Amur Minerals — Recovery abounds

Amur had a record field season in 2016, drilling 80 holes of 19,400.8m in aggregate length compared to a target of 15,000m. In so doing, the company confirmed the presence of continuous mineralisation along a 3,000m strike length at MKFL (cf 2,100m previously). Of the 59 (15,213.3m) resource definition holes, those intersecting ore grade mineralisation recorded an average mineralised thickness of 13.5m per interval, an average cumulative thickness of 23.6m per hole and an average length weighted grade of 0.76% Ni and 0.21% Cu (using a 0.2% nickel cut-off grade). In addition, the previously identified high grade structure (defined as >0.5% Ni) has been confirmed to exist along the entire 3,000m strike length of MKFL, with an average mineralised thickness of 10.5m per interval, an average cumulative thickness of 19.7m per hole and an average length weighted grade of 0.90% Ni and 0.25% Cu.

Of note is the fact that the headline grades of the holes drilled during the 2016 field season were higher both on average and specifically within the high grade domain than those recorded in the Kun-Manie May 2016 resource estimate. Cumulative widths in excess of 20m in the western extension of MKFL also compare with widths of c 12m elsewhere in the deposit. All MKFL drill identified mineralisation has now been conducted at a spacing deemed by SRK to be suitable for the definition of indicated resources.

On 10 February, Amur announced an updated mineral resource estimate, as compiled by consultants RPM. Unlike the previous resource estimate, dating from May 2016, which was calculated at a cut-off grade of 0% nickel, the new resource estimate was calculated at a cut-off grade of 0.4% Ni. As a result, sub-economic mineralisation is no longer included in the global resource inventory. A comparison between the January 2017 mineral resource estimate (at a cut-off grade of 0.4% Ni) and the May 2016 mineral resource (at a cut-off grade of 0% Ni) is as follows:

Of note, within this context, is the material increase in the resource at Maly Kurumkon, as a result of the exploration conducted during the 2016 field season and despite the increase in the cut-off grade. This contrasts with three other deposits (Vodorazdelny, Ikenskoe and Kubuk), at which no additional drilling has been conducted since May 2016 and at which the changes in resources therefore only reflected a recalculation based on a higher cut-off grade. As expected, in these cases, the higher cut-off grade resulted in a decline in the overall ore tonnage, an increase in the in-situ grade of nickel and a proportionately smaller decline in the number of contained nickel tonnes. One interpretation of the 2016 field season therefore is that it has more than replaced hitherto sub-economic resources with economic ones. Note that no recalculation was performed at Gorny, which is a relatively small deposit and the only one with an average grade below the 0.4% cut-off grade at the time of the H116 mineral resource estimate. As a result, Gorny has been excluded from the above table, although future drilling may result in its re-inclusion once again, at a later date.

Nevertheless, direct comparison may be made between three deposits, for which a range of resource estimates were provided at varying cut-off grades (including 0.4% Ni) in Q216:

As such, whereas the headline increase in resources (Exhibit 1) is 50,040t (+6.8% including Gorny or +10.3% excluding Gorny from May 2016 estimate) of contained nickel, compared to the previous resource, once the change in cut-off grades is taken into account, the underlying like-for-like change is 213,700t, or 41.7%, of contained nickel. Of note, once again, is the disproportionate increase in the resource at Maly Kurumkon on account of the exploration work performed in 2016, compared to the three other deposits, as a result of which it now comprises 61% of the Kun-Manie resource (cf 50% previously):

In nickel and gold (as an illustrative exercise) equivalent, Amur’s total mineral resource can therefore be stated as follows (at prevailing metals’ prices at the time of writing):

Based on our end-FY16 cash estimate, Edison estimates an enterprise value for Amur of US$56.3m, which equates to US$73.47 per total nickel resource tonne. That being the case, a headline resource increase of 50.0kt Ni should have added US$3.7m in (pro rata) value to Amur, while a like-for-like increase of 213.7kt should have added US$15.7m – compared to an estimated cost of the associated drilling of c US$1.3m (being 15,213.3m of resource definition drilling at management’s all-in estimate of drilling costs of US$75-100 per metre drilled).

After a 2016 field season focused on MKFL, Amur is now targeting Kubuk in the forthcoming 2017 field season. The existing Kubuk resource is contained within a 1km strike length.

Currently, 10,000m of drilling is planned in the area of the existing resource to upgrade a 10.9Mt inferred resource block into the indicated category (similar to MKFL drilling in the 2016 field season). The inferred block is currently the largest single continuous block of inferred resources within Kun-Manie and will account for c 6,000m of (in-fill) drilling by the company’s LF90 drill rig working at depths typically exceeding 200m. In addition, a minimum of seven metallurgical holes are planned, totalling a further 1,500-2,000m. The remaining c 1,000m will then be conducted as 1) step-out drilling over an additional 1km of strike length extension to the east of the existing resource and 2) to the west of the existing resource area in the 3km gap between Ikenskoe-Sobolevsky and Kubuk in an initiative to prove continuity between the two. NB In the event of continuity, this is likely to prove the deepest part of the orebody within Amur’s licence area.

