PRELIMINARY ECONOMIC STUDY

ROCK CREEK PROJECT
NOME, ALASKA

Submitted to:
NOVAGOLD RESOURCES INC.

August 13, 2003
Amended September 22, 2003

Prepared by:

Ken Kuchling, P.Eng. - Qualified Person:
Norwest Corporation
Suite 400, 205 – 9^th Ave SE
Calgary, Alberta T2G 0R3
Tel:      (403) 237-7763
Fax:      (403) 263-4086
Email    calgary@norwestcorp.com

www.norwestcorp.com

NovaGold Resources Inc. (NovaGold) commissioned Norwest Corporation (Norwest) to complete a preliminary economic study, mine plan, and cost estimate for the Rock Creek Project. In addition a summary of key environmental issues and permitting requirements has also been compiled. All amounts in this study are in US dollars unless otherwise stated.

Various sources of information have been used to develop the preliminary mine plan and cost estimate. NovaGold provided the geological data base, interpretation and the geological block model for Rock Creek. Information related to milling and ore recoveries are based on independent test work previously supervised by NovaGold. A single gold recovery test was completed by Norwest to verify and build on the previous test work. Capital and operating costs and equipment productivities are based on data from Norwest's in-house database and costing information from published sources.

This Preliminary Assessment includes the use of inferred resources that are considered too speculative geologically to have economic considerations applied to them that would enable them to be categorized as mineral reserves. Inferred gold resources will require further exploration to upgrade them to the higher Measured and Indicated categories.

The Rock Creek Project is located on the Seward Peninsula along the west coast of Alaska, north of Norton Sound. The project area lies about 13 km north of Nome and is accessed via state maintained roads. The city of Nome (population 4,000) is situated on the Bering Sea coast and serves as the logistical and administrative center for this portion of western Alaska. Nome has daily commercial jet service from Anchorage and has large container barge service from June through October. The project occurs partly on patented mining claims owned 100% by the Alaska Gold Co., a wholly owned subsidiary of NovaGold, and partly on land owned by Bering Straits Native Corporation (BSNC) subsurface, and Sitnausuak (the Nome native village corporation) surface, under agreement with NovaGold. The project site is characterized by cool summers and cold winters. Summer temperatures range from +8⁰C to +15�‹ C and winter temperatures average around -15°C.

Mineralization at Rock Creek is confined to quartz veining in two ore types; tension veins and the Albion shear zone. The resource in the tension veins accounts for approximately 90% of the tonnes and 85% of the gold content in the deposit. The two ore types each have different process recovery criteria, for reasons that may be related to sulfo-salt and silicification. Previous studies indicate that the Albion vein quartz poses the greater challenge in gravity recovery circuits, with recovery values in the range of 30%-40% for coarse grind material. The tension vein material yields higher gravity recoveries in the range of >80%. Cyanidation of the whole rock yields

fairly high recoveries for both ore types. Finer grinding tends to liberate gold and helps to improve recovery. For design purposes, it is assumed that gold recoveries for the two ore types will be 94% for the Albion zone and 97% for the tension vein material, based on a process relying on a fairly fine grind (80% < 140 microns).

The site specific geotechnical data gathered to date is limited to core descriptions, rock quality (RQD) and hardness information from 2002 and 2003 drilling. It is recommended that an inter-ramp pit wall angle of 45 degrees be adopted in scoping pit design. Although there is no geotechnical information available on the proposed plant site, foundations for buildings and the various items of plant are not expected to require any special methods of construction other than those dictated by permafrost conditions. As the project proceeds through the various levels of design, site specific information will need to be gathered to support the basic mine, development rock and tailings facility designs, as well as the design of building and plant foundations. To the best of our knowledge no thermistors or permafrost measurements have been made at the Rock Creek site although published data indicates that permafrost depths in Nome could be approximately 60 to 100 metres.

The proposed mine development and operation would have to obtain numerous State and Federal permits before development and operation can proceed. Various Federal and State agencies have jurisdiction over many of the activities proposed for the Rock Creek Project. The types of permits and approvals, which may be needed in order to construct and operate the Rock Creek Project, will depend on the design and operation of the mine. It is assumed that all project activities will occur on private or state lands and that no federal land use or rights-of-way will be required. Environmental baseline studies are required in support of the preparation of a number of State and Federal permit applications. Baseline information and data regarding surface water, groundwater, tailings and development rock geochemistry soils, vegetation, fish and wildlife, and threatened/endangered species, are necessary for design of the mining (i.e., extraction) and reclamation plan that must be prepared in support of the Plan of Operations application and the Solid Waste permit application. It is estimated that baseline collection will take about a year and then permitting will take an additional year.

