(England, Wales and Northern Ireland), of QMM's Social and Environmental Impact Assessment (SEIA) for the
Fort Dauphin titanium project.
Summary The following review is based on a thorough analysis of the Social and Environmental Impact Assessment (SEIA), recently published by QMM on its Madagascar Ilmenite project. It examines the criteria on which the SEIA is based and the nature of mitigating, conservation and habitat restoration measures. It also scrutinises the processes which have been used to evaluate data, and compares the company's own record, and project proposals, with similar mining ventures elsewhere. Our overwhelming conclusion is that there are numerous defects in the methodology of the SEIA. These include: absence of critical data, failure to undertake studies of some potentially very negative potential impacts, and reliance on highly speculative assertions, which are not backed by firm evidence. We also challenge the company's contention that it will consistently follow Abest practice@ or observe the highest industry standards in performance of the project. Specifically, the SEIA neglects to examine worst case scenarios; what the actual returns to the government and people of Madagascar of income generated by the mine, or how they will be distributed; and whether tourism can co-exist with such a large extractive venture. It severely underestimates the likely impacts of the dredging project on bio-diversity and forest loss; of the port construction and operation on marine life; of the proposed weir on both human and saltwater species; and of the new roads on local people. The SEIA also fails to properly evaluate the social and health impacts of an in-migrating workforce on residents in the Mandena-Fort Dauphin area. One of our most serious judgments is that the Astake-holder' process, in which the company has placed such confidence, was not properly conducted or the views of local communities reflected in the report. Viable plans have not been set out, to mitigate the large number of serious cultural and economic losses which they will have to make. We also cast considerable doubt on the economic security which the project will supposed provide to the Malagasy government and people. Not only is the Mandena mining project dependent on external funding for the new port, but many other aspects appear to be contingent on new public funding which will not derive from the company. Because of these multiple deficiencies, in our view the project would not be compatible with true sustainable development in southeast Madagascar, or for the country as a whole. We therefore urge the Malagasy government not to grant a permit.
1) Critique of the methodology of the SEIA At first sight the four volumes of the SEIA (Social and Environmental Impact Assessment) seem admirably comprehensive. However, they are decidedly not. We can compare them, for example, with the seven volume SEIA which TIOMIN recently presented to the Kenyan government in support of its dry sands heavy minerals project Yet TIOMIN's studies have satisfied neither the government (which this month set new conditions on the project) nor a Kenyan court, which has imposed an injunction on the project. Overall, the SEIA information is very badly organised, confusing (there is no consecutive page numbering in the weighty Appendix - indeed some pages are not numbered at all) and repetitive. There is also rarely any reference to the supporting study or author on which an assertion relies. No data whatever is provided on the management structure of QMM, including company response mechanisms to worst case failure or accidents, as required under regulations in the company's country of incorporation [see Canadian Environmental Assessment Agency Comments on the Voisey's Bay EIS, March 31 1998],
Although brief references are made throughout to comparable mineral sands operations elsewhere, these are generally not identified - including those managed by QMM's own head company Rio Tinto (through QIT Quebec), and its 50%-owned Richard's Bay Minerals (RBM) in South Africa. This makes it impossible - on the evidence offered in the SEIA - to test QMM's contention that its mine management and "restoration" proposals reflect good practices elsewhere.
The SEIA's authors evaluate the numerous residual impacts of the project according to "intensity", "reach", "duration", and "importance" [SEIA Appendix 1-50] - definitions with which we have little quarrel. Anticipated negative impacts are to be addressed by two types of measures: "Mitigation" and "Development" The first are intended either to eliminate, or decrease the intensity of, the identified impacts "in order that effects are socially and environmentally acceptable". However, no procedure is proposed to "measure these measures". Indeed, it is quite clear that many of the impacts have not been, or cannot be adequately anticipated, at this stage. Some of them - such as loss of bio-diversity and brackish waters of the inland waterways to the weir - may well be irreversible. The SEIA then addresses the residual impacts, "based on what remains following application of [mitigation measures]", as "Zero", "Low", "Medium" and "High" . In the latter case the company concedes that impacts of "a major importance" will remain and "no mitigation measures [are] applicable" (our italics). In these instances "compensation" will be made. But compensation is precisely that: an admission that actions to mitigate damage have failed. It is quite wrong for QMM to pass off compensation - whether in the form of money, services, goods or property - as a way of making unacceptable social or environmental impacts acceptable.