A total of 5,000m of drilling are also planned for Amur’s LF70 drill rig at Ikenskoe-Sobolevsky (where holes are typically shallower – eg less than 200m). Approximately 2,000-2,500m is planned for completion on the Sobolevsky part of the deposit, located to the south and east of the currently defined drill identified mineralisation, with the objective of expanding the resource in the direction of Kubuk. As at Kubuk, additional drilling will also be completed to generate a metallurgical sample for this deposit.

Beyond the availability of water from the Maya River, Amur plans to complete six to eight holes to establish potential sources of groundwater for industrial use to process the ore. Finally, drilling will also be completed at Maly Kurumkon-Flangovy to establish potential water inflow rates within the planned open pits and underground operations in order to establish dewatering requirements for the mine.

Inevitably, the amount of drilling completed will be weather dependent. Typically, c 15,000m of drilling can be completed by two drill rigs in four to five months. However, sufficient supplies have been procured to support 20,000m of drilling, should the opportunity for an extended season present itself. Additional equipment purchased to facilitate the completion of work necessary for the ongoing development of Kun-Manie includes a 25t truck mounted crane, a Caterpillar 320D2L excavator, a Caterpillar D6RII bulldozer, a water well drilling truck, Ural trucks (for fuel, drop side loading and personnel transport), portable cabins, drill water pumps and additional power generators at cost of c US$1.04m.

At an average cost of US$40/m for drilling alone, this 15,000m programme is anticipated to have a total direct cost of c US$0.6m.

The 2017 field season is anticipated to conclude the major period of Amur’s exploration activity. In due course, Amur intends to explore the 900m gap between the MKFL and Gorny ore-bodies (of which 400m lies on the easternmost side of the MKFL area and 500m lies within the Gorny area). If drilling of these extensions proves successful (ie intersects mineralisation), Amur believes that it is possible that the MKFL and Gorny deposits will prove to be one larger deposit with a strike length of up to 5km.

On 11 January, Amur announced the results of Gipronickel’s metallurgical test programme on a c half tonne sample of ore derived from half core from three drill holes located within the Maly Kurumkon-Flangovy (MKFL) deposit, which comprises the majority (55% by ore tonnage) of its Kun-Manie production licence area.

The Gipronickel results represent the first production scale test work from the Kun-Manie licence area and, owing to their larger size, are expected to be more reflective of the actual production process than those calculated in bench-scale tests previously conducted by SGS. Flotation test work on the sulphide ores by SGS was concluded on 12 samples covering six incremental grade ranges distributed throughout the JORC-drilled areas of MKFL and Kubuk. The results of these tests were released to the market in August 2016 and are summarised in Exhibit 7, below, for an average 0.7% nickel grade and are compared to the Gipronickel results for the half tonne bulk sample (actually 443.9kg), which had an actual nickel grade of 0.7% nickel.

In addition to issues of comparability with SGS’s result, the 0.7% Ni grade of the half tonne sample also approximates the 0.75% Ni average grade of the mineable reserve calculated by RPM in its open pit/underground production trade-off study. Notwithstanding its larger size, recoveries from the Gipronickel half tonne sample are self-evidently materially higher than the results achieved by SGS (see Exhibit 7). In part, this may be attributed to the fact that the SGS test work involved coarse pulverisation of the samples and included older material that is likely to have partially oxidised, resulting in lower recoveries. In addition, however, Gipronickel employed a two-stage grinding process such that, after initial grinding and concentrate generation, the reject stream was reground to allow for the recovery of a second concentrate.

Gipronickel will progress its analysis of the metallurgical characteristics of the Kun-Manie ore bodies via the processing of a 7.5t bulk sample recovered in the 2016 drill programme, which is currently inventoried in Amur’s core and storage sample facility in Khabarovsk. In the meantime, the Gipronickel results also indicate a higher recovery to concentrate than previously calculated and higher concentrate grades of 8.58% Ni and 2.10% Cu. An improved mass-pull of 6.6% (vs 7.0% previously) implies the production of 394kt of concentrate on average per annum from 6.0Mt of ore, compared to 420kt previously and a consequent reduction in fleet transport and in capex (eg via a smaller concentrate treatment facility) in the event that Amur opts for a toll smelting development option. Further capex savings (eg via the requirement for a smaller flash furnace) are also possible in the event that Amur opts for a matte or refinery development option.