Prior to developing the pit designs and mine layouts, the Lerchs-Grossmann (LG) pit optimizer was used to evaluate pit geometries based on variable economic factors. Each variation in revenue or cost would result in a different optimized pit. The Rock Creek pit has been design based on a gold price of US$325/oz and a mining cut-off grade of 1.0 g/t.

Production scheduling and mining equipment requirements are based on an assumed project operating schedule of two12-hour shifts per day for 360 days per year. Based on the waste to ore ratio, it is expected that the mining operations will either focus on ore or waste mining operations on a shift-by-shift basis. Probably four out of every five shifts would be assigned to waste stripping operations. During the periods when the fleet is stripping waste, the mill will continue to operate by reclaiming ore from the live ore stockpile. Ore mining rates are about 1.78 million tonnes per year while waste volumes range from 6.8 to 9.2 million tonnes per year.

The two main waste products to be generated by the Rock Creek mining operation will be the development rock stripped from the mine and tailings generated by the milling process. Each of these products will be disposed of separately. Development rock stripped from the pit will be placed into development rock dumps located adjacent to the pit. If pit sequencing permits, some development rock may be backfilled into mined out portions of the pit. Some development rock will also be used periodically for raising the tailings dam. A single tailings facility will be used to store the mill tailings, provide seasonal water inventory for processing needs and act as a storm water runoff buffer. The tailings will be retained behind a centerline constructed containment dam, which will be constructed using mined development rock and will incorporate a graded filter zone to mitigate piping issues.

The major mining equipment fleet for the Rock Creek Project will be modest in size. Two 12 m³ front end loaders, equivalent to a CAT 992G loader, will provide sufficient capacity for the ore and waste loading operations. A truck fleet of four 100 tonne trucks will provide operating capacity. Other mining equipment will consist of a blasthole drill, dozers, graders, and various pickup trucks and service vehicles.

A screening of nine different milling options was undertaken, including gravity, flotation, and cyanidation processes. The preferred case consists of a gravity circuit supplemented with a flotation circuit. One benefit of flotation is that it may be possible to achieve good recoveries at a relatively coarser grind than required for gravity alone. The gravity circuit would produce a relatively clean gold stream that would be refined into gold dore bars at site. The flotation circuit would produce a concentrate that would be shipped off-site for further processing, such as cyanidation. Approximately 82% of the recovered gold would report to the dore circuit. Further test work is required to support these assumptions.

The plant site area includes space for the primary crusher and crushed ore stockpile, the mill, a maintenance shop, an administration and mine dry building, and fuel storage. Ammonium nitrate and explosives will be stored remotely away from the main plant area. It is estimated that a total power requirement of 4 to 5 MW will be needed and will be supplied from the Nome Joint Utility System or on-site generation. The estimated cost for electric power from the local utility is US$0.12/kwh however further discussion with the power utility is required to confirm the responsibility for installing the step-up transformer and upgrading the power line to 25 kv from 125 kv.

The total project manpower is estimated at about 123 personnel, as summarized in the table below.

The project economics are most sensitive to the gold price and the grade of the ore body. Each 10% change affects the NPV by about 77% and the IRR by approximately 51% in the Base Case.

Operating cost changes of 10% will impact the NPV by 45% and the IRR by 31%. The capital cost affects the NPV by about 22% and the IRR by 20% for each 10% increase or decrease in the capital cost in the Base Case.

NovaGold geological staff indicate that there is a reasonable expectation that the in-pit resource and block model gold quantity at Rock Creek will increase due to additional resource drilling within the main orebody, drilling along untested extensions of the ore zone, and improved

TERMS OF REFERENCE

NovaGold Resources Inc. (NovaGold) commissioned Norwest Corporation (Norwest) to complete a preliminary economic study, mine plan, and cost estimate for the Rock Creek Project. In addition a summary of key environmental issues and permitting requirements has also been compiled.

SOURCES OF INFORMATION

Various sources of information have been used to develop the preliminary mine plan and cost estimate.

NovaGold provided the geological data base, interpretation and the geological block model for Rock Creek. Norwest validated the model based on data provided and assumptions used. There was no testing of assumptions or verification of the database.

Information related to milling and ore recoveries are based on independent test work previously supervised by NovaGold. A single gold recovery test was completed by Norwest to verify and build on the previous test work.

Operating and capital costs and equipment productivities are based on data from Norwest's in-house database and costing information from published sources.

DISCLAIMER

This report has been prepared for NovaGold Resources Inc. (NovaGold) by Norwest Corporation (Norwest). The findings, conclusions, and cost estimates provided are based on information developed by Norwest available at the time of preparation and data supplied by outside sources, including NovaGold staff and consultants working for NovaGold.

