Electronic down-hole survey files were read for the drill holes and compared to those stored in the resource database.
	NovaGold verified the integrity of the 2005 data and it is sufficiently free of error to be adequate for resource estimation.
14.3	2006 and 2007 Drilling Campaigns
	Drilling data were captured using acQuire software and stored in MS SQL Server. Geologic logs, collar and down-hole survey data were entered on site using acQuire data-entry objects. Assay data were imported directly from electronic files provided by the laboratories. The master Donlin database was moved from the Donlin camp to the Anchorage office in the middle of 2006, and about 50% of the 2006 assay data were imported directly into the master database in Anchorage.
	The acQuire database was converted from the standard acQuire data-model to the more robust acQuire “Corp” data-model to aid in data verification. Further verification of legacy data took place by Barrick when data were migrated to the new data-model.
	NovaGold verified the drill hole data in the following manner:
	Collar Surveys
	Collar survey information is transferred electronically from the electronic Ashtech survey instrument to the database, minimizing the chance of input error. The Ashtech output files and geologic logs were compared to 5 percent of the electronic collar surveys. There was one unexplained 20-cm discrepancy between the elevation file and the database. Strangely, the error rate against the geologic logs was much

Page 14-1

higher; most collar coordinates were off by several metres. NovaGold does not know the reason but it is possible that the geologic logs have the proposed coordinate and not the final coordinate written on them, or there are transcription errors from the electronic database to the paper log. NovaGold is satisfied that the collar surveys from the Ashtech data files are sufficiently error free to be used for resource estimation.

Down-hole Surveys

Down-hole surveys are transcribed by hand from the survey instrument to paper survey forms. The forms in turn are entered into the electronic database manually. Ten percent of the drill holes were checked and an unacceptable error rate of 4.4 percent was measured. The primary error was that the down-hole survey was omitted from the electronic database. Other errors were incorrect azimuth conversion, incorrect feet to metre conversion, incorrect depth and incorrect priority code. NovaGold recommends that the Donlin Creek LLC review their down-hole survey transcription protocols and complete a 100% check of the down-hole survey database. Despite the high error rate, the magnitude of the errors was small; therefore, in NovaGold’s opinion the impact on the estimation of grade will be minimal.

Assays

Electronic assay certificates made available by Barrick were merged into a single file and by matching on sample number were directly compared to the electronic database. For 2006 drilling, 70 percent of the assays were compared and an acceptable discrepancy rate of 0.4% was measured. For 2007, 99% of the assays were compared to the electronic assay certificates and a discrepancy rate of 1% was measured. Although an acceptable error rate, NovaGold recommends that the source of the discrepancies be identified. NovaGold believes that the assay database is sufficiently error free to be used for resource estimation.

Page 14-2

15.0	ADJACENT PROPERTIES
	Adjacent properties are not relevant for the review of the Donlin Creek project.

Page 15-1

16.0	MINERAL PROCESSING AND METALLURGICAL TESTING
	Key testwork programs on Donlin Creek ores have been conducted at a number of laboratories over a period of approximately seven years as directed by the project manager (primarily Barrick). Major programs at the bench-scale level were initiated in 2006 to test grinding, flotation, pressure oxidation (POX) and neutralization. In addition to bench-scale work, major pilot-plant runs were performed in flotation, POX and neutralization at the Barrick Technology Centre, SGS-Lakefield, G&T, and Hazen Research. Both levels of testwork were conducted to optimize process parameters and develop engineering information for use in pre-feasibility studies. NovaGold assumes that the work was completed by, or under the supervision of, a qualified person.
	The testing has shown that the ores require pre-treatment ahead of cyanidation to recover the gold. The preferred method of pre-treatment is POX of the sulphide concentrate produced from flotation. Overall gold recovery is estimated to be 89.8%, based on the combined life-of-mine (LOM) average of 92.9% recovery from flotation and 96.6% from POX of the concentrate.
	The metallurgical testwork, key results and recommendations are summarized below. Table 16-1 contains a summary of metallurgical reports.
16.1	Mineralogy
	Sulphur occurs primarily as pyrite and arsenopyrite. The pyrite contains only 10 to 20% of the gold, while arsenopyrite is the main carrier of gold in solid solution (sub- microscopic) form. The finest arsenopyrite has the highest gold grade.
	Little or no visible gold has been identified from all mineralogical investigations.
	Mercury (2 ppm), arsenic (2,800 ppm), antimony (80 to 90 ppm) and chloride (20 to 25 ppm) are present in the ore. Carbonate content is 3.3 to 3.4%. These elements all affect the process.
16.2	Direct Leach / CIL
	The ore is refractory to direct and carbon-in-leach (CIL) cyanidation and yields very low recoveries (<15%) from either leaching methodology. High gold recovery is only achieved by destruction of the sulphidic host matrix before cyanidation.

Page 16-1

Table 16-1: Summary of Metallurgical / Process Reports on Donlin Creek

Author	Report Date	Report Reference	Title
A. Lanfranco & Assoc.	2007 (Jun)	June 2007 Trial Report	Summary of Autoclave gas Testing Results
Air Pollution Testing	2006 (Oct)	APT Project BAR6243	Emission Testing Results of Autoclave
AMEC	2007 (May)	152628-001	Trade-off Study No. 1 – BIOX® Study
AMEC	2007 (May)	152628-002	Trade-off Study No. 2 – Comminution Circuit Study
AMEC	2007 (Apr)	152628-003	Trade-off Study No. 3 – Feasibility of Calcining CSS
AMEC	2007 (Apr)	152628-004	Trade-off Study No. 4 – De-Chlorination Treatment
AMEC	2007 (Apr)	152628-005	Trade-off Study No. 5 – AARL vs. Zadra Elution Processes
AMEC	2007 (Apr)	152628-006	Trade-off Study No. 6 – Roasting as an Alternation for Oxidation of
			Donlin Creek Flotation Concentrate
AMEC	2007 (Apr)	152628-007	Trade-off Study No. 7 – Pressure Oxidation (POX) of a Flotation
			Concentration vs. Pressure Oxidation of the Whole Ore (WOP)
AMEC	2007 (Jun)	Project No. 155096	Donlin Creek Gold Project Metallurgical Model Report
AMEC	2007 (Sep)	Project No. 155096	Review of 2006 Flotation Testwork
AMEC	2007 (Sep)	Project No. 155096	AMEC, Coarse Ore Storage: Covered Stockpile vs. Concrete Silos
AMEC	2007 (Sep)	Project No. 155096	AMEC, MF2 vs. Conventional Grinding Circuit
AMEC	2007 (Sep)	Project No. 155096	AMEC, Limestone Report
AMEC	2007 (Sep)	Project No. 155096	AMEC, Flat Bottomed Thickeners
Amtel	2004 (Aug)	Report 04-24	Gold Deportment in ACMA Composite of Donlin Creek
Amtel	2005 (Feb)	Report 05-01	Flotation of Au-Rich Arsenopyrite from Donlin Creek Ore
Amtel	2006 (Feb)	Report 06-01	Testwork on Donlin Creek Ores Using Novel Flotation Scheme
Amtel	2006 (Nov)	Report 07-03	Process Mineralogy of Donlin Creek Ores – Gold Deportment in PP
			Testwork Products
Amtel	2007 (Jul)	Report 07-04	Process Mineralogy of Donlin Creek Ores – Deleterious Elements in PP
			Testwork Products
Amtel	2007 (Jul)	Report 07-17	Variability Mineralogy of Donlin Creek Ores
Amtel	2007 (May)	Report 07-26	QA/QC on Sulphate Assaying of Donlin Creek Flotation Concentrates
Amtel	2007 (Jul	Report 07-28	Chlorine and Fluorine Assay Data of Donlin Creek Ores
Amtel	1996 (Nov)	Report 96-11	Deportment of Gold in the Donlin Creek Ores
Barrick Corporate	2007 (Sep)	N/A	2005 ACMA Intrusive Composite
Barrick Corporate	2007 (Sep)	N/A	2005 Lewis Intrusive Composite
Barrick Corporate	2007 (Sep)	N/A	2005 Sediment Composite
Barrick Corporate	2007 (Sep)	N/A	2006 Dec SGS PP Composite
Barrick Corporate	2007 (Sep)	N/A	2007 Jan SGS PP Composite
Barrick Corporate	2007 (Sep)	N/A	2007 Flotation Variability Samples (x98)
Barrick Corporate	2007 (Sep)	N/A	2006 Flotation Variability Samples (x12)
Barrick Corporate	2007 (Sep)	N/A	2006 Grinding Variability Samples (x155)
Barrick Corporate	2007 (Sep)	N/A	2007 Grinding Variability Samples (x149)
Barrick Corporate	2007 (Sep)	N/A	2006 Grinding PQ Core Variability Samples (x9)

Page 16-2

Author	Report Date	Report Reference	Title
Barrick Corporate	2008 (Sep)	N/A	2006 & 2007 Metallurgy Sample Plots
Barrick Technology	2007 (Apr)	59001-01	Donlin Creek Pre-feasibility Study Pressure Oxidation Pilot-plant
Centre			Results
Barrick Technology	2007 (Apr)	59001-02	Donlin Creek Batch Pressure Oxidation Tests Summary : Pre-feasibility
Centre			Study
Barrick Technology	2007 (Apr)	59001-03	Donlin Creek Feasibility Pressure Oxidation Piloting – Phase 1
Centre
Barrick Technology	2007 (May)	59001-04	Donlin Creek Pre-feasibility Neutralization Piloting
Centre
Barrick Technology	2007 (Mar)	59001-05	Summary of Donlin Creek Feasibility Study Batch Neutralization Test
Centre			Results
Barrick Technology	2007 (Apr)	59001-06	Pilot-plant CIL Cyanidation of Donlin Creek Concentrate After Pressure
Centre			Oxidation
Barrick Technology	2007 (May)	59001-07	Donlin Creek Feasibility Pressure Oxidation Product CCD and
Centre			Neutralization Piloting - Phase 1
Barrick Technology	2007 (Aug)	59001-08	Donlin Creek Feasibility Pressure Oxidation Piloting – Phase 2
Centre
Barrick Technology	2007 (Sep)	59001-09	Donlin Creek Batch Pressure Oxidation Summary – Feasibility Study
Centre
Barrick Technology	2007 (Aug)	59001-10	Donlin Creek Feasibility Phase 2, Pre-Acidification and Neutralization
Centre			Bench Test Results
Barrick Technology	2007 (Aug)	59001-11	Donlin Creek Feasibility Phase 2, Neutralization Pilot-plant
Centre
Barrick Technology	2007 (Aug)	59001-12	Donlin Creek Feasibility Miscellaneous Testwork (CIL, Detoxification,
Centre			Mercury Precipitation)
Barrick Technology	2007 (May)	Memorandum	Confirmatory Cyanide Detoxification Tests on Donlin Creek CIL Tails
Centre			from Feasibility Study
Barrick Technology	2007 (Jul)	NA	Mineralogy on Concentrate, Pre-acid Conc and AC discharge DC
Centre
Canadian Light Source	2007 (Sep)	CLSI Contract #5025	X-Ray absorption Near-Edge Structures (Xanes) Spectroscopy on
			Barrick Tailings Samples
Delkor	2006 (Nov)	Test Report 069301	Delkor Testwork on Thickening & Filtration Applications, Donlin Creek
			Autoclave Feed
Dorr-Oliver Eimco	2004 (Oct)	NA	Report on Testing of Donlin Creek Tailings applying Deep Cone Paste
			Thickening for Tailings Disposal
Dorr-Oliver Eimco	2006 (Oct)	NA	Report on Testing for Barrick Gold Donlin Creek Project Sedimentation
			and Rheology Tests on Autoclave Feed and Stage 1 CCD Feed
Dorr-Oliver Eimco	2006 (Sep)	NA	Summary of Thickening Results Donlin Creek Concentrates
Dorr-Oliver Eimco	2007 (Apr)	NA	Sedimentation and Rheology Tests on Flotation Concentrate,