Amur’s metallurgical test work announcement follows the release of the results of a trade-off study between open pit and underground mining, conducted by RPM in late December, which identified a potential mineable reserve of 44.5Mt of ore at grades of 0.75% Ni and 0.19% Cu at MKFL (cf a mineable reserve of 45.5Mt identified in Amur’s preliminary economic assessment (PEA), at an average grade of 0.53% Ni and 0.15% Cu). Key differences between the RPM study and the original PEA are a materially higher underground component of the overall mining operation and also materially higher grades overall, as shown below:

The RPM study assumes that the nickel and copper are together recovered into an aggregate 2.5Mt of concentrate (ie a 5.6% mass-pull) containing an average of 9.9% Ni and 2.9% Cu (ie based on SGS, rather than Gipronickel, grade-recovery curves, above). In addition, the RPM study may prove conservative in that it assumes Western Australian underground mining costs, contributing to a total operating cost of US$40.02 per ore tonne. It also did not include the results of 2016 field season considered above.

Amur’s continuing focus on the high-grade domains within the ore-body in conjunction with the results of the RPM mining trade-off study, in particular, are changing management’s perception of the likely mining outcome at Kun-Manie, with an increasing focus on underground operations. Where before Kubuk has been presumed to support both open pit and underground operations, it is now increasingly presumed to be predominantly open pit (although there could be a small, start-up open pit) with the result that the mine plan is evolving from the 50:50 underground:open pit operation envisaged in the most recent operational blueprint (on which Edison’s valuations are based) into something closer to a 75:25 production split. Even this may understate the extent of ultimate underground mining, given that reserves are calculated with reference to Australian underground mining costs, which could prove to be in the order of 100% higher than Russian ones. Using the 0.4% cut-off grade and a metallurgical recovery of 80%, Amur now projects the break-even price of nickel for Kun-Manie to be US$3.40/lb based on Russian mining costs and US$5.70/lb based on Australian mining costs (cf a price of US$5.03/lb Ni at the time of writing).

In addition to the mine plan, capex estimates relating to the project are also continuing to evolve. Amur has recently completed a survey of the terrain to be covered by the proposed access road to site, for example, including identifying sources of gravel and bridge locations. Notable developments in this respect are reported to be that it may be possible to utilise existing (albeit primitive) logging roads for part of the distance. In addition, initial indications are that the average cost of the road is likely to be in the order of US$400,000 per kilometre in mountainous regions and c US$150,000 per kilometre in other areas – both of which are substantially less than the company’s currently budgeted US$1m per kilometre. Although the precise savings cannot be established in the absence of specific and detailed road design, Amur does nevertheless anticipate a substantial reduction in this capital cost category with a significant portion of the decrease being attributable to the c 50% devaluation of the Russian rouble since the compilation of the original estimate. Note that Amur is currently compiling detailed topographic maps along the entire planned road route for use in the commissioning of a desktop study to develop more accurate costings in respect of its construction, prior to ‘walking the course’ as soon as the spring thaw sets in.

At the same time as advancing the technical aspects of the project, Amur has been cultivating its relationships with parastatal and banking organisation. These include:

Ahead of development funding therefore, Amur already has relationships with a range of sovereign funds, parastatal organisations, Chinese funds, western and Russian banks and overseas corporations.

In the absence of an updated mine plan, Edison’s forecasts and valuations are necessarily still based on Amur’s 2015 operational blueprint, although Amur estimates that the results of Gipronickel’s metallurgical test work alone could result in revenues being “increased by as much as 10% for nickel and 6% for copper” (note: to which end, readers are specifically directed towards Exhibit 11). At the same time, given the size of the newly defined resource and substantially higher grades, the previously projected annual production of nickel and copper at Kun-Manie could, in management’s opinion, “be increased by as much as 68% for nickel and 54% for copper”.

With this caveat in place, in our most recent valuation of Amur, we estimated values of the concentrate, low-grade matte, high-grade matte and refined metal options for Kun-Manie of 29c, 34c, 26c and 31c, respectively, using a 10% discount rate and at our long-term nickel price of US$22,355/t (and assuming 80:20 debt:equity funding). Updating these valuations to reflect Amur’s prevailing share price (vs c 5p previously) modifies these estimates to 39c, 51c, 41c and 50c, respectively, as shown below:

An analysis of the major component parts contributing to the change in our valuations of the separate development options is as follows:

Once again, the low-grade matte option prevails as the most efficient deployment of capital, although investors should note that this could change if the resource and mine plan are materially reconfigured as a result of the advancement of high-grade production from underground (which seems increasingly likely).

As previously, the above valuations were conducted at Edison’s long-term metals price forecasts of US$10.14/lb Ni, US$2.75/lb Cu, US$13.52/lb Co, US$1,123/oz Pt and US$768/oz Pd and are sensitive to them to the following extent:

Amur’s 2016interim results reveal that it had a net cash position of US$11.5m as at 30 June 2016, after US$3.8m of cash outflows before financing activities – ie a cash burn rate of US$0.6m per month. Cash outflow is forecast to reduce in H216 in the absence of any material capital investments in property, plant or equipment, such that Edison forecasts a year-end cash position of US$9.3m as at 31 December 2016.