Moreover, there are several instances where the SEIA clearly underestimates - and therefore wrongly defines - the nature of the impacts: for example that of the weir on both fauna and human livelihoods. QMM commissioned a large number of individuals and organisations to work on the SEIA. While their professionalism is not called into question. it is nonetheless disturbing that many of them have been closely associated with QMM or Rio Tinto elsewhere; a few - for example Kew Gardens and MERN - continue to be partly dependent on funding from the companies. It is obvious that future expert investigations of the fauna and flora in the project area - in particular endemic species which may still be discovered [SEIA Vol 1 3-167/68]) - have become significantly dependent on the largesse of QMM or Rio Tinto. We may legitimately ask whether such dependency may blunt the critical objective edge which professionals should bring to the project? 2) Defectiveness of data The research has major deficiencies, some of which get passed over by the use of euphemism. For example, we are told that the sampling of benthic fauna was "less significant" at Lake Andriambe than in the Mandena estuarine environment "which may explain the fewer numbers of species recorded" [vol 1 3-167]. Presumably the truth is that the sampling was not as thorough as it should have been? No work is recorded as having been undertaken on the impact of the mine on marine turtles. No data is provided to support the contention that reforestation will increase carbon sequestration. There is no study of the impact of noise pollution on fauna in the conservation zone [SEIA Vol 11 5-31] . There are instances where information which should have offered in the SEIA will only come at a later, unspecified date. We are promised that collected data on aquatic ecosystems "will constitute" a comparison base for a monitoring programme which "will allow for evaluating the changes likely to occur as a result of the projectY" [SEIA vol 1 136] . Details of the progamme are not given. We also informed that "No provision was made in this study for analytical sampling of nutrient elements soon after heavy rainfall" - once again, this is consigned to future monitoring [SEIA vol 1 p. 139]. One amphibian study - by the University of Michigan - was not even completed by the time the SEIA was published. Some assertions are highly speculative, such as: * That pollutants from transport "can be easily recovered and eliminated according to a code of good practice". [Vol 11 5-83] - but there is no definition of "good practice"'
* "Unless a permeable fault passes under the site, the bedrock permeability is expected to be low enough that the presence of groundwater will be limited" (our italics) [vol 1` 3-29]/. Why was no fault testing carried out? At least one statement is dangerously deceptive. We are informed that "There is no barrier or coral reef in the ocean off of (sic) the primary study zone (Mandena) " [SEIA Vol 1 3-175]. Yet, a few pages later, we are told that coral colonies, encrusting, tabular and branching" do exist off the Evatraha peninsula . Although these are not reefs as such, they could well serve as important refugia in the case of continued global warming and the loss of coral diversity further north We believe that port construction and operations are bound to increase marine turbidity, destroying some local coral through direct impacts and sedimentation. 3) Should a permit be granted before the final feasibility study is done? The SEIA is being presented to the government of Madagascar before completion of the final economic feasibility study. While this is not exceptional practice in the mining industry, its appropriateness here must be strongly questioned. QMM is relying heavily on profits from the mine, over a long period of time, to finance the conservation and restoration programme which it claims to be integral to its project. But, as Conservation International points out, at no point does QMM state what financial responsibility it will assume for the conservation zones and rehabilitation forest, or who will assume management of these programmes. This is a "significant omission" [CI Review page 27] . QMM also states that "the financing of the port must be resolved before a decision can be taken to proceed with the mining project" and that it will bear only part of this cost [SEIA Summary p31]. The company therefore intends to "propose a feasibility study that will guarantee the project's financial profitability, while ensuring external financing for the port." [SEIA Vol 1 4-40]. There are thus critical uncertainties over external funding for the port; over the nature of linkage between this funding and the project's viability; and about how finance for the conservation and post-mine rehabilitation programme is to be generated. We strongly advise that, at the very least, no permit be granted to the project until the final feasibility study is performed, addressing these matters. This must be accompanied by a thorough social and environmental assessment of the port and approach road proposals. The SEIA takes a particularly pessimistic view of the capacity of the local people in the Mandena region, backed by the administration, to limit or halt current over-exploitation of forests and their products. Ironically, while claiming that the capacities and good sense of the Antanosy are vital to the success of the QMM community development programme, the company ignores the capacity of Malagasy people to manage this on their own. QMM"s attitude is reminiscent of its parent company, Rio Tinto when, in the 1980's, it boasted it was teaching Namibian farmers in its Rossing concession, to plant their own gardens - although this was an activity they had been engaged in for generations.