Page 16-3

Author	Report Date	Report Reference	Title
			Neutralized Tailings and Flotation Tailings
Dorr-Oliver Eimco	2007 (Jul)	NA	Sedimentation and Rheology Tests PP5 and PP10 Combined Tailings
Dorr-Oliver Eimco	2007 (Aug)	NA	Summary of Thickening Results of Autoclave Hot Cure Slurry
Dorr-Oliver Eimco	2007 (Sep)	NA	Thickener Testing on Concentrate, AC Discharge and Float Tails
Dynatec Corporation	2004 (Nov)	217-RS-101-000-R0	Donlin Creek Project Concentrate Pressure Oxidation, Cyanidation and
			Neutralization Studies
Environmental	2004 (Aug)	2509/649	Acid Forming Characteristics of Waste Rock and Low Grade Ore
Geochemistry Int Pty Ltd			Samples from the Donlin Creek Project, Alaska
G&T Metallurgical	2004 (Jul)	KM 1523 / P0407	Production of Concentrates from Donlin Creek Ores
G&T Metallurgical	2006 (Feb)	KM 1708	A Preliminary Metallurgical Study Model and Flotation Test Work
G&T Metallurgical	2006 (Mar)	KM 1787	A Preliminary Assessment of Flotation Response
G&T Metallurgical	2006 (Aug)	KM 1806	A Preliminary Assessment of Flotation Performance
G&T Metallurgical	2006 (Oct)	KM 1852	The Flotation Response of Donlin Creek Mineralization – Pilot-plant
			Testing
G&T Metallurgical	2006 (Oct)	KM 1904	An Assessment of Metallurgical Response of the Blended Ore
G&T Metallurgical	2007 (Mar)	KM 1931	Metallurgical Assessment of Aging on Donlin Creek Ore
GeoBiotics	2006 (Mar)	N/A	Conceptual Capital and Operating Cost Estimates GeoBiotics’
			GEOCOAT Heap Biooxidation Technology for Barrick Gold Corporation
			Donlin Creek Project
GeoBiotics	2006 (May)	N/A	Preliminary Assessment of GeoBiotics’ GEOLEACH™ Technology for
			Barrick Gold Corporation Donlin Creek Project
Gold Fields Ltd	2006 (Dec)	DCK OOM Rev-A.06	Order of Magnitude Capital Cost Estimate for BIOX Plant
Gold Fields Ltd	2007 (Aug)	GFL 097/06/07	Progress Report for Phase 2
Hatch	2007 (Aug)	H322685	MetSim Model Design Criteria
Hatch	2007 (Jul)	H322685-00-GMP-F-0001	DC Feasibility Study Autoclave Design / Residence Time Control
Hatch	2007 (Aug)	H322685-00-GMP-F-0002	Pressure Oxidation and Carbon in Leach Gold Recovery
Hatch	2007 (Jul)	H-322685-00-GRP-F-0001	Mercury Reduction Technology Assessment
Hatch	2007 (Sep)	H322685-00-GRP-N-0002-	Hatch Area 17 and 33 Report
		Rev 0
Hatch	2007 (Jul)	H322685-17-GMP-F-0001	Revised Decant Return Water Recycle Assessment
Hatch	2007 (Jul)	H322685-17-GMP-F-0002	Autoclave Off-Gas Mercury Removal Process Description
Hatch	2007 (Aug)	H322685-DC-F-0001_2_V4	Autoclave Design Criteria
Hatch	2006 (Nov)	H-322685-MP-0001-CA01	Autoclave Feed Pre-Acidulation Trade-Off Study
Hatch	2007 (Mar)	H322685-MP-CA01-1000	Factored Autoclave Capital Cost Estimate
Hatch	2006 (Nov)	H-322685-RPT-0007-CA01-	Autoclave Trade-Off Study One vs. Two Trains
		R0
Hatch	2007 (May)	H-322685-RPT-0007-CA01-	Autoclave Trade-Off Study One vs. Two Trains
		R1
Hatch	2006 (Dec)	H-322685-RPT-0008-CA01	Autoclave Sizing Comparison

Page 16-4

Author	Report Date	Report Reference	Title
Hatch	2007 (Apr)	H322685-RPT-CA01-10001	Minor Species Dissolution Study
Hatch	2007 (Jul)	Rev 3, 27 July 2007	MetSim Feasibility Model (Summer, average)
Hatch	2007 (Jul)	Rev 3, 27 July 2007	MetSim Feasibility Model (Winter, average)
Hatch	2007 (Jul)	Rev 3, 27 July 2007	MetSim Feasibility Model (Summer, maximum)
Hatch	2007 (Jul)	Rev 3, 27 July 2007	MetSim Feasibility Models (Winter, maximum)
Hazen Research	2007 (Aug)	Hazen Project 10367-05	N2TEC Flotation of Donlin Creek Ore
Hazen Research	2006 (Aug)	Hazen Project 10367-06	Microscopic Examination of Flotation Concentrates – Memo
Hazen Research	2007 (Jan)	Hazen Project 10367-06	Re-flotation of Donlin Creek Pilot-plant Tails
Intec Ltd	2005 (Dec)	N/A	Scoping Leach Trials Via The Intec Gold Process Utilising CIL, Donlin
			Creek Project
JKMRC	2007 (Jun)	JKTech Job No. 06148	Flotation Circuit Analysis of the Donlin Creek Pilot Campaign (First)
JKMRC	2007 (Jul)	JKTech Job No. 06430	Flotation Circuit Analysis of the Donlin Creek Pilot Campaign (Second)
Jones, Hackl	1997 (Dec)	N/A	Continuous Bacterial Leaching and Cyanidation of a Donlin Creek
			Concentrate
Larox	2006 (Nov)	Test No. 15567T1	Filter Tested Slurry – Donlin Creek Low Grade Sulphide Concentrate
Minnovex	2005 (Jun)	NA	Review of Donlin Creek Flotation Testing
Mutis Liber (Mike	2007 (May)	Report No. 155	Donlin Creek Kiln and Gold Room Off-Gas Estimation Study
Adams)
Orway Mineral	2006 (Aug)	Report No. 46343	Donlin Creek Comminution Circuit Option Study
Consultants
Orway Mineral	2006 (Nov)	Report No. 6456	Donlin Creek Comminution Circuit Option Study Phase II
Consultants
Outokumpu Technology	2007 (Jan)	N/A	Thickening of Acidified and Non-Acidified Flotation Concentrates
Outokumpu Technology	2006 (Jul)	TH-0379	High Rate Thickening Flotation Concentrate and Tailings Samples
Outokumpu Technology	2006 (Sep)	TH-0383	Thickening of Concentrate and Autoclave Discharge
Outokumpu Technology	2006 (Oct)	TH-0387	Thickening of Neutralization Discharge
Outokumpu Technology	2007 (Jun)	TH-0403	High Rate Thickening of Flotation Tailings
Outokumpu Technology	2007 (Apr)	TH-0407	High Rate Thickening of Flotation Tailings, Concentrate, Acidified
			Concentrate and Autoclave Product
Placer Dome Research	1999 (Feb)	File No. E98012N	Static Net Acid Generation Tests
Placer Dome Research	1997 (Nov)	N/A	Donlin Creek Project Autoclave Pilot-plant Testwork Report No. 1
Placer Dome Research	1995 (Nov)	N/A	Donlin Creek Project Bench Scale Grinding, Cyanidation, CIL, Flotation
			and Pressure Oxidation Tests Report No. 1
Placer Dome Research	1996 (Nov)	N/A	Donlin Creek Project Compilation of Results of 1996 Metallurgical Test
			Programme Report No. 2
Placer Dome Research	1999 (Dec)	N/A	Donlin Creek Project Bench Scale Grinding, CIL, Flotation, and
			Pressure Oxidation Tests on ACMA, Rochelieu, North Lewis, and South
			Lewis Ore Zones Report No. 3
Placer Dome Research	2006 (May)	T1010C11-02	Determination of the Neutralization Capacity of Calcareous Sandstones

Page 16-5

Author	Report Date	Report Reference	Title
			and Flotation Tailings
Polysius Research	2005 (Nov)	Project 2337 2152	High-Pressure Grinding Tests on Gold Ore
Centre
R&C Environmental	2006 (Nov)	NA	Laboratory Evaluation of the SO2 / Air and Prussian Blue Process
Services
RDi Resource Devel.	2002 (Nov)	Project No. 02-018	Flotation and Leaching of Donlin Creek Sample
Inc.
SGS (Minnovex)	2007 (Aug)	LR 11328-007 (Design) and	Grinding Circuit Study for the Donlin Creek Project
		11490-001 (Geostatistics)
SGS Lakefield	2004 (Sep)	LR 10044-104	An Investigation into the Grindability Characteristics of Two Samples
Research Ltd			from the Donlin Creek Deposit – Progress Report
SGS Lakefield	2005 (Jan)	LR 10044-154 Report 1	The Grindability Characteristics of Samples (CSS) from the Donlin
Research Ltd			Creek Deposit
SGS Lakefield	2005 (Feb)	LR 10829-001 Report 1	The Separation of Arsenopyrite and Pyrite from a Donlin Creek Intrusive
Research Ltd			Composite
SGS Lakefield	2007 (Feb)	LR 11328-001	The Variability in Comminution Characteristics of Samples from the
Research Ltd			Donlin Creek Deposit
SGS Lakefield	2006 (Dec)	LR 11328-001 Report 141206	Geostatistical Analysis and Estimation of Grindability Data from the
Research Ltd			Donlin Creek Deposit
SGS Lakefield	2007 (Mar)	LR 11328-001 Report 2	An Investigation into the Cyclone Separation of a Flotation Scavenger
Research Ltd			Concentrate
SGS Lakefield	2007 (Sep)	LR 11328-001 Report 3	An Investigation into the Production of Mineralogy Samples Via
Research Ltd			Flotation of the Donlin Creek Composites
SGS Lakefield	2007 (Sep)	LR 11328-003	Environmental & Geotechnical Testing of Donlin Creek Tailings
Research Ltd
SGS Lakefield	2007 (Sep)	LR 11328-004	Geochemical Characterization – BTC 2006 Pilot Autoclave Test
Research Ltd			Program
SGS Lakefield	2007 (Nov)	LR 11328-005	An Investigation into Proposed Grinding Circuit for the Donlin Creek
Research Ltd			Deposit Based on Small-Scale Tests
SGS Lakefield	2007 (Apr)	LR 11328-006	Environmental Testing of Donlin Creek Pilot-plant Feed Samples
Research Ltd
SGS Lakefield	2007 (Apr)	LR 11328-006 LIMS No.	An Investigation by QemScan into the Mineralogical Characteristics of
Research Ltd		MI5011-JAN07 Report 1	Pilot-plant Product Streams from the Donlin Creek Project
SGS Lakefield	2007 (Sep)	LR 11328-006 Report 1	A Preliminary Pilot-plant Investigation into the Recovery of Gold from a
Research Ltd			Donlin Creek Composite
SGS Lakefield	2007 (Jun)	LR 11328-006 Report 2	An Investigation into the Metallurgical Performance of the Donlin Creek
Research Ltd			Composites
SGS Lakefield	2007 (Apr)	LR 11328-006 Rev 1	An Investigation into the FLEET Modelling of the Donlin Creek Flotation
Research Ltd			Circuit

Page 16-6

Author	Report Date	Report Reference	Title
SGS Lakefield	2007 (Apr)	LR 11328-007	The Variability in Comminution Characteristics of Samples from the
Research Ltd			Donlin Creek Deposit
SGS Lakefield	2007 (Aug)	LR 11328-008 Report 1	An Investigation into the Recovery of Gold from Variability Samples
Research Ltd
SGS Lakefield	2007 (Sep)	LR 11328-008 Report 1	A Summary of Batch Neutralization Test Results for the Donlin Creek
Research Ltd			Project Variability Study
SGS Lakefield	2007 (Sep)	LR 11328-008 Report 4	Summary of Results – Mineralogical Variability in Donlin Creek Ore
Research Ltd			Samples
SGS Lakefield	2007 (Apr)	LR 11328-009	Environmental Testing of Donlin Creek Phase 1 Pilot-plant 7 Flotation
Research Ltd			Samples
SGS Lakefield	2007 (Jun)	LR 11328-010	Environmental Testing of Donlin Creek Phase 2 Pilot-plant Feed
Research Ltd			Samples
SGS Lakefield	2007 (Sep)	LR 11328-010 Report 1	A Pilot-plant Optimization Study on the Recovery of Gold from a Donlin
Research Ltd			Creek Composite
SGS Lakefield	2007 (Sep)	LR 11328-010 Report 2	An Investigation into the Development of a Flowsheet for the Donlin
Research Ltd			Creek Composites
SGS Lakefield	2007 (Jun)	LR 11328-011	Continuance/Decommissioning of MWMP Residue Humidity Cell Tests
Research Ltd			& Test Data
SGS Lakefield	2007 (May)	LR 11328-011-MI5008-	An Investigation into Mineralogical Characterization of Three Final
Research Ltd		MAR07	Tailings Samples
SGS Lakefield	2007 (Aug)	LR 11328-011-MI5023-	An Investigation into the Mineralogical Characterization of a Final
Research Ltd		MAY07	Tailings Sample,
SGS Lakefield	2007 (May)	LR 11328-013	Geochemical Characterization – SGS Pilot Flotation PP-03 Survey
Research Ltd			Samples – February 2007 Pilot – Conventional
SGS Lakefield	2007 (May)	LR 11328-014	Geochemical Characterization – SGS Pilot Flotation PP-12 Survey
Research Ltd			Samples – February 2007 Pilot – MF2
SGS Lakefield	2007 (Aug)	LR 11328-015	Geochemical Characterization – BTC February 2007 Pilot-plant Tails
Research Ltd			Samples
SGS South Africa	2006 (Nov)	BIOMET 06/18	Bacterial Oxidation Amenability Testing of the Donlin Creek
			Concentrate – Phase 1
Signet Technology Inc	1999 (Mar)	N/A	Process Design and Cost Study for Bacterial Oxidation of Concentrates
			at Donlin Creek
Surface Science	2006 (Nov)	SSW Ref. 59606brh.BAR	Evaluation of Particle Surfaces using ToF SIMS
Western
Surface Science	2007 (May)	SSW Ref. 90906brh.BAR.1	Statistical Differentiation of the Surface Species from a Pilot-plant Study
Western
Surface Science	2007 (Jul)	SSW Ref. 90906brh.BAR.2	Statistical Differentiation of the Surface Species of Primary Mill
Western			Discharge Samples.
University of Alberta	2004 (Jun)	N/A	Separation of Pyrite and Arsenopyrite – A Literature Review