This preliminary economic study has been prepared under the supervision of Ken Kuchling, P.Eng. the designated Qualified Person. He completed a site visit to the Rock Creek property on April 22 and 23, 2003. Ken Kuchling has not verified any of the existing geological data nor has he verified the assumptions used to build the block model. Data verification and block modelling was undertaken by NovaGold technical staff. The completed block model was provided to Norwest by NovaGold for use in this study.

PROPERTY LOCATION AND DESCRIPTION

The Rock Creek Project is located on the Seward Peninsula along the west coast of Alaska, north of Norton Sound. The project area lies about 13 km north of Nome and is accessed via state maintained roads, as shown in Figure 2.1.

The terrain is fairly hilly with broad and narrow valleys. Nome is located at sea level while the Rock Creek plant site would be at an elevation of about 70 masl (metres above sea level). The mine area is higher, at elevations ranging from 100 masl to 150 masl.

Vegetation at the site consists mainly of low shrubs and grasses. Forested areas and trees are non-existent in the mine area.

The project occurs partly on patented mining claims owned 100% by the Alaska Gold Co., a wholly owned subsidiary of NovaGold and partly on land controlled by the Bering Straits Native Corporation (BSNC). Bering Straits Native Corp also owns local mineral rights with surface rights owned by Sitnausuak (the Nome native village corporation). The Bering Straits agreement involves escalating annual payments up to production, a 2.5 % Net Smelter Return (NSR) royalty and a 5% Net Profits Interest from production from BSNC lands. The known resource at Rock Creek lies within land owned approximately 66% by Alaska Gold Co. with the remainder within Bering Straits Native Corporation lands.

The nearest area to the Rock Creek prospect that is closed to mineral entry is the Bering Land Bridge National Preserve which is over 60 miles northeast of the Rock Creek prospect at its closest point. There currently are no unusual social, political or environmental encumbrances to exploration, development or production on the prospect.

CLIMATE

The project site is characterized by cool summers and cold winters. Summer temperatures range from +8⁰C to +15⁰C and winter temperatures average around -15⁰C. Figure 2.2 provides a chart of seasonal air temperatures, showing mean, and average high and low temperatures.

The project site is also characterized by relatively low annual precipitation averaging less than 42 cm with the majority of precipitation falling as rain in summer. Figure 2.3

LOCAL INFRASTRUCTURE

The city of Nome (population 4,000) is situated on the Bering Sea coast and serves as the logistical and administrative center for this portion of western Alaska. Nome has daily commercial jet service from Anchorage and has large container barge service from June through October.

The Rock Creek prospect is currently road accessible via the Glacier Creek Road and the state maintained Teller-Nome Highway, an all-weather paved and gravel road. There are plans by the State of Alaska to construct the Glacier Creek Road By-Pass, shown in Figure 2.1, and engineering for the road is underway. Construction of the 5 km long by-pass road will simplify the road access distance to site.

The city of Nome has provided electricity to past mining operations and has offered that service for future operations if necessary. Current generating capacity is about 12 MW with three diesel generators and there are discussions that local capacity will be increased up to 15 to 20 MW with newer more energy efficient diesel generators. The current local power consumption is in the range of 4 to 6 MW.

No camp facilities are required at the Rock Creek Project due to its close proximity to Nome, which is well serviced with accommodations.

REGIONAL GEOLOGY

The Seward Peninsula is composed of a complex series of metamorphic rock packages that have been affected by large-scale accretionary, rotational and transcurrent events.

The project area is underlain by metamorphic rocks belonging to the Nome Group, which is thought to constitute a coherent litho-stratigraphic succession consisting of four major units;

LOCAL GEOLOGY

The Rock Creek mineralized zone covers an area of approximately 500 by 1500 meters. The main rock types are schist and quartz muscovite schist (QMS). QMS is dominant but other types of schist are common in the deposit area. The overall bedding is relatively flat with a slight dip to the northwest. Post-mineralization faulting is evident in the core with gouge zone typically 1 to 3 meters wide. Mineralization at Rock Creek is confined to quartz veining. The gold grains are coarse and visible gold is very common.

Exploration efforts have determined that gold was deposited at Rock Creek in replacement bodies, in tension veins, and in the Albion shear zone. Only the latter two, i.e. tension veins and Albion shear veins, were modeled. The resource in the tension veins accounts for approximately 90% of the tonnes and 85% of the gold content in the deposit.

Computer Modelling (supplied by NovaGold)

The resource model for the Rock Creek Gold Deposit is built on the geological model as defined in drill holes. Geological continuity is established and defined by the stratigraphic model. Structural and mineralization controls are somewhat poorly captured and are not modeled and the current mineralized shell is purely based on assay.