4) Failure to examine worst case scenarios In contrast, while deploring the degradation of the environment by human pressures, charcoal burning, and general over-exploitation of forest resources, the SEIA fails almost completely to address "worst case scenarios" resulting from its own operations. These include: road transport accidents causing spills of low-radioactive materials; a vessel collision or capsize in the port, resulting in discharges of large amounts of mineral product; leakage or incendiarisation of fuel oil; excessive flooding in the mining zone; droughts causing exposure of low-radioactive minerals dumped back into the dredged areas. None of these possibilities is unthinkable: indeed, they have occurred at several major mines in recent years - including those operated by Rio Tinto. For example, more than six hundred chemical drums were washed by flash floods into the local water supply at Kelian, east Kalimantan a few weeks after opening of the eponymous gold mine. Dry toxic "islands" have appeared in the tailings pond at the same mine It is also worth noting that the prospect of product spills at the ports in two of Rio Tinto's projects (the Pasminco base metals project - now sold - and the Kalimantan coal mine) had to be assessed in the EIA's prepared for both these projects. Even if there were no such troubling precedents, we would urge that no permit be granted for the project until QMM has posted a comprehensive environmental performance bond, in an amount calculated by an independent assessor, to cover all worst case scenarios and possible operational failures. 5) Questions over Petriky and St Luce Rio Tinto has clearly stated it wants to maintain its global lead in the mining and production of titanium materials. It was the need to provide feedstock for its QIT smelter in Quebec which, during the 1990's, motivated the company to apply for permission to mine the dunes and wetlands of St Lucia in Natal. This particularly ill-advised project was rejected by a judicial commission and later by virtually all the South African cabinet (President Nelson Mandela himself signed a national petition against it). Instead St Lucia is now being developed around agriculture and eco-tourism, as farmers return to the area following the depredations of apartheid. There is the prospect that it will be declared a World Heritage site from which mining would be permanently banned. The QMM SEIA covers mining only in the Mandena sector. Several references are given to future exploitation of Petriky and St-Luce, but little assessment is made
of possible impacts in these zones: these SEIAs are promised at a much later date. Yet QMM makes it clear that, unless it eventually proceeds with exploitation of these sectors, the financial viability of the entire project may be at risk. In theory, granting consent for mining in Mandena will not carry with it any implication that QMM will automatically gain a permit to mine at Petriky and St. Luce. In practice, this may be extremely difficult for a future government of Madagscar to resist.