Page 16-7

Author	Report Date	Report Reference	Title
University of British	2006 (Oct)	N/A	Rheology Testing on Donlin Creek Samples (Hot Cure, CN Detox
Columbia			Slurry)
University of British	2007 (Mar)	N/A	Rheology Testing of Donlin Creek Samples (Hot Cure, CIL Feed)
Columbia
University of British	2007 (Jun)	N/A	Rheology Testing of Donlin Creek Samples (Hot Cure, Pre-acidified
Columbia			Conc)
University of British	2007 (Jul)	N/A	Rheology Testing of Donlin Creek Samples (CN Detox, Neut Product)
Columbia
University of British	2005 (Nov)	Project Number T1010A03	Quantitative Phase Analysis of Four Pulp Samples (CSS) Using the
Columbia			Rietveld Method and X-ray Powder Diffraction Data.
University of Cape Town	2007 (Jun)	MPTech C-06-20	Donlin Creek Gold Ore Use of Reactivity Number and Nitrogen based
			flotation

Page 16-8

16.3	Crushing / Grinding
	More than 300 samples from the deposit have been bench-tested for hardness.
	Geostatistical analyses have been undertaken on the complete hardness dataset. The resulting hardness model was used for circuit design and capacity estimation.
	The ores are considered moderately hard, with an average Ball Work Index of 13.61 kWh/st (15 kWh/t) (Table 16-2). The ores are amenable to SAG milling.
	The hardness properties of core samples that have been exposed to the freeze/thaw cycle of an Alaska winter are considered unreliable because of alteration and cannot be used on their own (without proper calibration) for the purpose of grinding circuit design.

Table 16-2: Estimated Bond Ball Work Index (BWI)

	BWI Mean
Domain	(kWh/t)	Std. Dev.	Sample Count
RDX + RDF + MD	15.5	1.3	107
GWK + SHL	14.6	1.5	48
RDA	13.8	1.3	46
RDXL	14.4	1.1	33
RDXB	16.4	1.2	39
All	15.0	1.5	273

16.4	Flotation
	Extensive bench flotation testwork has been undertaken. Flotation recoveries are highest from intrusive ores, at 94.7 to 97.5%, lower from the sedimentary ores at 89.7 to 91.3%, and problematic for partially geologically oxidized or altered ores, averaging 75.7% (see Table 16-3 and Figure 16-1).
	Testwork confirmed the very close relationship between arsenic and gold recovery.
	An MCF2 (mill chemical float, twice) style flowsheet with 100 minutes of residence time increases the gold recovery to a 7% sulphur flotation concentrate by an estimated 1.8%.
	CIL leaching of the flotation tails does not yield economical gold recovery.

Page 16-9

Table 16-3: Summary of Final Flotation Recovery by Geological Domain (without oxide)

	Tonnes	Adjusted Recovery* to MCF2 Pilot Result
	(%)	(%)
AKIVIK	6.2	97.48
400	7.3	96.84
ACMA	16.0	96.32
AURORA	5.3	96.09
VORTEX	13.0	95.05
LEWIS	26.7	94.74
GWK	21.8	91.33
SHL	3.9	89.71
Overall	100.0	94.49
* Not corrected for oxide content

Figure 16-1: Variability of Flotation Gold Recovery Results for Oxidation Samples

16.5	Pressure Oxidation (POX)
	Dynatec and Barrick Technology Center have completed numerous pressure oxidation bench-testing programs. Three separate autoclave pilot test programs, each incorporating a number of different test runs, were conducted in 2006 and 2007.
	POX allows for 96.6% recovery of the gold in CIL following oxidation.

Page 16-10

	A counter-current decantation (CCD) washing circuit has been incorporated ahead of pressure oxidation to control soluble species from the concentrator in the feed to the autoclave.
	Autoclave gas emissions will pass through a mercury recovery system before release to the atmosphere.
16.6	CIL / Gold Recovery
	CIL processing of the washed autoclave product produces optimized gold recoveries of 96 to 97% and requires low amounts of cyanide.
	Lime consumption by the CIL feed can be minimized by operating the CIL circuit at a pH of ~9.0 to avoid precipitation of magnesium hydroxide.
	Ensuring high-efficiency washing of the autoclave product, preferably to 98% or more, minimizes lime and cyanide reagent addition in CIL.
	A Cherokee Chemical Company UNR reagent can be used to control mercury leached into solution from the cyanidation of the autoclave product (mercury that is leached by cyanide and does not adsorb onto carbon) to low levels within the recirculating process water streams.
	Mercury recovery systems for off-gas/vapour streams have been designed for the carbon regeneration kiln, electrowinning, retort system and smelting furnaces.
16.7	Environmental Considerations
	The high temperature POX process is generally considered best practice for the generation of stable arsenic compounds suitable for long-term disposal in a tailings storage facility, depending on the amount of iron available for co-precipitation with soluble arsenic. The Donlin Creek mill feed has sufficient iron content to provide the recommended minimum stoichiometric ratio of 4:1 iron to arsenic. Iron is mainly associated with pyrite, arsenopyrite and ferroan dolomite; iron content as siderite is minor.
	The tailings decant water from the process plant will likely contain levels of arsenic, mercury, magnesium, molybdenum, selenium and antimony above applicable standards. The tailings water will also be elevated in sulphates (greater than 10 g/L), particularly due to the presence of magnesium, which increases the solubility level of sulphate in solution.
	Recycling of tailings decant waters will be considered.

Page 16-11

	Testing has been undertaken to confirm the use of the Air/SO2 process for cyanide destruction in the CIL tails before being blended with the neutralization circuit. The backup use of soluble iron sulphate for the destruction of weak acid dissociable cyanide (WAD CN) has also been shown to be effective.
	All plant tailings will be neutralized before discharge to the tailings storage facility.
	The ABA and humidity cell testing results to date indicate that the final plant tailings stream will be maintained as non-acid-generating.
16.8	Conceptual Process Plant Design
	The design criteria for the conceptual process plant facilities are primarily based on testwork results stemming from pilot-plant runs and variability testwork programs. Modern modelling techniques were utilized in the design of the grinding, autoclave and flotation circuits. SGS Lakefield developed a CEET (Comminution Economic Evaluation Tool) model for the Donlin Creek project to determine the power requirements of the grinding circuit; Barrick and JK Tech developed a JKSimFloat model to size the flotation circuit. In addition, various aspects of the process design were benchmarked to compare the Donlin Creek project with other similar operations.
	The design of the Donlin Creek process plant is based on the most current, modern- day technology for both the process circuits and equipment selection. Particular attention was paid to incorporate state-of-art technology—such as mercury abatement systems to control mercury—for safety and environmental protection.
	The conceptual process plant is designed to recover a sulphide flotation product and oxidize the refractory gold concentrate in a pressure oxidation circuit prior to cyanidation. Highlights of the process plant are as follows:
		The gyratory crusher with a capacity for 45,000 t/d (49,600 st/d).
		The design is based on the MCF2 grinding and flotation circuit. The grinding circuit is SABC with a single SAG mill in closed circuit with parallel cone crushers followed by primary ball mill in closed circuit with cyclones. Primary ball mill product reports to primary rougher flotation. Rougher flotation tailings report to the secondary ball mill circuit, while in closed circuit with cyclones. Secondary ball mill product at P80 50 µm reports to secondary rougher flotation. Secondary rougher flotation concentrate reports to cleaner flotation. A cleaner scavenger flotation circuit treats the cleaner flotation tailings.
		Combined flotation concentrates from primary rougher and cleaner flotation are dewatered in a thickener before acidulation and CCD washing to remove

Page 16-12

solubilized species from the concentrator and acidulation portions of the process. It has been shown that high levels of soluble species in the feed to the autoclave have detrimental effects on the POX/CIL gold recovery.

The autoclave circuit includes two autoclaves operating in parallel. The autoclaves are designed to operate at 437°F (225°C) with a retention time of 50 minutes.

Thickened flotation tailings slurry is used as a cooling medium in the autoclave letdown circuit.

Flotation tailings are combined with the flotation concentrate wash solution product to neutralize the acidic solution before discharge to the tailings storage facility (TSF). The carbonate in the flotation tailings slurry will provide primary neutralization. The final pH level will be adjusted by adding slaked lime.

Flashed and cooled autoclave discharge slurry is cured for 6 hours at 212°F (100°C) before POX discharge CCD. The acidic solution recovered by CCD is recycled to acidulation and flotation feed conditioning.

POX CCD product slurry should be neutralized with lime ahead of cyanidation.

The CIL circuit will operate at pH 9.0. The CIL tanks should be fully enclosed and vented to a caustic scrubbing system to recover cyanide and recycle it back to the CIL circuit. The CIL circuit has a residence time of 24 hours. Carbon will be handled with in-tank revolving screens.

The carbon handling area for the loaded carbon consists of an acid wash circuit and a modified pressure Zadra circuit for stripping carbon. Carbon will be reactivated in an electric kiln.

Gold will be recovered in an electrowinning circuit. The electrowinning sludge will be treated in a retort before being melted in an induction furnace.

Mercury that evolves in the process plant will be captured in a number of mercury abatement systems, which will treat the following streams: autoclave flash vent, regeneration kiln feed and discharge vents, electrowinning vents, retort furnace exhaust, induction furnace vent and general refinery area ventilation stream.

Page 16-13

17.0	MINERAL RESOURCE AND MINERAL RESERVE ESTIMATES
17.1	Introduction
	The mineral resource estimates for the Donlin Creek project were calculated by Barrick in a model referred to as “DC7a”. NovaGold has independently verified and validated the results. Kevin Francis, P.Geo of NovaGold is the qualified person regarding this work.
	The Donlin Creek model (project area) was built using geological information and assay data collected from 1,000 core holes totalling 286,120 m (938,714 ft), 296 RC holes totalling 29,294 m (96,109 ft) and 21,897 m of trenching. A total of 160,041 assays were used in the model. The estimates were made from a 3D block model utilizing Vulcan™ mine modelling software. The cell size for the model is 6 m east by 6 m north by 6 m high.
	There are presently no mineral reserves at Donlin Creek.
17.2	Estimation Approach
	At Donlin Creek, the felsic dykes and sills are the main host for gold mineralization. In Lewis, and to a lesser extent at ACMA and Aurora-Akivik, the hosting sediments can also contain gold mineralization. As gold mineralization is not pervasive in these host units (the mineralization only occurs in areas which have been structurally prepared), Barrick chose to define mineralized envelopes within intrusive and sedimentary rocks using probability-assisted domaining. Sulphur was constrained in a similar manner. Once the mineralized envelopes are defined, grade is estimated outside and within the envelopes using ID3.
17.3	Geologic Model
	Three-dimensional (3D) solids of the geological model were generated from polygon interpretations constructed from cross-section and level plans. Tools available in Vulcan were used to create the polygons, which were digitized directly on a computer screen snapping to drill holes in section. Once digitized, the polygons were used to develop 3D wireframe solids to incorporate geologic control into the grade model for the intrusive rocks.
	The solids were used to assign the corresponding geological code to the 3D block model. Priorities were assigned to each solid depending on the timing of the dyke or sill and to account for any overlapping. To limit the size of the model, blocks were assigned a default code of greywacke (rock = 93) and were then overprinted with rock

Page 17-1

values according to the established priorities specified in Table 17-1. Rocks assigned a greywacke code had the lowest priority value.