Two geologic domains were modeled to restrain the interpolation at Rock Creek.

(a) The Albion zone is a persistent sub-vertical shear zone that extends beyond the resource area along strike and at depth. It is composed of several tightly spaced quartz veins within a corridor of strong shearing and alteration. It consists of the Albion east and Albion west structures.

(b) The tension vein quartz muscovite schist (QMS) zone is relatively flat lying and appears to be controlled in part by host stratigraphy. The tension vein zone contains multiple narrow sub-vertical sheeted tension veins that have the same strike and dip as the Albion zone shear. Cross-section work shows that only the tension veins host significant tonnage of sheeted veins so the tension veins are used to limit kriging of the sheeted veins both laterally and vertically. Below the tension veins, the rocks are poorly mineralized and have been excluded from the resource model.

The Rock Creek model (project area) was built using geological information and assay data collected from 99 core holes (10,092 m), 118 RC holes (8,807 m). A total of 10,222 assays were used in the model. The modeling philosophy is based on using geologic features and distribution of gold mineralization in controlling the extent of interpolation. The Albion veins and the QMS were modeled as 3D solids.

The block model used for the scoping study resource estimation of the Rock Creek Gold Deposit was constructed using MineSight® mine modeling software and SaGe® for variography. The final model consists of 4,080,000 blocks, and all blocks have dimensions of 5m x 5m x 5 m. Sub-blocking was used along the Albion contacts. The model was formulated in local mine grid coordinate system which is rotated 50 degrees from true north.

The items modeled include topography, felsic intrusive, faults, specific gravity, mineralized zone and in situ gold grade. A current surface topographic model that covers the resource model was used to code that portion of the blocks below the surface. A field in the block model named TOPO was initialized and coded with the percentage of each block below the surface. Overburden is minimal in the Rock Creek resource area and therefore no overburden has been coded in the model.

Mineralized shells were developed for both the Albion zone and the tension vein zone. All blocks, or portion of blocks, within the Albion zone are considered mineralized. For the tension vein zones, acceptable mineralized envelops were defined by utilizing Probability Assisted Constrained Kriging or PACK (Pan, 1995). NovaGold's project geologists and a geologist representing AMEC selected the shell used to outline the populations. Finally, it was assured that the nominated boundaries did not violate current geologic understanding of mineralization controls.

Ellipsoidal search was used to limit the interpolation and to avoid smearing grade laterally. Rotations and ranges are based on geological information and geostatistical analysis.

Geological Model Review

The geological model development for pit designs and subsequent economic evaluations is outside Norwest’s scope of work. However, prior to proceeding with this type of assessment several preliminary checks are completed to ensure the model is reasonable. This level of checking does not test or validate assumptions, rather the methodology generating the model is reviewed and the model is compared to the base data and assumptions.

on a gold price of US$325/oz is estimated to be about 0.1 g/t less than the breakeven grade. Table 3.2 also summarizes the sensitivity of the mill cut-off grade for gold price. The mill cut-off grade is not used in the production schedule for this study but may be considered at the feasibility stage for stockpiling of marginal ore.

When describing the in-pit resource in Section 7.1, the breakeven cut-off grade of 1.0 g/t will be used to summarize the resource. However the additional resource contained between the actual breakeven cut-off (0.85 g/t) and 1.0 g/t cut-off grade will also be quantified.

Recovery Criteria

The two ore types that have been modelled each have different process recovery criteria. The reasons for the differences in recovery are uncertain at this time but may be related to sulfosalt and silicification.

Section 10.1 describes the results of metallurgical test work. These studies indicate that the Albion vein quartz poses the greater challenge in gravity recovery circuits, with values in the range of 30%-40%. The tension vein material yields higher gravity recoveries in the range of 80%. Cyanidation of the whole rock yields fairly high recoveries for both ore types.

The application of recovery criteria to the tension vein ore is fairly straightforward. Processing the Albion zone represents a more complex situation. The Albion zone can consist of a mix of both Albion quartz and tension vein type quartz. Geological logging records percent volume of Albion quartz (within a 5-level classification scheme) and percent volume of tension vein quartz and three other vein types, along with a count of tension veins. However logging records those values in natural geological intervals, not in the 2-metre long sample intervals. Thus identifying the percent volume of quartz types in a particular sample can be a tedious and non-automated effort if logged geological breaks don't coincide with sample breaks. Both Albion and tension vein quartz can be intimately

intermingled and a very small volume of one type can exist in an intercept dominated by the other.