6) Questioning the standards followed by QMM/Rio Tinto QMM says it will observe Rio Tinto's in-house standards and, in particular, the parent company's Statement of Business Practice ("The Way We Work" 1998) [SEIA Summary page 11] . It also promises to follow the ICC (International Chamber of Commerce) Business Charter and World Bank guidelines on mining best practice. (Although several publications by the World Bank are cited in the SEIA's bibliography there is no reference to criticisms made by the Bank of the first phase project proposal). The ICC is a purely business organisation and its Charter is vague, rhetorical and purely voluntary. World Bank standards have also been criticised by many environmental and social observers as inadequate and flawed. The Bank's recently proposed revisions to its policy on Indigenous Peoples are considerably weaker than the existing policy. Rio Tinto was one of the first companies to sign onto the UN Global Compact for multinational companies. However, this year it was indicted by Corporate Watch-US for failure to observe both human rights and environmental safeguards at mines in Indonesia and Brazil, as laid down under this Compact. In addition, the mining ombudsman for Community Aid Abroad (Oxfam Australia) has condemned Rio Tinto for human rights violations (specifically regarding women) at its Kelian gold mine in east Kalimantan. In relation to specific standards for mineral sands mining, the company certainly cannot argue that its project, as presented, conforms to "best possible practice" around the world. The South African government's rejection of Rio Tinto's St Lucia project - also a wet dredge dune enterprise - has already been noted. The state government of New South Wales (Australia) has a policy that "Heavy mineral sands mining inYlittoral rainforests on the coast will be prohibited" (our italics). Mineral sands beach mining in general is prohibited, unless a company can "clearly [demonstrate] that it is environmentally acceptable [such as] where a dune area requires stabilisation or re-vegetation to protect the coastal environment" As long ago as 1981, Australia's leading conservation, the Australian Conservation Foundation (ACF), declared there should be "no further consents for mining for mineral sands in natural areas along Australia's east coast as distinct from inland reefs of rock rutile and ilmenite, because it has a "severe and permanent environmental impact" 7) The rights and welfare of the peoples of southeast Madagascar
The SEIA ignores the implications of the ILO Convention 169 on the rights of Indigenous Peoples, and the UN Draft Declaration on the Rights of Indigenous People. The latter stresses the necessity of "fully informed prior consent" to, and the right of veto over, natural resource projects. In contrast, Rio Tinto's "The Way We Work" merely states that "Yeconomic, technical. environmental and social factors are coordinated in an integrated process. In all cases this involves full consultations with local people Ywith the intention of securing the widest possible agreement and support for any activity proposed" ["The Way We Work" Rio Tinto 1998, pge 10]. This falls considerably short of the precepts soon likely to be adopted by the UN Sub-commission on Human Rights. So too do the widely-vaunted "stakeholder" dialogues which QMM conducted during the SEIA process - and which it promises to continue for the life of the mine. While a huge number of concerns and fears were raised by villagers at numerous meetings (and these are amply recorded in the Appendix), many of them have simply not been answered. (Or if they have, the answers are not recorded.) In particular (according to Conservation International's observations) QMM "failed to reach the poorest and most vulnerable local stakeholders" [CI Review p 28] and to inform holders of "usufruct rights Y[and] to full compensation for any lossesY" The engagement process purported to be an exchange of information between all parties. In fact, it seems to have corresponded to a previous exercise conducted by Rio Tinto in Subanen territory in the southern Philippines, during the late 1990's. Here, communities were continually told of company plans and invited to modify them, rather than permitted to reject mining per se, or empowered to present their own development plans. The process was dubbed "dialoguing to death" by one local leader. Rio Tinto was finally forced to acknowledge the opposition of a majority of the Subanen and withdraw from the region. It is somewhat disquieting to find that one of the two chief architects of QMM"s "stakeholder" dialogues in southeast Madagascar was the same anthropologist who had led the equivalent process in the Philippines It is clear from the SEIA summaries of these meetings that the experts and company representatives took charge of the discussions, rather than local leaders. Moreover, several anthropologists and sociologists who have worked with the Antanosy and Antandroy peoples over many years were not involved. (We are not told if they were invited and refused). QMM promises to continue the stakeholder dialogues through the life of the project. How will these be conducted? Will they be independently monitored? If villagers want the project - or any of its components - drastically modified or halted, what process will be employed to address their demands? Indeed, will they have any substantive rights in this regard at all? Or will they simply be consigned to a culturally alien "compensation" process which, so far, the SEIA has not even defined? 8) Regional Development plan There can surely be few people, however sceptical of the value of the proposed mine, who do not applaud the intentions behind the Regional Development Plan, and QMM's stated adherence to it. However, the SEIA treats this critical new development as almost wholly contingent on the mine proceeding.