Page 17-2

Table 17-1: Three-Dimensional Solids Used in the Modelling Process

		Block		Block	Composite	Composite		Rock
Triangulation Name	Block Variable	Value	Block Variable	Value	Field	Value	Domain	Type	Priority
other_1a.00t	rock	1	rock_mine	1	ore	1	ACMA	Intrusive	5
other_3b.00t	rock	2	rock_mine	1	ore	2	Lewis	Intrusive	5
other_5a.00t	rock	3	rock_mine	1	ore	3	Akivik	Intrusive	5
other_8.00t	rock	4	rock_mine	1	ore	4	400	Intrusive	5
other_9.00t	rock	5	rock_mine	1	ore	5	Aurora	Intrusive	5
							L.
rda_dyke_2.00t	rock	6	rock_mine	2	ore	6	Vortex	Intrusive	45
rda_dyke_3c.00t	rock	7	rock_mine	2	ore	7	Lewis	Intrusive	45
rda_dyke_4a.00t	rock	8	rock_mine	2	ore	8	Vortex	Intrusive	45
rda_dyke_4b.00t	rock	9	rock_mine	2	ore	9	Vortex	Intrusive	45
rda_dyke_4c.00t	rock	10	rock_mine	2	ore	10	Vortex	Intrusive	45
rda_rdxb_sill_9.00t	rock	11	rock_mine	2	ore	11	Aurora	Intrusive	50
rda_sill_1a.00t	rock	12	rock_mine	2	ore	12	ACMA	Intrusive	40
rda_sill_3c.00t	rock	13	rock_mine	2	ore	13	Lewis	Intrusive	40
rda_sill_3d.00t	rock	14	rock_mine	2	ore	14	Lewis	Intrusive	40
rda_sill_4a.00t	rock	15	rock_mine	2	ore	15	Vortex	Intrusive	40
rda_sill_5b.00t	rock	16	rock_mine	2	ore	16	Akivik	Intrusive	40
rda_sill_6.00t	rock	17	rock_mine	2	ore	17	Wedge	Intrusive	40
rda_sill_8.00t	rock	18	rock_mine	2	ore	18	400	Intrusive	40
rda_sill_9.00t	rock	19	rock_mine	2	ore	19	Aurora	Intrusive	40
rdf_dyke_3a.00t	rock	20	rock_mine	3	ore	20	Lewis	Intrusive	15
rdf_dyke_3b.00t	rock	21	rock_mine	3	ore	21	Lewis	Intrusive	15
rdf_dyke_3c.00t	rock	22	rock_mine	3	ore	22	Lewis	Intrusive	15
rdf_dyke_3d.00t	rock	23	rock_mine	3	ore	23	Lewis	Intrusive	15
rdf_dyke_5a.00t	rock	24	rock_mine	3	ore	24	Akivik	Intrusive	15
rdx_dyke_3a.00t	rock	25	rock_mine	4	ore	25	Lewis	Intrusive	25
rdx_dyke_3b.00t	rock	26	rock_mine	4	ore	26	Lewis	Intrusive	25
rdx_dyke_3c.00t	rock	27	rock_mine	4	ore	27	Lewis	Intrusive	25
rdx_sill_1a.00t	rock	28	rock_mine	4	ore	28	ACMA	Intrusive	20
rdx_sill_3b.00t	rock	29	rock_mine	4	ore	29	Lewis	Intrusive	20
rdx_sill_3c.00t	rock	30	rock_mine	4	ore	30	Lewis	Intrusive	20
rdx_sill_3d.00t	rock	31	rock_mine	4	ore	31	Lewis	Intrusive	20
rdx_sill_3e.00t	rock	32	rock_mine	4	ore	32	Lewis	Intrusive	20
rdx_sill_5b.00t	rock	33	rock_mine	4	ore	33	Akivik	Intrusive	20
							Torture
rdx_sill_7.00t	rock	34	rock_mine	4	ore	34	d	Intrusive	20
rdx_sill_8.00t	rock	35	rock_mine	4	ore	35	400	Intrusive	20
rdx_sill_9.00t	rock	36	rock_mine	4	ore	36	Aurora	Intrusive	20
rdxb_dyke_3a.00t	rock	37	rock_mine	5	ore	37	Lewis	Intrusive	55

Page 17-1

		Block		Block	Composite	Composite		Rock
Triangulation Name	Block Variable	Value	Block Variable	Value	Field	Value	Domain	Type	Priority
rdxb_dyke_3b.00t	rock	38	rock_mine	5	ore	38	Lewis	Intrusive	55
rdxb_dyke_3c.00t	rock	39	rock_mine	5	ore	39	Lewis	Intrusive	55
rdxb_dyke_3d.00t	rock	40	rock_mine	5	ore	40	Lewis	Intrusive	55
rdxb_sill_1a.00t	rock	41	rock_mine	5	ore	41	ACMA	Intrusive	50
rdxb_sill_3c.00t	rock	42	rock_mine	5	ore	42	Lewis	Intrusive	50
rdxb_sill_3d.00t	rock	43	rock_mine	5	ore	43	Lewis	Intrusive	50
rdxb_sill_6.00t	rock	44	rock_mine	5	ore	44	Wedge	Intrusive	50
rdxb_sill_9.00t	rock	45	rock_mine	5	ore	45	Aurora	Intrusive	50
rdxl_dyke_1a.00t	rock	46	rock_mine	6	ore	46	ACMA	Intrusive	35
rdxl_dyke_5a.00t	rock	47	rock_mine	6	ore	47	Akivik	Intrusive	35
rdxl_dyke_5b.00t	rock	48	rock_mine	6	ore	48	Akivik	Intrusive	35
rdxl_sill_1a.00t	rock	49	rock_mine	6	ore	49	ACMA	Intrusive	30
rdxl_sill_5b.00t	rock	50	rock_mine	6	ore	50	Akivik	Intrusive	30
rdxl_sill_8.00t	rock	51	rock_mine	6	ore	51	400	Intrusive	30
basal_shales.00t	rock	92	rock_mine	8	ore	92	All	Shale	1
main_shl.00t	rock	92	rock_mine	8	ore	92	All	Shale	1
mid_shl.00t	rock	92	rock_mine	8	ore	92	All	Shale	1
upper_shl.00t	rock	92	rock_mine	8	ore	92	All	Shale	1
dc5_whit_air.00t	topo	99				99	All	Air	10
ovb_topo_edited_2007								Overburde
0330.00t	topo	2				2	All	n	1
topo_surface_061107.
00t	topo	3				3	All	Topo	5

Page 17-2

	In addition to the intrusive wireframes, solids were generated to represent overburden and existing wireframes were used to constrain areas dominated by shale. Topography was also incorporated along with a variable to account for the percentage of material above and below the topographic surface.
	NovaGold validated and checked for crossing errors, consistency and closure. NovaGold found that rdx_sill_3b_dc7.00t, one of the smallest intrusive triangulations, was improperly applied to the geologic model resulting in intrusive blocks being coded as sediment. The volume represented by this triangulation is minor and, in the opinion of NovaGold, immaterial.
17.4	Capping
	Raw core and RC assays were examined for the presence of local high-grade outliers. Once these outliers were identified, the overall grade distributions were utilized to establish capping values. The raw assay data were grouped by rock type and capping values for gold were determined for each major rock type. Total sulphur, arsenic, mercury and antimony assays were not capped.
	Rock codes present in the database and major rock types used in the statistical analysis are listed in Table 17-2. A review of the codes identified several lithological codes that are relict from early drill logs and represent a very small proportion of the logged intervals. Barrick determined that several of these codes could be grouped into the major rock types for the purpose of statistical analysis, reducing the number of rock types from 34 to 12 The numerical “grouped code” value identifies all rock types assigned to a given major rock type.
	Assay top cuts were selected from cumulative frequency plots. The distribution for all gold assays (Figure 17-1) shows a well-defined lognormal population with no obvious breaks in the higher-grade trend. However, the grade-frequency trend becomes erratic above 30 g/t Au. This is therefore the position at which the raw assay would be capped and represents a total metal loss of 2.1%.
	Individual frequency distribution plots were generated to determine the appropriate grade cap for each rock type. The selected grade cap for each rock type was applied to all raw assays prior to compositing. Cumulative frequency distribution plots for each of the major rock types are provided in Figure 17-2 to Figure 17-9.
	The cumulative frequency distribution for gold assays logged as greywacke (Figure 17-2) departs from the approximate lognormal trend at grades above 25 g/t Au, corresponding to the 98.6th percentile. This grade level cap was therefore applied to a total of 94 gold assays before compositing, representing a metal loss of 4.20%.

Page 17-3

The cumulative frequency distributions for gold assays from lithological units logged as shale or argillite (Figure 17-3), rhyodacite aphanitic porphyry (Figure 17-5), rhyodacite porphyry (Figure 17-7) and rhyodacite coarse-grained blue porphyry (Figure 17-8) depart from the approximate lognormal trend at about 30 g/t Au. This grade level cap was therefore applied to a total of 14 shale and argillite, 14 rhyodacite aphanitic porphyry, 26 rhyodacite porphyry and 13 rhyodacite coarse-grained blue porphyry gold assays before compositing, representing metal losses of 6.89%, 0.43%, 1.36% and 0.84%, respectively.

Table 17-2: Rock Codes in the Drill Database

Lithology Code	Rock Type Description	Grouped Code
CHT	Chert	14
CGL	Conglomerate	14
GWK	Greywacke	14
BHF	Biotite Hornfels	15
CHF	Calcsilicate Hornfels	15
HFL	Hornfels	15
MD	Mafic Dyke	17
GRN	Ground Core/Missing Core	19
NS	No Sample	19
OVB	Overburden	20
OVR	Channel Deposits	20
ARH	Aphanitic Rhyolite	26
RDA	Rhyodacite Aphanitic Porphyry	26
RDF	Rhyodacite Fine-grained Porphyry	27
GDR	Granodiorite	28
GP	Grey Porphyry	28
GPX	Grey Crystalline Porphyry	28
GRH	Grey Rhyolite	28
IBX	Intrusive Breccia	28
MZD	Monzodiorite	28
QM	Quartz Monzonite	28
QLT	Quartz-latite Porphyry	28
QMP	Quartz-monzonite Porphyry	28
RD	Rhyodacite	28
RDX	Rhyodacite Porphyry	28
RHY	Rhyolite	28
BP	Blue Porphyry	29
BR	Blue Rhyolite	29
	Rhyodacite Coarse-grain Blue
RDXB	Porphyry	29
RDXL	Rhyodacite Lath-rich Porphyry	30
ARG	Argillite	33
SED	Sediments	33
SHL	Shale	33
SLT	Siltstone	34

Note: Major rock types used in the block model are highlighted in bold type. Assignment of minor rock types is shown in the grouped code

Page 17-4

Figure 17-1: Raw Au Assays – Cumulative Frequency Distribution of All Rock Types

Figure 17-2: Raw Au Assays – Greywacke Cumulative Frequency Distribution Plot

Page 17-5

Figure 17-3:	Raw Au Assays – Shale and Argillite Cumulative Frequency
	Distribution Plot

Figure 17-4: Raw Au Assays – Mafic Dyke Cumulative Frequency Distribution Plot

Page 17-6

Figure 17-5:	Raw Au Assays – Rhyodacite Aphanitic Porphyry Cumulative
	Frequency Distribution Plot

Page 17-7

Figure 17-6:	Raw Au Assays – Rhyodacite Fine-Grained Porphyry Cumulative
	Frequency Distribution Plot

Figure 17-7:	Raw Au Assays – Rhyodacite Porphyry Cumulative Frequency
	Distribution Plot

Page 17-8

Figure 17-8:	Raw Au Assays – Rhyodacite Coarse-Grained Blue Porphyry
	Cumulative Frequency Distribution Plot

Figure 17-9:	Raw Au Assays – Rhyodacite Lath-Rich Porphyry Cumulative
	Frequency Distribution Plot

Page 17-9

The cumulative frequency distribution for gold assays for rock types logged as mafic dyke (Figure 17-4) departs from the approximate lognormal trend and becomes erratic at grades above 35 g/t Au, corresponding to the 99.74 ^th percentile. This grade level cap was therefore applied to a total of 13 gold assays before compositing, representing a metal loss of 10.58% .

The cumulative frequency distribution for gold assays logged as rhyodacite fine-grained porphyry (Figure 17-6) departs from the approximate lognormal trend and becomes more erratic above 19 g/t Au, corresponding to the 99.27 ^th percentile. This grade level cap was therefore applied to a total of 13 gold assays before compositing, representing a metal loss of 2.71% .

The cumulative frequency distribution for gold assays logged as rhyodacite lath-rich porphyry (Figure 17-9) departs from the approximate lognormal trend and becomes more erratic above 12 g/t Au, corresponding to the 98.4 ^th percentile. This grade level cap was therefore applied to a total of 19 gold assays before compositing, representing a metal loss of 1.52% .

The effect of capping on the co-efficient of variance (COV) is summarized in Table 17-4. Values represent all assays greater than 0.1 g/t Au.

Table 17-3: Donlin Capping

Major Group	Proposed Threshold	Actual Threshold
	and Cap (g/t)	and Cap (g/t)
GWK	25	25
HFL	10	10
MD	35	35/30
No Sample / Ground Core	30	30
OVB	5	5
RDA	30	30
RDF	19	19/16
RDX	30	30/28
RDXB	30	30/28
RDXL	12	12/10
SHL	30	30
SLT	30	30/20

Page 17-10

Table 17-4: Coefficient of Variance for Au Greater than 0.1 g/t, by Rock Type

Rock Type Code	COV - Uncapped	COV - Capped
GWK	2.04	1.71
SHL/ARG	3.93	1.86
SLT	4.09	1.76
MD	2.73	0.92
OVB	2.61	1.28
RDA	1.25	1.20
RDF	1.78	1.42
RDX	1.64	1.46
RDXB	1.45	1.38
RDXL	1.09	0.99
All Rock Types	1.87	1.51

	NovaGold has reviewed the cumulative distributions, the capping script and compared the capped and uncapped assays. NovaGold has found that the capping script has errors where the threshold and cap value are not the same. The result is that there are 57 assays that should have been capped, based on the plan, that were not capped. In NovaGold’s opinion, it is unlikely that the error would result in a materially different global resource.
17.5	Specific Gravity Data and Statistics
	Specific gravity (SG) data were gathered from the 1996-1997 and the 2006 drilling program. As of March 5, 2007, the total number of SG measurements collected amounted to 7,370 points. The weighted average of these SG data points is 2.69 (g/cm3).
	Table 17-5 summarizes the average SG values by rock type.
	The SG data were further evaluated based upon the grouped rock codes used when estimating gold. The data were grouped and then divided into three main rock types: intrusive rocks, greywacke and shale. For each grouped rock type, data points clearly identified as outliers were removed before the average was determined. As is observed in Table 17-5, the SG data are quite similar within each grouped unit. Based on these statistical results, it was decided that the grouped average SG values listed in Table 17-6 would be sufficient for tonnage estimation and that a block model of SG estimates was not warranted.