One could expect that theoretically each Albion zone mining block might potentially have a different recovery depending on the ratios of Albion quartz to tension vein quartz contained within. Furthermore the results of the metallurgical lab testing will also depend on the respective quartz ratios. Since all of the metallurgical samples tested likely consisted of mixed quartz there maybe no real data on the recovery rate for "pure" Albion quartz.

For design purposes, Table 3.4 summarizes the recovery factor used in pit optimization and project design, assuming a gravity and flotation milling operation.

Block Model Summary

NovaGold staff has prepared the geological block model for the purposes of this study. Table 3.5 summarizes the unclassified block model resource estimate. Further documentation on the block model resource and capping methodologies are described in NovaGold internal documents.

Pit Slopes

At this stage of study, the site specific geotechnical data gathered to date is limited to core descriptions, rock quality (RQD) and hardness information from 2002 and 2003 drilling. No geotechnical testing of core to determine shear strength has been carried out and no instrumentation has been installed to confirm permafrost or groundwater conditions. It is expected that this and other information will be gathered during subsequent stages of the study and specific recommendations for further geotechnical work are included in Section 14.

In the design of pit slopes in the relatively competent rock at Rock Creek, failure through the intact rock will generally not be a realistic failure mode. Wall stability will be governed by the presence of discontinuities such as jointing, shearing and faulting, which can act as planes of weakness along which movement can occur. Depending on the orientation and frequency, these discontinuities can also act as release surfaces that allow blocks of rock to separate from the rock mass.

A review of the available RQD data from 45 and 55 degree inclined drill holes around the potential pit limits, clearly shows a marked variation in fracture/joint frequency with hole depth. Figure 4.1 shows the RQD data from holes RKDC02-108,109 and 110 outside the north end of the ore body, RKDC02-106,107 and 111 at the south end and RKDC02-110, RKDC03-117,121 and 123 near the centre of the ore body. All holes were drilled in the same direction (140 degrees azimuth) at dips of 45 or 55 degrees from horizontal.

There is a wide spread of relatively low RQD values (average 10 to 25) through what is assumed to be the main ore bearing shear and tension zones (Albion shear and QMS) down to an inclined hole depth of 24 metres (80 feet) or a vertical depth of about 17 metres, with a marked increase in RQD (average 50 to 70) at an inclined hole depth of 24 to 34 metres (80 to 110 feet) or a vertical depth of about 17 to 24 metres. A second zone of lower RQD (average 15 to 30) is indicated between 34 and 40 metres inclined depth (110 and 130 feet), again increasing below. The zone of increased RQD may represent what may be expected in the pit walls outside the main ore zone, although none of the holes with RQD data appear to be deep enough to have penetrated beyond the QMS. This would explain the second zone of lower RQD below an inclined depth of 34 metres.

Based on the indications of the limited data available on hardness (R3+) and assuming pit walls formed in hard rock with an RQD of around 60, stable inter-ramp pit wall angles in the range of 40 to 50 degrees should be possible for wall heights in the range of 150 metres. If subsequent investigations into the proposed pit walls indicate higher average RQD values of over 75, then wall angles in the upper end of the range may be adopted. However, if lower RQD values of less than 40 are found to be more representative, then wall angles at the lower end of the range will be required. The latter may well apply at shallow depth within the QMS. Along the final pit walls it is possible to use double or triple benching to increase pit wall angles.

At this scoping level of study it is recommended that an inter-ramp pit wall angle of 45 degrees be adopted in design but that sensitivity to the range of 40 to 50 degrees should also be evaluated.

These slope angles assume that any excess pore water pressures below permafrost in the walls are can be controlled or naturally dissipate (see Permafrost Section 4.2). Further investigation work is required to confirm this assumption.

Development Rock Dumps

Development rock dumps located on the valley side slopes to the east and west of the pit will be constructed on regolith that is typically residual soil comprised of silty sand and gravel gradations. No geotechnical investigations have been conducted in these materials to date, but from inspection of core photos (see RKDC02-107 in Figure 4.2), these surficial materials are expected to provide a competent foundation for the dumps.

With maximum side hill slopes at the potential dump sites of about 10 degrees (Figure 4.3), stable rock dump slopes of 2.5H:1V should be possible and stable soil dumps slopes of 3.5H:1V may be assumed in design. Mixed rock and soil dump slopes should be stable at 3H:1V. It is assumed that only nominal thicknesses of organics will be present and that dumps can be developed in lifts of about 5 metres, without the need for any pre-stripping. However, these assumptions need to be confirmed at the next stage of design.

Tailings Impoundment

No geotechnical information is available on the proposed site for the tailings impoundment at the lower end of the Rock Creek valley and no information has yet been gathered on suitable construction materials for the impoundment dykes.

