Ø Harvest Minerals - Positive initial 105.3Mt @ 21.35% KCl inferred resource including 62Mt of sylvinite at 25% KCl for Sergi potash project in Brazil. We estimate pre-tax NPV8% of US$350-US$408m.
24 July 2015
Harvest Minerals (HMI AU, A$0.022, Mkt. Cap A$9.7m) – Initial resource for Sergi potash project in Brazil
Key Points:· JORC inferred mineral resource at Sergi of 105.3Mt a 21.3% KCl· Includes 62.0 Mt of sylvinite at 25.0% KCl· Exploration target of a further 28-32Mt at 22-28% KCl· Planned drill programme to upgrade resource to indicated by end Q1 CY2016 and progress BFS· Plan to drill second potash project at Capela and complete test work on phosphate project at Arapua in coming months· We estimate pre-tax NPV8% at Sergi of US$350-US$408m depending on development route
Harvest is a fertilizer development company listed on the ASX and currently in the process of listing on AIM. The company has several projects in Brazil including a phosphate project which could provide potential near-term cash flow whilst developing its two potash projects in the Sergipe Alagoas Basin, close to the only producing potash mine in Brazil, Vale's Taquari-Vassouras.
Brazil largest importer of fertilizersAgriculture is the key growth industry in Brazil, but whilst the country has abundant sun, water and land, it lacks the fertilizers required to maximise yields. Brazil is currently the largest importer of fertilizers (70% usage, India 43%, USA 36% and China 7%). The Government wants the country to be self-sufficient in fertilizers by 2020, but as fertilizer use continues to increase (+12% between 2012-2014) so does the level of imports (+23%), whereas domestic production continues to fall (-9%). Brazil currently imports over 90% of its potash and 51% of its phosphate and the country’s only potash mine, Vale’s Taquari-Vassouras, is due to be depleted by 2018.
Initial resourceAgainst this background there has been considerable interest in Harvest Minerals projects, particularly given that the potash projects are between 13km and 48km from Taquari-Vassouras which suggests there could be significant capex savings if agreement could be reached with Vale for access to the plant. Harvest acquired Sergi in April but from 2011 to 2014, four holes were drilled between 1.6-2.7km apart. Of these only one intercepted mineralised formations of sylvinite and carnallitite but, combined with later 3D seismic work, Coffey Minerals was able to estimate an inferred resource of 62Mt sylvinite at 25% KCl and 43.4Mt carnallitite at 15.9% KCl for a total 105.4Mt at 21.3% KCl. Coffey also identified a further 28-32Mt at 22-28% KCl exploration target. The sylvinite and carnallitite both occur in two layers. The sylvinite layers occur at 1,190m (8.27m thick) and 1,221m (8.46m thick) depth and the carnallitite layers at 1,299m (7.25m thick) and 1,314m (5.5m thick) respectively.
Our own conceptual estimates indicate pre-tax NPV8% of between US$350 and US$408mWhen Harvest went into a trading halt ahead of the announcement, the company stated it expected to release a “Reasonable Prospects for Extraction” report allowed to publish that report. The company announced today that although the report had been completed under ASX rules (Guidance Note 31), unlike under AIM regulations, inferred resources could not be used to support a production target and financial projections. However, they went on to state that the results of the report were such that they would pursue further exploration and development of the project “with vigour”.
As a result, we have estimated our own conceptual cost estimates on the viability of the project. Under section 3 of the JORC statement released yesterday, it was proposed to use room and pillar mining to extract the ore. This is the same mining method used at Taquari-Vassouras, which has a plant capacity to produce up to 625kt of K60 potash per year. We used this as the basis of our estimates under several different scenarios.
1. Harvest-owned mine (1.4km shaft and room-and-pillar mining) and processing plant
2. Harvest-owned mine and infrastructure including 35km conveyor belt to toll treat ore at Taquari-Vassouras from H2 2018 on a cost-plus-20% basis.
For the mine and infrastructure costs we used comparable costs from other projects we have worked, plus guidance from industry. For the processing plant we based our costs on several recent technical reports on potash projects and, in particular, of Highfield Resources recent scoping study on Sierra del Perdón which was based on processing 3.15Mtpa of 50/50 carnallite/sylvinite to produce 560Ktpa of K60 product. We also used the Brazilian potash prices forecasts in that report from independent fertilizer consultants Integer Research.
As the resource is inferred and the assumed mining method was room and pillar we assumed only 40% of the current resource is mineable and ignored the upside from the exploration target.
We also adopted the scenarios using just sylvinite and a carnallitite/sylvinite blend. Using carnallitite/ sylvinite increases the estimated mineable resource from 24.8Mt to 42.12Mt and mine life from 9 to 15.5 years, but drops the average grade from 15.8% K2O to 13.42%K2O, slightly decreases recoveries and increases the capex and opex as a carnallite decomposition circuit would need to be added to remove the magnesium.
Due to the distance from Sergi to Taquari-Vassouras, our provisional capex suggest there is little difference between building a mine and new plant (US$450m) and building a mine and 35km conveyor to Taquari-Vassouras (US$425m). However, given the availability of water at site, we believe the infrastructure capex could be substantially reduced by using a slurry pipeline to Taquari-Vassouras.
Based on these assumptions, our pre-tax estimates for processing the sylvinite only were NPV8% US$385m and 28% IRR for own mine and plant compared with NPV8% US$350m and 27% IRR for toll treating. For blending the ore we estimated an NPV8% of US$454.5m and 24% IRR for own mine and plant and NPV8% US$405.6m with a 23% IRR for toll treatment.
We also looked at the impact of increasing the size of the mineable resource, given the current exploration target and the potential for using alternative mining methods. For a rough guide we doubled the mineable resource to 84.2Mt and, assuming using own mine and plant, estimated a pre-tax NPV8% of US$750m and IRR 31% for processing a total of 49.6Mt of sylvinite and NPV8% US$687.5m and IRR 25% for processing 84.2Mt of blended ore.
This obviously has a large impact improving both the IRR and the NPV. However, it is worth noting that anything over ~20 years had little impact and that by blending ore and lowering grade the NPV and IRR of processing a sylvinite and carnallitite blend was lower than just processing sylvinite despite the longer mine life.
Sergi development not dependent on ValeThese estimates are obviously conceptual in nature and we will have a better idea once the company upgrades the resource and releases its own numbers. However, based on these estimates the key inference is that the company is not dependent on Vale to develop the project. If a resource can be generated for Capela, we believe the financial incentives to reach an arrangement with Vale will be more compelling as it is only 13km away from Taquari-Vassouras.
Next StepsWe view the initial resource as extremely positive and paves the way for further work both at Sergi and at Capela as well as completing the test work at Arapua on the direct application natural fertilizer (DANF). The plan is now once the dual listing is completed over the summer, to upgrade the resource and start the BFS at Sergi whilst carrying out initial drilling and a scoping study at Capela. If mineralisation is intersected at Capela, due to it being far shallower (at 250-330m depth) and closer to Taquari-Vassouras, it may take priority over Sergi.
HMI Price at posting:
2.2¢ Sentiment: Buy Disclosure: Held