At one point in the SEIA the key admission is made that: "The question of whether the Anosy region can adequately accommodate the project is a critical issue, one that affects several elements of the QMM project, including the building of infrastructure, managing conservation and rehabilitation, lodging and health services Y" as well as new pressures on sanitation, education and other provisions [SEIA Vol 1 2-64]. But QMM is asserting that its role in the provision of new services is essential in overcoming the current inadequacies. This is highly disingenuous. For the biggest chunk of new services is not to be funded by the QMM but has still to be found from public or multilateral sources. It is a serious defect that the components of the regional development plan have not been assessed independently of the project. QMM appears to place itself both as a keen proponent of Regional Development (supporting it whether or not the mine goes ahead) and as a primary architect of the development plan. At the same time, it is an admitted benefactor of that plan - at least in relation to the siting and financing of the port. It is difficult to resist the conclusion that QMM is trying to be all things to all people. We believe that, if QMM were refused a permit to proceed with mining, it would not remain for long in Madagascar (perhaps first selling-on its interest, as Rio Tinto has done on several occasions in the past). The relationship between the mine and sustainable development in southeast Madagascar goes to the heart of the vital question as to what should come first: a clear and carefully evaluated plan for development, in which mining is considered only one possible (albeit important) factor. Or acceptance of the mine now, in the hope that promises made by the company will be fulfilled over a lengthy period of time. In view of the absence of numerous pieces of data in the SEIA, accompanied by the long term uncertainty over the project's financial viability, we do not believe this second option to be a viable or prudent one. 9) Bio-diversity questions There is huge optimism in the SEIA about the project management's ability to maintain and encourage bio-diversity, which would be dependent largely on the three proposed conservation zones. [see, for example, SEIA Vol 11 5-70]. In fact, the zones now identified are considerably smaller than those originally proposed by Lewis Environmental Services in 1992. QMM's optimism depends on unjustified and unrealistic assessments of the benign nature of mining. As pointed out by Conservation International, the SEIA "does not deal adequately with increased pressures on faunal populations [from mining] Yboth within the littoral forest zone and Yelsewhere in the region." [CI review page 23] .
This failure is compounded by the fact that in, certain key instances, QMM admits that it does not even know of the presence and whereabouts of assumed endemic and possibly threatened species. QMM makes the unwarranted assumption that conservation is maintained, so long as species per se are not lost. For example: "No rare or endemic species are to be found in the swamps", hence "there is no specific issue in terms of the conservation and protection of the plant life and wildlife of the environment" [SEIA Vol 11 5-70]. This is a non-sequitur, predicated on an unacceptably narrow definition of conservation as widely understood. 10) Problems of the weir The proposed weir is a core component of the project. But it will effectively transform a major saline water source into a freshwater one, with (in our opinion) irreversible results.
The SEIA acknowledges that this structure will "alter the quality of the [existing ] landscape", sacrificing some of the wilderness characteristics and aesthetic appeal, producing an "impact of major importance" [SEIAVol 11 5-99] - especially to mangrove plantations [SEIA vol 11 5-75] with loss of seawater shrimps, crabs and eels [SEIA vol 11 5-79] and the "modification" of "the entire benthic community located upstream from the structure" . QMM also concedes that the weir "could have a negative impact of medium importance on the tourism operators" in the area [SEIA Vol 11 5-108], resulting in a reduction in tourism. The only compensation offered is a vague prospect that "Tourist activities related to the Mandena ecological research centreYcould possibly be developed" [SEIA ibid ] (our italics).
However, eco-tourism is intrinsically dependent on the very "peacefulness and visual appeal" which the company agrees will largely disappear. In this respect we again point out that the blithe hope that tourism could co-exist or "hitch" onto mining, was dismissed in the context of RBM's failed application for its St Lucia mineral sands project in South Africa. QMM identifies as an "impact of medium importance" [Vol 11 5-107] obstacles posed by the weir in the transport of local goods between Lake Lanirano and Fort Dauphin and Evatrala village . The company does not, however, address the changes which will result from construction of the weir, except to promise (unspecified) new installations by way of mitigation. These constructions are to be backed by a "mediation procedure" in the face of anticipated resistance by local traders. QMM fails to identify what this procedure will be. This is quite unsatisfactory.