However, from inspection of some typical regolith core, it is likely that impoundment starter dykes may be constructed on natural grade stripped of organics without the need for any special measures to enhance bearing capacity or limit seepage. It will be very important to investigate the full perimeter of the proposed impoundment to confirm that a uniform thickness of low permeability regolith and or alluvium is present and to determine a suitable depth of key to be constructed under the main dykes. It will also be important to determine the depth to bedrock and the nature of jointing and fracturing in the bedrock surface.

At this level of study it is assumed that sufficient fine-grained material will be available from the regolith excavated during the initial pit opening to construct a tailings dyke with an impermeable core or upstream blanket over a shell of development rock. If sufficient fine-grained materials are available, then a homogeneous dyke section may be considered. Figure 4.4 is a conceptual dam cross-section.

Under the ponded surface of the facility it is likely that thawing of the permafrost foundation will occur. For this reason, it will be important to thoroughly investigate permafrost in the area, including ground temperatures and any excess ice conditions.

Infrastructure

Although there is no geotechnical information available on the proposed plant site, foundations for buildings and the various items of plant are not expected to require any special methods of construction other than those dictated by permafrost conditions. For suitable types of foundation in areas of permafrost refer to Section 4.2.3

Recommendations

Fieldwork
Other than some RQD and rock hardness data generated from the 2002 and 2003 core drilling, no geotechnical site investigation has yet been carried out at the proposed mine site. As the project proceeds through the various levels of design, site specific information will need to be gathered to support the basic mine, development rock dump and tailings facility designs, as well as the design of building and plant foundations. The following summarizes recommendations for gathering geotechnical information considered necessary for the next level of study.

Introduction

Permafrost is defined as a state of ground that remains at or below a temperature of 0ºC for a minimum period of at least two consecutive years. No distinction is made in the permafrost definition for whether the frozen ground is rock or soil as long as the temperature regime meets the criteria. Permafrost develops in areas where the heat loss from the ground during winter exceeds the combined energy gain during the summer and energy radiated by the local geothermal gradient (i.e. heat radiating upwards from depth). Consequently the thickness of permafrost and actual permafrost temperatures can vary locally, depending on air temperatures, the geothermal gradient, snow cover, exposure direction and topographic conditions. Typically north facing exposures will have greater permafrost depths than south facing exposures. Areas overlain with organic soils and snow cover (i.e. insulators) will have less permafrost than barren rock exposures.

Average annual air temperatures in Nome are about -3°C; seasonal fluctuations are shown in Figure 2.2. Published map data indicates that the Rock Creek Project site is located near the regional boundary between continuous and discontinuous permafrost.

To the best of our knowledge no thermistors or permafrost measurements have been made at the Rock Creek site. Published data indicates that permafrost depths in Nome could be approximately 100 metres. The lack of trees and the presence of frozen placer gravel deposits are indicators of the permafrost regime. In the NovaGold report, "1999 Rock Creek Drilling & Metallurgical Program", mention is made on Page 15 that some of the drilling at Rock Creek is shallow and "cuts off at the approximate depth of the water table" at about 60 metres. This could be an indicator of the permafrost depth in the mine area rather than the actual water table.

Within the permafrost horizon, a near surface zone of the permafrost will thaw and re-freeze on an annual basis as seasonal air temperatures change. This upper zone is termed the “active layer” and could be about 2 to 3 metres at the Rock Creek site. In hard rock, the active layer will be deeper due to a higher thermal conductivity, and could be closer to 5 to 6 metres.

At a certain depth below the ground surface, the seasonal air temperature fluctuations become attenuated and the ground mass essentially remains at a

Engineering Properties of Permafrost

A key characteristic that impacts on the engineering properties of permafrost is that the majority of the groundwater or soil moisture is frozen. At temperatures a few degrees below freezing, a small amount of water remains unfrozen, bound to silt and clay sized particles in the soil matrix. The presence of ice in the pores of a soil or in rock fractures tends to increase the shear strength of the material. It also reduces the hydraulic permeability such that the material can be considered essentially impervious. The presence of interstitial ice, particularly ice lenses, causes the soil to exhibit creep behaviour under loading. This means the material may very slowly strain or deform under loading. The colder the permafrost temperature is, the higher the developed shear strength will be and the lower the creep rate.

Permafrost soils can be classified as “thaw-stable” or “thaw-unstable”. Thaw unstable permafrost loses a significant amount of strength upon thawing, due to presence of ice and excess water when melted. Thaw stable permafrost is generally ice-poor and will not settle appreciably when thawed. Generally more conventional type foundation designs can be used in thaw-stable permafrost while special permafrost protection measures may be required in the thaw-unstable areas. Where bedrock is near surface, it maybe best to seat the footing in the bedrock and excavate frozen soils entirely.