Page 17-11

Table 17-5: Specific Gravity Values by Rock Type

Rock Code in			Specific Gravity
DH Database	Rock Types and Domain	No. of Samples	(g/cm3 )
ARG	Argillite	272	2.67
CGL	Conglomerate	9	2.71
FTZ	Fault Zone	25	2.75
GWK	Greywacke	2,368	2.71
MD	Mafic Dyke	473	2.73
MZD	Monzodiorite	2	2.70
RDA	Rhyodacite Aphanitic Porphyry	499	2.64
RDF	Rhyodacite Fine Grained Porphyry	315	2.67
RDX	Rhyodacite Coarse Grained Porphyry	1,339	2.66
RDXB	Rhyodacite Coarse Grained Blue Porphyry	520	2.63
RDXL	Rhyodacite Lath Rich Porphyry	216	2.64
SLT	Siltstone	838	2.72
SHL	Shale	387	2.70
All	All Rock Types	7,370	2.69

Table 17-6: Specific Gravity Values by Grouped Rock Type

			Specific Gravity
Grouped Rock Type	Individual Rock Types	No. of Samples	(g/cm3 )
Intrusive Rocks	RDA, RDX, RDXB, RDXL & RDF	2,889	2.65
Greywacke	GWK & CGL	2,377	2.71
Shale	SHL, SLT & ARG	1,497	2.70
All	All Rock Types	6,763	2.68

17.6	Assay Statistics
	Most of the potentially economic grades are hosted by the felsic dykes, with significant gold grades and grade thicknesses also hosted in shale and other sedimentary rocks, greywacke, and in rocks logged as mafic dykes.
	The logged rock code was used to cap raw assays but was not used to constrain the estimates during grade estimation.
17.7	Compositing
	Compositing is done prior to grade estimation to place the assay data on a near- constant support. Metal grades, geology and alteration data were composited “down- the-hole” into equal length, 6 m (20 ft) composites. Integer codes for domain and lithology were added to each composite by back tagging from the block model. NovaGold carried out checks on the calculation of the length-weighted grade and assignment of the integer codes. NovaGold confirmed these were correctly performed.

Page 17-12

Three composite databases were generated: one for gold values where non-assayed (missing) intervals are set to zero, one for sulphur, and one for arsenic, antimony and mercury (multi-elements). In the latter two databases, assay values are not capped and missing intervals are ignored. Principal composite fields are described in Table 17-7.

Table 17-7: Composite Database – Fields Used to Generate Estimates

Database Field	Description	Use During Estimation Process
AU_USE	Capped Gold Grade	Raw assays are capped prior to compositing. This variable is used to estimate gold.
	Composite	Composite is reported in g/t
LENG02	Gold Composite	Records the composite assay length for gold. Block estimates for gold are weighted
	Length	by length.
SU_USE	Sulphur Composite	Used to estimate total sulphur. Composite is reported in percent. Sulphur assays
		were not capped prior to compositing.
LENG03	Sulphur Composite	Records the composite assay length for total sulphur. Block estimates for sulphur are
	Length	weighted by length.
AS_USE	Arsenic Composite	Used to estimate arsenic. Composite is reported in ppm. Arsenic assays were not
		capped prior to compositing
LENG04	Arsenic Composite	Records the composite assay length for arsenic. Block estimates for arsenic are
	Length	weighted by length.
HG_USE	Mercury Composite	Used to estimate mercury. Composite is reported in ppm. Mercury assays were not
		capped prior to compositing.
LENG05	Mercury Composite	Records the composite assay length for mercury. Block estimates for mercury are
	Length	weighted by length.
SB_USE	Antimony Composite	Used to estimate antimony. Composite is reported in percent. Antimony assays
		were not capped prior to compositing.
LENG09	Antimony Composite	Records the composite assay length for antimony. Block estimates for antimony are
	Length	weighted by length.
INTIND	Intrusive Indicator	Contains integer value 0, 1 or -99. For assays with:
	Field.	INTIND =1, if Rock >=1 and <= 55 & Au >= 0.25 g/t
		INTIND =0, if Rock >=1 and <= 55 & Au < 0.25 g/t
		INTIND =-9, if Rock >=1 and <= 55 or = 1-6 & Au < 0 g/t
SEDINT	Sediment Indicator	Contains integer value 0, 1 or -99. For assays with:
	Field	SEDINT =1, if Rock >=92 and <= 93 & Au >= 0.25 g/t
		SEDINT =0, if Rock >=92 and <= 93 & Au < 0.25 g/t
		SEDINT =-9, if Rock >=92 and <= 93 or = 1-8 & Au < 0 g/t
ORE	Rock Code Field	Rock codes were assigned to the composite using the rock code in the block model
		(see Table 3.6). This field is used to restrict composites used during the estimation of
		gold and other elements.

17.8	Block Model Construction
	The block model for the Donlin Creek resource model was created using Vulcan™ mine modelling software. There was no sub-blocking used against the geological surfaces or topography and all blocks are 6 m × 6 m × 6 m in size. The coordinate system is UTM Zone 4 metres – NAD83. The model framework is provided Table 17- 8.

Page 17-13

Table 17-8: Block Model Origin and Extent

Model Start Offset (UTM metres)			Block Size (m)			Model End Offset (UTM metres)
X	Y	Z	X	Y	Z	X	Y	Z
539,000	6,877,000	-500	6	6	6	543,002	6,881,002	454

17.9	Definition of Mineralized Envelopes
	Probability shells were constructed for use in constraining gold and sulphur grade estimates in intrusives and sediments.
17.9.1	Gold
	The first stage in defining the mineralized envelopes is to divide the composites into intrusive and sediment, mineralized and non-mineralized groups using an indicator grade of 0.25 g/t gold. An indicator value1 was set to one if the composite value was equal to or greater than the chosen threshold and to zero if the composite value was less than the threshold. More formally:
	i( x, z c ) = 1 If the grade zx is threshold
	i( x, z c ) = 0 If the grade zx is < threshold
	Where zx is the composite defined in X, Y and Z space.
	The probability that gold in any block in the model exceeded the threshold was estimated using the indicator data and inverse distance to a power squared (ID2). After visual inspection, a threshold of 0.5 was selected to divide mineralized from non- mineralized domains.
17.9.2	Sulphur
	Similarly, indicator values were assigned to the sulphur 6 m composite database such that all intrusive data that were below 0.50% S were assigned an intrusive indicator value of 0 and above 0.5% S a value of 1. All sediment data with a composite sample grade below 0.50% S were assigned a sediment indicator value of 0 and above 0.5% S a value of 1.

^{______________________________________
1} The indicator value is a binary code with the value 1 or 0.

Page 17-14

17.9.3	Shell Construction
	A block discriminator model was then developed by estimation using the assigned indicator values. Indicator values were estimated in two passes for each major rock type. The inverse distance to a power squared (ID2) was utilized for block indicator assignment. The estimation and search parameters are summarized in Table 17-9 and Table 17-10.
	In the first pass, a relatively large number of samples and drill holes were used to estimate the block probabilities. At least three drill holes were required to create an indicator value for each block based on the following sample selection criteria: a minimum number of six composites per estimate, a maximum of 13 composites per estimate, and a maximum of two composites per drill hole.
	As a result, some areas in the indicator model for intrusive rocks, greywacke and shale did not receive an indicator value in the first pass. A second indicator pass was performed with search and selection criteria the same as pass one except that a minimum number of four composites was required. This change required two drill holes per estimate instead of three. A flag variable was stored in the model in order to track blocks that were assigned an indicator during this pass.
	The resulting block estimates were values between 0 and 1. For the gold indicator values, a threshold was used in intrusive rocks, shale and greywacke to separate non- mineralized material from mineralized material. These indicator values are best considered as probabilities of having a block grade above a specific threshold. An indicator block value of 0.5 equates to a 50% probability of a block having a gold grade equal to or greater than 0.25 g/t.

Table 17-9: Discriminator Model Search Parameters

Indicator Pass		Variables		Search Distance (m)			Search Rotation
		Composite	Block Model		Semi-
	Indicator	Input	Output	Major	major	Minor
Rock	Pass	Variable	Variable	Axis Z	Axis Y	Axis X	X	Y	Z
Intrusive	1	INTIND	INTIND	175	175	100	024	0	-68
Intrusive	2	INTIND	INTIND	175	175	100	024	0	-68
Shale	3	SEDINT	SEDINT	175	175	100	024	0	-68
Shale	3	SEDINT	SEDINT	175	175	100	024	0	-68
Greywacke	4	SEDINT	SEDINT	175	175	100	024	0	-68

Page 17-15

Table 17-10: Discriminator Model Sample Constraints

Indicator Pass		Variables		Sample Selection Criteria
		Composite	Block Model	Minimum	Maximum	Maximum
	Indicator	Input	Output	Samples per	Samples per	Samples per
Rock	Pass	Variable	Variable	Estimate	Estimate	Drill Hole
Intrusive	1	INTIND	INTIND	6	13	2
Intrusive	2	INTIND	INTIND	4	13	2
Shale	3	SEDINT	SEDINT	6	13	2
Shale	3	SEDINT	SEDINT	4	13	2
Greywacke	4	SEDINT	SEDINT	6	13	2
Greywacke	4	SEDINT	SEDINT	4	13	2

	The sulphur indicator model was created using the same search parameters and sample constraints as shown in Table 17-9 and Table 17-10; however, estimated indicator values were output to the INTIND_S and SEDINT_S variables in the block model.
	Composites inside a model cell that met or exceeded the probability thresholds were classified as within a mineralized envelope. NovaGold confirmed that the models were properly constructed and that the composites were properly back-coded with their respective domain.
17.10	Estimation Parameters
17.10.1	Gold Grade Estimation
	Gold grades were then estimated into the block model using an ID3 methodology for the two populations: (1) internal to the mineralized envelope, defined as blocks with indicator values greater than or equal to 50%; and (2) external to the mineralized envelope, defined as blocks with indicator values less than 50%. Composites 6 m long were flagged as being either inside the 0.25 g/t Au indicator threshold (i.e., passing through blocks with an estimated probability of at least 50%) or outside the 0.25 g/t Au indicator threshold.
	Estimation of grade into the blocks was broken into five passes based upon increasing search distances. Gold grades were estimated separately for intrusive rocks, shales and greywackes, and further sub-divided based upon whether blocks were internal or external to the mineralized envelope.
	The initial grade estimation pass used a “box search” with a search range having the same dimensions as a single block. Each successive estimation pass used increasingly longer ranges out to a maximum of 125 m (410 ft). Search ellipse and sample weights were adjusted based on an anisotropic model. Once estimated,

Page 17-16

blocks could not be overwritten by subsequent estimation passes. Table 17-11 and Table 17-12 summarize the grade interpolation parameters and constraints for the intrusive rocks. Table 17-13 and Table 17-14 document the estimation parameters and search constraints for shale and greywacke. Example bench sections through the gold grade model are shown in Figure 17-10 and Figure 17-11.

Page 17-17

Table 17-11: Au Estimation Parameters Internal to Intrusive Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Samples/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
1	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites (BIND) and	1
													blocks with intrusive
													indicator INTIND >= 0.5 &
													assigned rock values 1-55.
2	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites (BIND) and	2
													blocks with intrusive
													indicator INTIND >= 0.5 &
													assigned rock values 1-55.
3	AU_USE	AUE	ID3	75	75	30	024	0	-68	2	3	1	Composites (BIND) and	3
													blocks with intrusive
													indicator INTIND >= 0.5 &
													assigned rock values 1-55.
4	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites (BIND) and	4
													blocks with intrusive
													indicator INTIND >= 0.5 &
													assigned rock values 1-55.
5	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites (BIND) and	5
													blocks with intrusive
													indicator INTIND >= 0.5 &
													assigned rock values 1-55.