11) Losses of littoral forests The mining of the Mandena sector will, according to QMM's own data, destroy 203 hectares of littoral forest. The company also concedes that "forest fragments may be required to be removed sooner than they would otherwise have been cut down by villagers under the no-mining scenario" [Summary page 21]. Extraordinarily, although the company maintains that it has factored in several key improvements to the original project plan, it now plans to leave behind far fewer fragments of rehabilitated land than was envisaged under the earlier EIA of Lewis Consultants The company's calculations may also go awry should "significant migration to the area occur without proper planning" [Summary page 10].
QMM has said it will "preserve" 160 ha. as a conservation area and, after mining operations, replant another 212 ha. (about 10% of the total 2,120 hectares in the concession [SEIA vol 11 5-71, 5-74]. Nonetheless, it confesses that "it isYdifficult to accurately estimate what forest area will remain in Mandena once the mining operations are completed", although "historical trends lead us to believe that the residual forest will have disappeared even before the onset of dredging operations". We found this crucial section of the SEIA unacceptably speculative, It hinges on an unsubstantiated conjecture - indeed the false syllogism - that, without the mine's conservation components, all the littoral forest will disappear over the next 20 years [SEIA vol 11 5-71]. What is certain is that the mine will destroy a substantial part of the albeit already-degraded forest. As Conservation International points out, it is possible that "the mining operation [could cause] a net loss of access to the services and goods on which communities are dependent [from the rainforest]" Crucial resources of mahampy could be jeopardised, both in quality and quantity [CI Review page 25] We support Conservation International's contention that "The SEIA's straight line projection of loss of littoral forests, based on data on forest loss over a 50-year period, does not represent an objective analysis of the available data" and that this "Ymay be overly pessimistic in regard to the present degree of deforestation" [CI Review page 24]. " No management plan has yet emerged for the conservation areas. In particular, no details are provided as to which parts will be excluded to local people and which (if any) may be managed by them. This is a serious reflection on the company's oft-recited claim that it will involve all stakeholders in building a sustainable future for the areas that their operations will impact. 12) Burial sites QMMs cavalier attitude to the observance of fundamental indigenous beliefs is evidenced by vague and patronising language used in the SEIA to quell fears that burial sites and their zones of influence will not be unduly disturbed. Extraction "could" raise local fears, we are told, but their concerns "could" be minimised, by "allowing villagers to participate in establishing and monitoring protection measuresYas well as ensuring their access to the sites at all times". Nonetheless it is conceded that villagers " will remain worried and consider that the sense of the place has been affected". No indication is given as to what these protective measures will be, nor how local people will be involved in them. Once again QMM downplays the issue as of "medium importance" [vol 11 5-99], although it is surely one of major cultural significance. Our evidence is that presently unidentified graves will continue to be found after construction and mining operations begin. The SEIA also passes over the fact that the swamps (which cover a significant 17% of the Mandena mining sector [SEIA Vol 1 3-118]) are used by the Antanosy for the burial of young children and foetuses [Vol 1 3-127], and makes no attempt to address the consequences of disturbing the swamps.