Where engineering designs purposely utilize the favorable strength or permeability characteristics of the permafrost, it is imperative that the cold temperature regime is maintained in order to retain the desired engineering properties. Construction activity can warm or degrade the permafrost in several ways. For example, the removal of layers of insulating organic vegetation, the ponding of water due to changes in topography, or heat radiating from buildings can change the temperature regime at the ground surface and cause permafrost thaw. This thawing could then lead to a reduction in shear strength or even melting of in-situ ice lenses, creating ground settlement or seepage pathways. Settlement could then lead to additional water ponding, progressively degrading the permafrost. The thawing of ice filled rock fissures can create potential water seepage pathways if incorporated in a water retaining structure. The convective heat transfer caused by the flowing water may result in even further permafrost degradation in such cases.

of the side slopes. Consequently, constructed embankments and roads can be used to provide runoff containment, where practical. However in order to maintain proper drainage grades, it may be necessary in some locations to excavate ditches. Armouring of ditch slopes in thicker overburden can help to maintain stable slopes and will prevent till erosion during periods of water flow.

In some places it may be necessary to excavate ditches in bedrock and no permafrost protection measures will be required.

Where settling ponds are required small, low-height dams may be needed. These dams should be of sufficient height and dimension that they eventually freeze. The freeboard should be at least 2 metres above the maximum water level to be contained so that active layer in the dam remains above water level.

Small Structure Foundations
Small structures with low ground bearing pressures may be constructed with several alternate footing designs. The selection of the specific foundation design will depend on the depth to bedrock at a given site and the type of overburden cover present.

Where bedrock is at or near surface, shallow piles socketed directly into the bedrock can be used to support the structure. Alternatively a shallow thickness of frozen overburden can be sub-excavated down to bedrock and then replaced with compacted fill. Slabs on grade can also be utilized where the bedrock is at surface.

Where bedrock is at depth other means of support are available, although we expect such measures will likely not be needed at the Rock Creek site. Adfreeze piles frozen into the permafrost are one option available. These piles are placed to depths sufficient to resist annual freeze/thaw uplift forces and also support the loads imposed by the structure. A layer of compacted fill can be used to support the structure on top of the frozen soil and insulate the sub-grade. It may also be possible to elevate the structure off the ground surface, providing a ventilated pad beneath such that cold winter air can assist in keeping the ground frozen. Consideration will also be given to the incorporation of synthetic insulation in the sub-base design to reduce the thickness of fill required

Large Structure Foundations
Large structures, such as the mill building, truck shop, and diesel storage tanks should be constructed directly on bedrock foundations. Localized pockets of overburden and regolith (i.e. weathered bedrock) should be excavated down to

competent rock, where necessary. Heat radiating from these structures will likely cause some thaw of the underlying permafrost however since it will consist of bedrock, no settlement is expected. No special permafrost protection measures are envisioned in these areas.

Development Rock Dumps
The development rock dumps will be used for disposal of rock stripped during the mining operations. The rock piles will be constructed over various permafrost foundation conditions including bedrock, with some overburden veneer. No special construction methods are envisioned for the development rock dumps, although if geotechnical investigations indicate the presence of ice-rich soils, then some provisions may be required to minimize slumping of dump slopes on thawing soils.

Over time the rock dumps will freeze into permafrost, however the large pore spaces will probably still allow seepage through the dumps. Studies have shown that sulphide oxidation and acid generation rates may be reduced in frozen rock but would likely still continue.

Tailings Management Facility
The tailings storage facility will consist of sub-aqueous deposit tailings covered by a water cap. The downstream slopes of the tailings dams will likely freeze over time.

Some thaw of the permafrost bedrock underlying the interior of the pond is likely. The thaw zone should neither extend laterally through the dams or downwards very deep. Therefore some rock fractures at depth should remain frozen and ice-filled, maintaining an impermeable barrier to subsurface seepage.

Upon project closure, once the ponded tailings water is discharged the exposed tailings beaches should begin to freeze, eventually re-establishing the permafrost regime within the impoundment. Most likely the entire tailings area will become permafrost except for the upper 2 to 3 metre active layer.

Mining
The Rock Creek pit areas are expected to be contained within permafrost rock. The geotechnical design of the pit slopes may incorporate rock shear strength increases due to permafrost. As well seepage and pore pressure conditions should take into the presence of permafrost. Where present in the pit wall, the permafrost zones are considered impermeable rock.