Page 17-18

Table 17-12: Au Estimation Parameters External to Intrusive Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Samples/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
6	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites and blocks with	6
													intrusive indicator INTIND <0.5
													and assigned rock values 1-55
7	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites and blocks with	7
													intrusive indicator INTIND <0.5
													& assigned rock values 1-55
8	AU_USE	AUE	ID3	75	75	35	024	0	-68	2	3	1	Composites and blocks with	8
													intrusive indicator INTIND <0.5
													& assigned rock values 1-55
9	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites and blocks with	9
													intrusive indicator INTIND <0.5
													& assigned rock values 1-55
10	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites and blocks with	10
													intrusive indicator INTIND <0.5
													& assigned rock values 1-55

Page 17-19

Table 17-13: Au Estimation Parameters Internal to the Shale Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Samples/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
1	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites and blocks with	21
													SEDINT indicator >= 0.5 &
													Rock 92
2	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites and blocks with	22
													SEDINT indicator >= 0.5 &
													Rock 92
3	AU_USE	AUE	ID3	75	75	30	024	0	-68	2	3	1	Composites and blocks with	23
													SEDINT indicator >= 0.5 &
													Rock 92
4	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites and blocks with	24
													SEDINT indicator >= 0.5 &
													Rock 92
5	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites and blocks with	25
													SEDINT indicator >= 0.5 &
													Rock 92

Page 17-20

Table 17-14: Au Estimation Parameters External to the Shale Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Samples/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
6	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites and blocks with	26
													SEDINT indicator <0.5 &
													Rock 92
7	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites and blocks with	27
													SEDINT indicator <0.5 &
													Rock 92
8	AU_USE	AUE	ID3	75	75	30	024	0	-68	2	3	1	Composites and blocks with	28
													SEDINT indicator <0.5 &
													Rock 92
9	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites and blocks with	29
													SEDINT indicator <0.5 &
													Rock 92
10	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites and blocks with	210
													SEDINT indicator <0.5 &
													Rock 92

Page 17-21

Table 17-15: Au Estimation Parameters Internal to the Greywacke Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Samples/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
1	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites and blocks with	31
													SEDINT indicator >= 0.5 &
													Rock 93
2	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites and blocks with	32
													SEDINT indicator >= 0.5 &
													Rock 93
3	AU_USE	AUE	ID3	75	75	30	024	0	-68	2	3	1	Composites and blocks with	33
													SEDINT indicator >= 0.5 &
													Rock 93
4	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites and blocks with	34
													SEDINT indicator >= 0.5 &
													Rock 93
5	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites and blocks with	35
													SEDINT indicator >= 0.5 &
													Rock 93

Page 17-22

Table 17-16: Au Estimation Parameters External to the Greywacke Indicator

	Input	Output			Semi-					Min	Max	Max/		Resultant
	Composite	Block	Est.	Major	Major	Minor	Bearing	Plunge	Dip	Sample/	Samples/	Drill		Estimate
Pass	Variable	Variable	Method	Axis	Axis	Axis	(Z)	(Y)	(X)	Estimate	Estimate	Hole	Conditions	Flag
6	AU_USE	AUE	ID2	3	3	3	024	0	-68	1	99	99	Composites and blocks with	36
													SEDINT indicator <0.5 &
													Rock 93
7	AU_USE	AUE	ID3	75	75	15	024	0	-68	2	3	1	Composites and blocks with	37
													SEDINT indicator <0.5 &
													Rock 93
8	AU_USE	AUE	ID3	75	75	30	024	0	-68	2	3	1	Composites and blocks with	38
													SEDINT indicator <0.5 &
													Rock 93
9	AU_USE	AUE	ID3	125	125	55	024	0	-68	2	3	1	Composites and blocks with	39
													SEDINT indicator <0.5 &
													Rock 93
10	AU_USE	AUE	ID3	30	30	10	024	0	-68	1	3	1	Composites and blocks with	310
													SEDINT indicator <0.5 &
													Rock 93

Page 17-23

Figure 17-10:	Bench Section Depicting Estimated Measured and Indicated Au Blocks Plotted Against
	Drill Holes – 0 masl – 50 m Section Width (source Barrick)

Page 17-24

 

Figure 17-11:	Bench Section Depicting Estimated Measured and Indicated Au Blocks Plotted Against
	Drill Holes – 100 masl – 50 m Section Width (source Barrick)

Page 17-25

17.10.2	Sulphur Grade Estimation
	Sulphur grades were estimated using the same methods and parameters as for the gold grade estimation. A series of five passes was used for blocks inside and outside the 0.50% sulphur grade indicator populations. Separate estimation runs were generated for intrusive rocks, shale and greywacke. Composites 6 m long were flagged as being either inside the 0.50% sulphur indicator threshold (i.e., blocks with an estimated probability of at least 50% for intrusive and 50% for shale and greywacke) or outside the 0.50% sulphur indicator threshold.
	Sulphur data are less extensive than gold data; therefore, a number of blocks are not estimated during the inverse distance estimation runs. Regression curves were derived from the relationship between gold and sulphur for each of the major rock types. The regression formulae were then used to assign sulphur values to un- estimated blocks based on the estimated gold grade. Where gold grade was not estimated, a value of 0.001 g/t Au was assumed for the calculation. The formulae used to assign un-estimated sulphur blocks are summarized in Table 17-17.

Table 17-17:Sulphur Regression Formulae

	Block Model Rock Code
Rock Type	Minimum	Maximum	Regression Formula
‘Other’ Intrusives	1	5	sulphur = 0.8190*Au^0.3164
RDA dykes	6	11	sulphur = 0.8848*Au^0.2925
RDA sills	12	19	sulphur = 0.7203*Au^0.2623
RDF dykes	20	24	sulphur = 1.3896*Au^0.2473
RDX dykes	25	27	sulphur = 0.8619*Au^0.3584
RDX sills	28	36	sulphur = 0.7629*Au^0.3586
RDXB dykes	37	40	sulphur = 0.6576*Au^0.3748
RDXB sills	41	45	sulphur = 0.7345*Au^0.3115
RDXL dykes	46	48	sulphur = 0.6559*Au^0.2605
RDXL sills	49	51	sulphur = 0.7668*Au^0.2884
Shale	92	92	sulphur = 1.0409*Au^0.3406
Greywacke	93	93	sulphur = 0.9227*Au^0.3973

17.10.3

Grade Estimation for Other Elements

Arsenic, mercury and antimony grades were estimated using the same methods and parameters as for the gold grade estimation. A series of five passes was used for blocks inside and outside the 0.25 g/t gold grade indicator populations. Separate estimation runs were generated for intrusive rocks, shale and greywacke. Composites 6 m long were flagged as being either inside the 0.25 g/t Au indicator threshold (i.e., blocks with an estimated probability of at least 50% for intrusive and 50% for shale and greywacke) or outside the 0.25 g/t Au indicator threshold.

Page 17-26

Data for arsenic, mercury and antimony are limited. Regression curves were derived from the relationship between gold and each of these elements for each of the major rock types. The regression formulae were then used to assign arsenic, mercury and antimony values to un-estimated blocks based on the estimated gold grade. Where gold grade was not estimated, a value of 0.001 g/t Au was assumed for the calculation. Formulae used to assign un-estimated arsenic, mercury and antimony blocks are summarized in Table 17-18 to Table 17-20.

Table 17-18:Arsenic Regression Formulae

	Block Model Rock Code
Rock Type	Minimum	Maximum	Regression Formula
‘Other’ Intrusives	1	5	arsenic = 1625.1*Au^0.4618
RDA dykes	6	11	arsenic = 1842.5*Au^0.4759
RDA sills	12	19	arsenic = 1828.5*Au^0.4541
RDF dykes	20	24	arsenic = 1453.7*Au^0.4054
RDX dykes	25	27	arsenic = 1509.7*Au^0.516
RDX sills	28	36	arsenic = 1958*Au^0.5425
RDXB dykes	37	40	arsenic = 1511.4*Au^0.5284
RDXB sills	41	45	arsenic = 1788.2*Au^0.5699
RDXL dykes	46	48	arsenic = 1650.9*Au^0.4454
RDXL sills	49	51	arsenic = 2475.8*Au^0.5344
Shale	92	92	arsenic = 1693.5*Au^0.7126
Greywacke	93	93	arsenic = 1479.2*Au^0.6026

Table 17-19:Mercury Regression Formulae

	Block Model Rock Code
Rock Type	Minimum	Maximum	Regression Formula
‘Other’ Intrusives	1	5	mercury = 0.8172*Au^0.0832
RDA dykes	6	11	mercury = 1.4318*Au^0.2591
RDA sills	12	19	mercury = 1.1658*Au^0.1852
RDF dykes	20	24	mercury = 1.8751*Au^0.2441
RDX dykes	25	27	mercury = 1.7817*Au^0.2775
RDX sills	28	36	mercury = 1.1122*Au^0.1713
RDXB dykes	37	40	mercury = 1.481*Au^0.1772
RDXB sills	41	45	mercury = 1.1273*Au^0.2093
RDXL dykes	46	48	mercury = 0.9299*Au^0.106
RDXL sills	49	51	mercury = 0.8941*Au^0.0869
Shale	92	92	mercury = 1.1091*Au^0.1709
Greywacke	93	93	mercury = 1.2638*Au^0.1912

Page 17-27

Table 17-20:Antimony Regression Formulae

	Block Model Rock Code
Rock Type	Minimum	Maximum	Regression Formula
‘Other’ Intrusives	1	5	antimony = 15.711*Au^0.2532
RDA dykes	6	11	antimony = 21.859*Au^0.2295
RDA sills	12	19	antimony = 22.391*Au^0.338
RDF dykes	20	24	antimony = 46.548*Au^0.333
RDX dykes	25	27	antimony = 24.714*Au^0.254
RDX sills	28	36	antimony = 24.961*Au^0.4095
RDXB dykes	37	40	antimony = 24.653*Au^0.275
RDXB sills	41	45	antimony = 20.141*Au^0.2973
RDXL dykes	46	48	antimony = 25.098*Au^0.3221
RDXL sills	49	51	antimony = 21.306*Au^0.3204
Shale	92	92	antimony = 22.581*Au^0.4417
Greywacke	93	93	antimony = 28.796*Au^0.364

17.10.4	CO2, Ca, and Mg Assays
	Values for carbon dioxide, calcium and magnesium were estimated separately and incorporated into the current block model. Estimates of CO2 and Ca are used for waste rock management and environmental assessment. Magnesium content is used for metallurgical models.
17.11	Classification of Waste Rock Management Categories
	Several variables were included in the block model to aid with the geochemical classification of waste rock at Donlin Creek, as described below:
	Acid Potential (AP) was calculated from the estimated total sulphur concentration (ST) where:
	Acid Potential = 31.25 x Estimated ST (%)
	Neutralization Potential (NP) from carbonate minerals was estimated from:
	NPCO3 = 0.94.NP + 0.98
	To avoid a slight bias at low NP below 16.3 kg CaCO3/t resulting from the regression equation, the calculated NPCO3 should not exceed analytical NP when NP is below 16.3 kg CaCO3/t. Therefore, the following rules were applied to the calculation:
	If NP≤16.3 kg CaCO3/t:	NPCO3 = NP
	If NP>16.3 kg CaCO3/t:	NPCO3 = 0.94.NP + 0.98

Page 17-28

The variables NPCO3 and AP were estimated for each block separately and were then used to calculate acid rock drainage (ARD) potential. ARD was modelled using the ratio NPCO₃/AP. Recommended ARD categories were assigned to each blockaccording to calculated ARD potential, as shown in Table 17-21.

Table 17-21:ARD Categories

Category	Block Model Value	Category Description	NPCO3/AP Range
A	1	Very unlikely to generate ARD	NPCO3/AP >2
B	2	Unlikely to generate ARD	1.4 < NPCO3/AP ≤2
C	3	Potentially acid generating but with very	1.0 < NPCO3/AP ≤1.4
		long delays (several decades) to onset of
		ARD
D	4	Potentially acid generating in the life of the	0.2 < NPCO3/AP ≤1.0
		mine (possibly less than a decade)
E	5	Potentially acid generating but with shorter	NPCO3/AP ≤0.2
		delays to onset (less than a few years)

The block model estimates for arsenic and sulphur values were used to calculate the ratio of arsenic to sulphur (As/S) for each block. The ARD potential and As/S ratio were then used to classify blocks into seven waste rock management categories (WRMC) subdivided into potentially acid generating (PAG) and non-PAG groups. These WRMC codes are summarized in Table 17-22. The two non-PAG categories (A and B) are each split into two sub-categories to allow for arsenic leaching. The (ii) sub-categories indicate rock that has low potential to generate ARD but could result in arsenic leaching above 1 mg/L.

Table 17-22: Recommended Waste Rock Management Categories

	Block Model			As/S
Category	Value	Category Description	NPCO3/AP Range	(As in mg/kg and S in %)
A(i)	1	Very unlikely to generate ARD, and	NPCO3/AP >2	As/S<196 and As<250
		“low” arsenic leaching.
A(ii)	2	Very unlikely to generate ARD, and	NPCO3/AP >2	As/S>196 or As>250
		arsenic leaching potentially significant.
B(i)	3	Unlikely to generate ARD and “low”	1.4 < NPCO3/AP ≤2	As/S<196 and As<250
		arsenic leaching”.
B(ii)	4	Unlikely to generate ARD and arsenic	1.4 < NPCO3/AP ≤2	As/S>196 or As>250
		leaching potentially significant.
C	5	PAG but with very long delays (several	1.0 < NPCO3/AP ≤1.4	All
		decades) to onset of ARD.
D	6	PAG in the life of the mine (possibly	0.2 < NPCO3/AP ≤1.0	All
		less than a decade).
E	7	PAG but with shorter delays to onset	NPCO3/AP ≤0.2	All
		(less than a few years).

Page 17-29

17.12	Variography
	The 6 m composites were used to develop indicator and relative pair-wise variograms. Relative pair-wise variograms were generated for all sample data and by domain using orientations along the average strike and dip of the mineralized zones. This orientation was identified both geologically and through stereonet analysis of oriented vein data. The analysis defines a plane striking 024° and dipping 68° to the southeast and forms the basis for search orientation during block estimation.
	Indicator variograms were generated at 0.25 g/t Au for the 6 m composites. The correlograms at 0.25 g/t Au were fitted with a spherical model as shown in Figure 17- 12. Ranges of 30 m (98 ft) and 45 m (148 ft) can be observed at 80% and 90% of the total sill variance. Barrick used the variogram solely for classification of resources, described in Section 17.12.

Figure 17-12: 0.25 g/t Au Indicator Variogram for 6 m Composites

17.13	Model Validation
	To validate the gold and sulphur grade estimates a number of checks were carried out. These include:
		Visual inspection of gold estimation results on plans and sections

Page 17-30

	Comparison of estimated and nearest neighbour statistics
	A change of support check
	A check of “script files” used within the Vulcan™ software system
	A check of summation of the mineral resources
Validation by Visual Inspection on Sections and Plans
Estimated grades, resource classification, drill hole assays and mineralized shells were inspected on-screen by NovaGold. Overall, NovaGold found a reasonable agreement between the assays and the estimated gold and sulphur grades. The inferred estimate appears to be well constrained and no significant areas of overestimation were observed.
Comparison to a Nearest Neighbour Estimate
For the purposes of validation, nearest neighbour (NN) models with the 6 m gold and sulphur composites were generated.
Table 17-23 and Table 17-24 summarize the comparisons between nearest neighbour and estimated gold and sulphur grades. The results are tabulated at a zero cutoff. The differences shown are well within 5%, which NovaGold considers the accepted limit with the exception of the gold estimate of rock group “rck01_other”. This exception represents less than 0.1% of the global tonnage.
The comparison with the nearest neighbour statistics is considered acceptable for the grade estimation model.
Swath Plots
The term swath plot refers to an assessment of local trends. Groups of estimates and data are compared for different sections or plans, i.e., swaths of data, and these are compared to ensure that the estimation methodology honours the trends in the data.
For this review, the program tm_1davg.exe was used to determine the trends in the ID3 and NN models and the principle directions of easting, northing and elevation were examined by plotting graphs of grade versus distance.
Figure 17-13 shows an example of the trend check results for rck02_rda intrusive. Note that both the red and green lines display similar trends. A complete set of the trend plots is provided in Appendix D.

Page 17-31

Based on this assessment, the DC7a model appears to honour the local trends in the data.

Change of Support Check

Change of support issue becomes important when a cut-off is applied to a distribution. The validation of a model is to determine if it is conditionally unbiased and properly reflect tonnes and grade above a cut-off, or range of cut-offs. The range of economic cut-offs considered range between 0.7 and 1.5 g/t gold.

One measure of the change of support in the model is the coefficient of variation (CV). NovaGold calculated the theoretical CV for 6 m x 6 m x 6 m blocks and compared to the block model distribution using a Discrete Gaussian Method (DGM), which adjusts the variance of the point information (NN estimate) to represent the variance of the proposed 6 m by 6 m by 6 m SMU. The calculation of the expected CV can be made taking into account “perfect selection”. In perfect selection, the grade of the block is known “perfectly” and the discriminating decision of ore or waste is made without error.

Page 17-32

Table 17-23: Comparison of Estimated Gold Grade (au_use) to Nearest Neighbour (pau) Model

		Au_use (g/t)				Pau (g/t)					Au_use/Pau	Au_use/Pau
Rock_mine	no blks	mean	std dev	cv	max	no blks	mean	std dev	cv	max	mean	CV
rck01_other	4,409	1.250	1.229	0.983	9.010	4,409	1.159	1.419	1.224	11.420	1.079	0.803
rck02_rda	433,947	1.252	1.684	1.345	20.520	433,947	1.257	1.985	1.580	26.620	0.996	0.852
rck03_rdf	11,223	1.330	1.331	1.001	8.590	11,223	1.318	1.644	1.248	10.230	1.009	0.802
rck04_rdx	594,596	1.179	1.678	1.423	27.950	594,596	1.167	1.971	1.689	28.850	1.010	0.843
rck05_rdxb	369,325	0.979	1.561	1.595	28.460	369,325	0.990	1.894	1.913	28.460	0.988	0.834
rck06_rdxl	137,641	1.369	1.351	0.987	17.450	137,641	1.365	1.553	1.138	24.670	1.003	0.867
rck07_shale	3,728,838	0.220	0.882	4.000	22.660	3,728,838	0.216	0.978	4.534	22.660	1.022	0.882
rck08_gwk	1,085,041	0.176	0.829	4.701	24.910	1,085,041	0.180	0.928	5.168	24.920	0.982	0.910
All	6,365,020	0.436	1.180	2.705	28.460	6,365,020	0.436	1.333	3.062	28.850	1.001	0.884

Table 17-24: Comparison of Estimated Sulphur Grade (su_use) to Nearest Neighbour (psu_use) Model

	Su_use % (S)					Psu_use % (NN)					S/NN	S/NN
Rock_mine	no blks	mean	std dev	cv	max	no blks	mean	std dev	cv	max	mean	CV
rck01_other	4,408	0.939	0.408	0.434	2.656	4,408	0.921	0.469	0.509	2.953	1.020	0.853
rck02_rda	415,741	0.838	0.467	0.558	6.072	415,741	0.843	0.560	0.664	6.072	0.994	0.840
rck03_rdf	11,108	1.198	0.431	0.360	2.914	11,108	1.152	0.559	0.485	4.473	1.040	0.742
rck04_rdx	575,595	0.849	0.539	0.635	8.519	575,595	0.860	0.601	0.699	8.519	0.986	0.909
rck05_rdxb	336,631	0.753	0.485	0.645	5.680	336,631	0.757	0.547	0.722	5.680	0.993	0.893
rck06_rdxl	127,460	0.816	0.365	0.448	5.130	127,460	0.814	0.446	0.549	5.130	1.002	0.816
rck07_shale	3,038,497	0.452	0.557	1.231	5.886	3,038,497	0.450	0.601	1.336	5.886	1.004	0.922
rck08_gwk	982,375	0.420	0.508	1.209	5.997	982,375	0.422	0.556	1.320	6.012	0.997	0.916
All	5,491,815	0.544	0.556	1.021	8.519	5,491,815	0.544	0.608	1.117	8.519	0.999	0.915

Page 17-33

Figure 17-13: Trend Plot of RCK02_RDA Gold Grades ID3 (au_use) versus NN (pau)

Page 17-34

To determine the spatial continuity, correlograms were computed for each rock type that has sufficient 6 m composites. Rock types rck01_other and rck03_rdf only have 60 and 69 6m composites, respectively, so correlograms were not calculated. The correlograms were computed in the GSLIB program gamv.exe and fit using vmodel.exe. An example correlogram provided in Figure 17-14 shows the fitted model developed for rck02_rda.

Figure 17-14: Correlogram of 6 m Composite Gold Grade for RCK02_RDA Intrusive

The fitted model is then used to calculate the variance of a point within a block using the program gammabar.exe. Using Krige’s relationship, the variance of a block within the deposit can be determined along with the block variance. The block variance is then used as a parameter for the DGM program dgm.exe.

The ID3 grade estimates have been placed into a block model with a grid of 6 m by 6 m by 6 m and then an economic cut-off is applied to the model. In theory then, the SMU-sized ID3 block estimates should have the same distribution as the DGM-NN. If this were true, then the CV ratios should be close to 1.0. None of these ratios are close to 1.0; in fact, they are greater than 1.0, indicating that the ID3 gold grade estimation model is more variable than it should be for a model reflecting the appropriate change of support.

Page 17-35

The impact of having a grade estimation model that is too variable is that for any given cut-off, the tonnes and grade will not reflect the tonnes and grade of the SMU blocks and will likely overestimate the grade and underestimate the tonnes. A block model that is too variable is therefore conditionally biased.

Figure 17-15 displays grade-tonnage curves for NN (black), ID3 (green) and the DGM-NN (red). The black line for tonnes and grade is equivalent to separating ore from waste at 6 m drill core SMU, not the block SMU. The DGM-NN lines show the distribution of the SMU blocks as determined from the change of support method. The ID3 green line is directly from the ID3 grade estimates.

Figure 17-15: Grade – Tonnage Curve for Gold Estimate for RCK02_RDA Intrusive

Using the 1.2 g/t Au cut-off, the ID3 model predicts a proportional tonnage of 0.375 grading 2.83 g/t Au, while the DGM-NN predicts a tonnage of 0.375 with a grade of 2.60 g/t Au. While the tonnage is the same at this cut-off, the grade is about 7% more for the ID3 model than the DGM-NN. This is entirely due to the difference in the variability of the ID3 model.

Page 17-36

Grade-tonnage curves for the other rock types show a similar behaviour to that of the grade-tonnage curve of rck02_rda (see Appendix E). For all rock types, the ID3 model overestimates the grade and underestimates the tonnes at the range of economic cutoffs.

It is possible to adjust the variance of the ID3 grade estimates to better match those predicted by DGM-NN model. The blue lines shown in Figure 17-15 are variance-corrected ID3 grade estimates for rck02_rda. As the amount of the adjustment is small (for the intrusive), the affine correction method was used.

In NovaGold’s opinion, the gold and sulphur grade estimates are under-smoothed relative to the expected variability of perfect selection. It is possible that a mine plan generated from this block model will overestimate the gold grade. The Donlin Creek LLC is aware of the issue and is working internally and with outside experts to quantify the materiality and assess the impact of alternate estimation plans that utilize more composite samples. NovaGold agrees with this approach and suggests that mine planning and production scheduling regard this model as being an optimistic case.

A Review of Vulcan “Script Files”

NovaGold inspected the “script-files” created for each gold estimation run to ensure that the correct search, composite and block selection, and parameters were used to estimate grades in the block model. No errors were found in the run-files.

A Summation Check

NovaGold independently checked the summations for the grade and tonnage figures. No errors were found.

Page 17-37

17.14	Resource Classification and Summary
	The logic for mineral resource classification of Donlin Creek is consistent with the CIM definitions referred to in NI 43-101. The current level of drilling allows a reasonable assumption of geologic and grade continuity for the indicated mineral resource category. The Indicated mineral resource category is supported by the present drilling grids over the ACMA and Lewis deposits (nominal 25 m to 35 m). The measured mineral resource category is supported only in blocks pierced by exploration drill holes. Inferred mineralization is limited to a reasonable expectation of mining by a conceptual open pit shell using a metal price of US$650 per ounce of gold and recent estimates of mining, geotechnical and metallurgical parameters.
17.14.1	Resource Classification
	The resource model was classified using distance to nearest composite as stored in the model blocks during the nearest-neighbour grade estimate. Classification distances are based on the 80% and 90% of variance from the omni-directional indicator variogram model described in Section 17.12. The classification methodology is provided in Table 17-25.

Table 17-25: Donlin Creek Resource Classification

	Minimum	Maximum
	Distance to	Distance to	Minimum	Intrusive Indicator
	Nearest Drill	Nearest Drill	Number of Drill	Block Condition	Sediment & Greywacke
Category	Hole	Hole	Holes	Criteria	Indicator Block Criteria
Measured	0 m	3 m	Block pierced by	>= 0.0	>= 0.0
			drill hole
Indicated	0 m	30 m	>= 2	>= 0.0	>= 0.0
Indicated	30 m	45 m	>= 2	>= 0.5	>= 0.7
Inferred	30 m	45 m	>= 2	>= 0.0 & <0.5	>= 0.0 & <0.7
Inferred	45 m	60 m	>= 2	>= 0.5	>= 0.7

17.14.2

Resource Tabulation

The mineralization of the Donlin Creek project, effective date as of February 5, 2008, is classified as measured, indicated and inferred mineral resources. The classified mineral resources are shown in Table 17-26. The mineral resource has been constrained within a conceptual pit based on US$650 per ounce of gold and using recent estimates of mining, geotechnical and metallurgical parameters. Barrick’s Technical Services Evaluations Group estimated a variable net smelter return (NSR) cut-off grade based on recent estimates of mining costs, processing costs (dependent upon sulphur content), selling costs and royalties, rather than gold grade alone. The

Page 17-38

NSR cut-off equates to approximately 0.8 g/t gold at the average estimated sulphur grade of 1.12% .