SITE WATER MANAGEMENT

Surface runoff from undisturbed areas outside the project footprint will be collected in interceptor ditches and diverted around the project area. Settling or polishing ponds will be used to remove suspended solids prior to discharge to the environment. Figure 6.1 shows the location of the interceptor ditches. These settling ponds will also be equipped with submersible pumps, which can be used to withdraw water for process use.

Surface runoff from disturbed areas inside the project footprint will be collected in mine water ditches and storm water sumps. These sumps will be equipped with submersible pumps, which will discharge to the tailings management facility.

Runoff from the downstream slope of the tailings dam and dam seepage will be collected in a toe ditch and return to the tailings pond. No hydrological studies or detailed sizing of ditches and sumps has been undertaken at this time.

The pits are expected to be largely contained within permafrost and therefore minimal groundwater seepage is anticipated until mining at depth. However surface runoff and pit seepage will be pumped out of the pit into the mine water ditches for re-use in the process. Diesel power pumps will be used for mine dewatering.

Option 4 was selected as the preferred tailings facility option based on trade-of between footprint area and dam height, with footprint minimization be the driving factor. Option 4 results in a southward shift, providing more space for the mill site yet still locating the pond over the Rock Creek gulley. The Rock Creek channel provides a convenient place to locate the process water reclaim barge.

Once the final mine plan was developed, the tailings tonnage increased to about 11 million tonnes (8.5 million m³). The selected location was able to accommodate the added tailings by raising the dams higher by about 7 metres.

Condemnation drilling must be done to discount the presence of any mineable ore zones in this area and to evaluate the extent of the mineralization to the southwest of the main ore zone.

PIT OPTIMIZATION (LERCHS-GROSSMANN)

Prior to developing the pit designs and mine layouts, the Lerchs-Grossmann (LG) pit optimizer was used to evaluate pit geometries based on variable economic factors. Each variation in revenue or cost would result in a different optimized pit. As a first pass, the pit optimizer is run using estimated mining and processing costs. Once the final mine plan and production schedule is developed, more refined estimated costs can be developed. If these costs were significantly different than those used in the pit optimization, the Lerchs-Grossmann would be repeated with the new costs. As will be discussed later, the final operating costs were not significantly than those used herein and the optimization work was not repeated. The "capped" block grades were used for optimization purposes.

Table 6.2 summarizes the parameters (under the column labelled Base Case Values) used in the pit optimizations. In addition a series of sensitivities were run on different factors, also summarized in Table 6.2.

The base case pit slope assumed that the west wall would have an overall slope of 30° due to the likely presence of haul roads along this wall of the pit. The east pit wall was varied from 45° to 50°. The case of a 40° pit slope around the entire pit was also modeled to simulate a spiral ramp.

Gold Price Sensitivity
Table 6.3 and Figure 6.3 describe the impact of the recoverable gold resource with a variable gold price. A significant sudden increase in recoverable gold occurs at a price of $275/oz, which is the point at which the two smaller pits coalesce into one larger pit. Beyond that point, the recovered gold increases gradually, at a rate of about 10,000 ounces for each $25 increase in the gold price.

Milling & Admin & Sustaining Capital (MAS costs) cost sensitivity
The combined milling, administration, and sustaining costs (MAS costs) were varied above and below the base case cost of $7.50/t. The results, shown in Table 6.3, indicate that the recoverable gold would increase by about 28,000 ounces for each $0.50/t decrease in MAS cost. The milling cost sensitivity is therefore not as great as the gold price impact.

Albion vein recovery sensitivity
The Albion vein recovery was varied downward from base case value of 94%. The results, shown in Table 6.3, indicate that the recoverable gold would decrease by about 12,000 ounces for each 5% decrease in recovery. The relatively low impact of the Albion ore recovery is due to the fact that only 22% of the block model gold is contained in the Albion zone.

Tension vein recovery sensitivity
The Tension vein recovery was varied downward from base case value of 97%. The results, shown in Table 6.3, indicate that the recoverable gold would decrease by about 50,000 ounces for each 5% decrease in recovery. The large impact of the Tension ore recovery is due to the fact that most of the block model gold is contained in the Tension zone.

Pit slope sensitivity
The overall pit slopes were varied from base case value of 40° which is predicated on a 45° inter-ramp angle and the inclusion of haul roads. The results, shown in Table 6.3, indicate that the recoverable gold would increase by less than 10,000 ounces by steepening the pit slope from 45° to 50°.

Conclusion
Based on the foregoing sensitivities, it was decided to develop a pit design using the $325/oz LG pit as the starting point. This shell encompasses the single large pit configuration and should yield between 650,000 and 700,000 ounces of recoverable gold. The incorporation of ramps into the pit design will alter the ore and waste volumes somewhat from those estimated in the optimized pit shell.