The mineral resource estimates for Donlin Creek project show an increase in resources over the July 2006 mineral resource estimates. This increase is the result of additional drilling.

Table 17-26:	Donlin Creek Project Mineral Resource Summary(1)(2)(3)(4)
	Effective Date February 5, 2008

	Tonnes	Au	Contained Au
	(M)	(g/t)	(Million oz)
US$0.01 NSR/t Cut-off
Measured Mineral Resource	4.3	2.73	0.38
Indicated Mineral Resource	367.4	2.46	29.0
Measured + Indicated Mineral Resources	371.7	2.46	29.38
Inferred Mineral Resource	46.5	2.31	3.46

⁽¹⁾ Mineral resources that are not mineral reserves do not have demonstrated economic viability
⁽²⁾ Rounding differences may occur.
⁽³⁾ Resources are constrained within a Lerchs-Grossman (LG) open-pit shell using the long-term metal price assumption of US$650/oz of gold. Assumptions for the LG shell included pit slopes variable by sector and pit area: mining cost is variable with depth, averaging US$1.57/t mined; process cost is calculated as the percent sulphur grade x US$2.09 + US$10.91; general and administrative costs, gold selling cost and sustaining capital cost are reflected on a per tonne basis. Average estimated sulphur grade is 1.12% . Based on metallurgical testing, gold recovery is assumed to be 89.5% . Blocks with a cost margin of US$0.01/t or higher above the variable cut-off were reported.

NovaGold reviewed in detail the implementation of the classification criteria, and subsequent flagging of blocks. NovaGold also reviewed the tabulation of the resource blocks within the various domains using an independent computer program. No errors or omissions were found. In the opinion of the author, resource classification criteria are reasonable. NovaGold is not aware of any environmental, permitting, legal, title, taxation, socio-economic, marketing, political or other relevant issue that would materially impact the estimation of resources. NovaGold is also not aware of any mining, metallurgical, infrastructure or other relevant factors that would materially impact the estimation of resources.

Page 17-39

18.0	OTHER RELEVANT DATA AND INFORMATION
	There are no other relevant data and information available at the time of this report.

Page 18-1

19.0	ADDITIONAL REQUIREMENTS FOR TECHNICAL REPORTS ON DEVELOPMENT PROPERTIES AND PRODUCTION PROPERTIES
	A Preliminary Economic Assessment by AMEC was completed in March 2002 (Juras and Hodgson, 2002). NovaGold commissioned SRK to complete a Preliminary Economic Assessment in September 2006 (Dodd et al., 2006). Both of these studies are not reliable due to changes in mineral resources. No other studies have been finalized for the Donlin Creek project.
	The project is not presently a development or production property.

Page 19-1

20.0	INTERPRETATION AND CONCLUSIONS
		The geology of the Donlin Creek project and mineralization controls are well understood.
		The mineralization has been delineated by core and RC samples collected from drill holes. Key quality assurance and control steps have ensured the validity of the assay database. Auditing of the drill hole database has verified the integrity of the data used to estimate resources.
		Metallurgical testing completed to date appears to support the selected processing sequence.
		Probability shells were constructed to partition intrusive and sediments into mineralized and unmineralized domains during gold and sulphur estimation.
		Gold and sulphur grades were estimated by ID3. NovaGold has validated the resource estimate and believes it to have been completed to industry standard.
		The mineral resource at Donlin Creek is classified according to the CIM definitions referred to in National Instrument 43-101.
		In the opinion of the author the underlying data are adequate to define the mineral resources that are the subject of this report.

Page 20-1

21.0	RECOMMENDATIONS
	The project is of sufficient quality to warrant a feasibility study. The author is not aware of the exact cost of such a study, but assumes it could exceed two million dollars.

Page 21-1

22.0	REFERENCES
	AMEC, 2003, Prefeasibility Study – Status Report, Donlin Creek Project, Draft, Internal report for Placer Dome
	Juras, S., 2002: Technical Report, Donlin Creek Project, Alaska, unpublished Technical Report to NovaGold Resources, Inc. by AMEC E&C Services Limited, effective date January 24, 2002.
	Juras, S. and Hodgson S., 2002, Technical Report, Preliminary Assessment, Donlin Creek Project, Alaska, unpublished technical report to NovaGold Resources, Inc. by AMEC E&C Services Limited, effective date January 24, 2002.
	Miller, L., Ebert, S., Kowalczyk, P., Petsel, S., McAtee, J., Goldfarb, R., Miller, M.L. and Dodd, S., 2000, Geology, mineralization, and exploration at the Donlin Creek project, southwestern Alaska [abs.]: British Columbia and Yukon Chamber of Mines Cordilleran Roundup, Abstracts, p. 45.
	Piekenbrock, J.R. and Petsel, S.A., 2003, Geology and Interpretation of the Donlin Creek Gold Deposit, Alaska, Private Report to NovaGold Resources
22.1	Glossary
	Adit: An opening driven horizontally into the side of a mountain or hill for providing access to a mineral deposit
	Aeromagnetic survey: A geophysical survey using a magnetometer on board or towed behind an aircraft.
	Airborne survey: A survey made from an aircraft to obtain photographs, or measure magnetic properties, radioactivity, and so on.
	Anomaly: Any departure from the norm which may indicate the presence of mineralization in underlying bedrock.
	Anticline: an area of rock deformation that involves a downward slope to either side. In an exposed eroded anticline the oldest rock layers are in the center and the rocks on either side dip or slope away from the center of the structure. The rock at the center of the anticline is higher than the same stratum elsewhere. Anticlines typically form during crustal warping as the result of compression concurrent with orogenic mountain building.
	Assay: A chemical test performed on a sample of ores or minerals to determine the amount of contained metal.

Page 22-1

Base metal: A base metal is a common or at least inexpensive metal. Frequently, the term is used to refer to those that oxidize or corrode relatively easily, and react variably with dilute hydrochloric acid to form hydrogen. Examples include iron, nickel, copper, lead and zinc.

Bedding: The arrangement of sedimentary rocks in layers.

Breccia: A rock in which angular fragments are surrounded by a mass of fine-grained minerals.

Channel sample: A sample composed of pieces of vein or mineral deposit that have been cut out of a small trench or channel, usually about 10 cm wide and 2 cm deep

Cordillera: The continuous chain of mountain ranges on the western margins of North and South America.

Diamond drill: A rotary type of rock drill that cuts a core of rock that is recovered in long cylindrical sections, two cm or more in diameter.

Electromagnetic Survey: A geophysical survey method which measures the electromagnetic properties of rocks.

Footwall: The rock on the underside of a vein or ore structure.

Fold: geological process that causes a bend in a stratum of rock

Geochemical Survey: a study or sampling program undertaken to measure the quantities of selected elements in a certain area

Grab sample: A sample from a rock outcrop that is assayed to determine if valuable elements are contained in the rock. A grab sample is not intended to be representative of the deposit, and usually the best-looking material is selected.

Grade: Percentage of a metal or mineral composition in an ore or processing product from mineral processing.

Hangingwall: The rock on the upper side of a vein or ore deposit.

Host Rock: The rock within which the mineralization or ore occurs

Induced polarization: A method of ground geophysical surveying employing an electrical current to determine indications of mineralization.

Page 22-2

Partnership: a business undertaking entered into by two or more parties which is intended to terminate upon the completion of a specific project

Landsat: The generic name for a series of natural resource scanning satellites launched by the United States beginning in 1972.

Landsat TM: Landsat Thematic Mapper. Earth observation satellite with seven bands at 30 m spatial resolution

Mapping: Recording geological features on a map

Mine: An opening or excavation in the earth for the purpose of extracting minerals.

Mineral: A naturally occurring, solid, inorganic element or compound, with a definite composition or range of compositions, usually possessing a regular internal crystalline structure.

Mineral Resource: A concentration or occurrence of natural, solid, inorganic or fossilized organic material in or on the earth’s crust in such form and quantity and of such a grade or quality that it has reasonable prospects for economic extraction. The location, quantity, grade, geological characteristics and continuity of a mineral resource (deposit) are known, estimated or interpreted from specific geological evidence and knowledge.

Mineral Reserve: A mineral reserve is the economically mineable part of a measured or indicated mineral resource demonstrated by at least a preliminary feasibility study. This study must include adequate information on mining, processing, metallurgical, economic and other relevant factors that demonstrate, at the time of reporting, that economic extraction can be justified. A mineral reserve includes diluting materials and allowances for losses that may occur when the material is mined.

Occurrence: The area where a particular mineral is found

Ore: A natural deposit in which a valuable metallic element occurs in high enough concentration to make mining economically feasible. The term is proscribed under NI 43-101.

Overburden: Material of any nature, consolidated or unconsolidated, that overlies a deposit of ore that is to be mined.

pH: The negative logarithm of the hydrogen ion concentration, in which pH = -log [H+]. Neutral solutions have pH values of 7, acidic solutions have pH values less than 7, and alkaline solutions have pH values greater than 7.

Page 22-3

Prospect: A mining property, the value of which has not been determined by exploration

Reconnaissance: a general examination or survey of a region with reference to its main features, usually preliminary to a more detailed survey

Resistivity survey: A geophysical technique used to measure the resistance of a rock formation to an electric current.

Rift: a zone between two diverging tectonic plates

Royalty: a percentage interest in the value of production from a lease that is retained and paid to the mineral rights owner

Scarp: A steep cliff or steep slope, formed either as a result of faulting or by the erosion of inclined rock strata

Schist: Metamorphic rock dominated by fibrous or platy minerals. Schist has a schistose plain of cleavage, and is product of regional metamorphism

Sedimentary Rock: A rock formed from the consolidation of loose sediment or from chemical precipitation, such as sandstone and limestone

Shear Zone: A planar zone of weakness, similar to a fault, but consisting of several parallel displacement zones usually over a greater width than a single fault

Tectonic: pertaining to the rock structures and external forms resulting from the deformation of the Earth's crust.

Trench: long, narrow excavation dug through overburden or blasted out of rock to expose a vein or ore structure

Vein: A mineralized zone having a more or less regular development in length, width, and depth to give it a tabular form.

Zone: An area of distinct mineralization.

22.2 Abbreviations and Units of Measure

Acid mine drainage	AMD
Annum (year)	a
Average	AV
Best value	BV
Canadian Securities Administration	CSA
Centimetre	cm

Page 22-4

Check Samples	CS
Coefficient of determination	R2
Confidence interval	CI
Copper	Cu
Cubic centimetre	cm3
Cubic feet per minute	cfm
Cubic metre	m3
Copper Equivalent	CuEQ
Day	d
Days per week	d/wk
Days per year (annum)	d/a
Degree	°
Diameter	Ø
Dry metric tonne	dmt
Elevation (metres)	el
Environmental Evaluation	EA
Environmental Impact Assessment	EIA
Gram	g
Global positioning system	GPS
Gold	Au
Grams per tonne	g/t
Greater than	>
Hectare (10,000 m2 )	ha
Hour	h
Hours per day	h/d
Hours per week	h/wk
Hours per year	h/a
Inductively-coupled plasma (Chemical Analysis Instrument)	ICP
Kilogram	kg
Kilograms per cubic metre	kg/m3
Kilograms per hour	kg/h
Kilograms per square metre	kg/m2
Kilogram per year	kg/a
Kilometre	km
Lead	Pb
Less than	<
Litre	L
Litres per minute	L/m
Mass spectrometer (Analysis Instrument)	MS
Mass submerged in water	Mw
Measure of acidity or alkalinity of a solution	pH
Metre	m
Metres above sea level	masl
Metres per minute	m/min
Metres per second	m/s
Micrometre (micron) 10 -6 m	µm
Milliamperes	mA
Milligram	mg
Milligrams per litre	mg/L
Millilitre	mL
Millimetre	mm

Page 22-5

Million	M
Million Dollars (US)	US$M
Million tonnes	Mt
Ministry of Energy and Mines	MEM
Minute (plane angle)	'
Minute (time)	min
Month	mo
Natural Source Audio Magnetotelluric	NSAMT
Ounce	oz
Optical Emission Spectroscopy (Analysis Instrument)	OES
Overall bias	OABias
Parts per billion	ppb
Parts per million	ppm
Percent	%
Preliminary Economic Assessment	PEA
Probability Assisted Constrained Kriging	PACK
Quality Assurance and Quality Control	QA/QC
Reverse circulation	RC
Rock mass rating	RMR
Rock quality designator	RQD
Second (plane angle)	"
Second (time)	s
Silver	Ag
Standard deviation	SD
Specific gravity	SG
Square centimetre	cm2
Square kilometre	km2
Square metre	m2
Thermal Imaging	TM
Thousand tonnes	kt
Tonne (1,000 kg)	t
Tonnes (1,000 kg) per annum	t/a
Tonnes (1,000 kg) per day	t/d
Underground	UG
US dollar	US$
Universal Transverse Mercator (co-ordinate system)	UTM
X-Ray diffraction	XRD
Year (annum)	a

Page 22-6