13) Is eco-tourism compatible with mining? QMM/Rio Tinto regards tourism both as a means of consolidating the innovations in conservation and compensating for the medium or long term loss of resources to local people. But the company also seems to present tourism as a means of "selling" the less palatable parts of the project to a sceptical public, while the supposed "synergies" between mining and tourism are developed. This tactic will undoubtedly perturb many people concerned to promote culturally responsible, "eco-tourism" under indigenous control. However, these groups (such as Tourism Concern) have not been consulted. As Conservation International states, the SEIA is generally lacking in information on the negative impacts of mining on tour operations; for example dismissing the aesthetic downside of the port and weir and (we would add) impacts of the noise, dust and heavy traffic inevitably associated with this huge mine. The assessment also fails to address one of the core socio-cultural elements in Antanosy society about which QMM claims to be highly concerned; namely the disturbance - indeed violation - of the "sense of place". This is at risk both from mining and a significant increase in package tourism, such as the company proposes [SEIA Summary page 35] It is clear to us that eco-tourism cannot sustainably co-exist with the mining project. Nor is tourism value-neutral in itself. Some of the prospects for the industry mooted in the SEIA stretch credibility too far. For example, the assertion is made that swimming, scuba diving and whale watching would pose no "usage conflict" with fishing [SIEA vol 1 3-200]. But the SEIA also acknowledges that fishing is more important now in the Mandena zone than it was ten years ago [SEIA vol 1 3, 178] and that tour operators have already contributed to the erosion of river banks [vol 1 3 57] 14) Economic benefits? QMM has had more than enough time to thoroughly evaluate the project's economic benefits to local communities, Fort Dauphin, the South East region and the Malagasy exchequer, and it has not done so. No income statements are provided in the SEIA; no figures are given for government financial risks and responsibilities regarding the project; and there are no binding guarantees that a fixed percentage of returns to the government will be made to the region, whether to local administration or community organisations. (The disparity between royalty shares to central government as opposed to provincial government and local organisations has been at the root of considerable strife in Indonesia. This has prompted it to shift the balance in its recently-announced new Mining Bill).
QMM recites the familiar mantra that the project will necessitate construction of infrastructure and bring economic benefits which are bound to benefit the communities in the region. We are informed that "the construction and maintenance of lodgings for a pool of foreign workers and those from outside the Fort Dauphin region", carrying with them significant purchasing power, "should benefit the Fort-Dauphin land and real estate markets". [SEIAVol 11 5-103] QMM is supposedly committed to paying for the construction of new housing units and the provision of new public service infrastructure [SEIA Vol 11 5-105] This may be so, but it is also likely that there will be considerable conflict between the Antanosy and the outsiders. Even further pressures will be exerted on at-risk habitat in the area, along with inflation in real-estate prices and costs of essentials. No large-scale mine project has managed to avoid such outcomes, though some have tried to mitigate them. The SEIA fails almost completely to address these serious socio-economic consequences The SEIA also tells us that |Lake Ambavarano "should present an economic benefit for the nearby communities" due to the introduction of new freshwater, halieutic, species. There is no data to back this assumption and it ignores the likely impacts of the weir, already mentioned The promise of new economic activities and promotion of local business is severely qualified in the SEIA by the statement that "..it is impossible to assign a value to the expected results of these efforts at this time" Instead of quantifying any gains - sector by sector - QMM will rely on the formation of local associations, new support activities, and employment generated both by its extractive and conservation measures, somehow ending up with a ratio of 1 new job (in the project) to 1.7 other jobs in other sectors. [SEIA Vol 11 5-104]. Once more, it is unclear how the company could arrive at an optimistic prognosis, based on little or no evidence. Compounding this inadequacy is the fact that QMM does not quantify the number of people who would be removed if the project goes ahead under its present design (potentially hundreds of families), or those who would be hampered in their current economic activities by various aspects of the project. Nor does the SEIA specify what compensation would have to be found for them. In this respect it is valid to note that Rio Tinto has long been engaged in acrimonious debate with customary land users at its Kelian gold project. While many of the compensation issues may have finally been resolved, disparities in payments between individuals have compounded social conflict. Even after more than ten years, land rights issues remain largely unresolved. 15) Alternative mining Sapphire and gold mining is mentioned only in passing in the SEIA - and in a rather negative fashion [SEIA Summary, page 34]. No consideration is given to promotion of environmentally sound artisanal mining as an economic generator for the extractive sector, rather the mineral sands project. In fact, as your government will be aware, recent studies (by Dr Thomas Siepelmeyer, with whom we are in contact) have shown that gem mining could become a major income and employment generator, precisely for communities in the south east of the country. As we understand it, the government's 1998 mining policy aims to support small scale mining and to attract private investment for it. It is puzzling that potential growth of this sector is neglected by the SEIA.
16) New roads The company seems to attempt a precarious balance between restricting the construction of new roads, with their admitted negative impacts (dust, noise, accidents. pollution, land disturbance) and promoting a proposed new road between Fort Dauphin and Ehola, as an integral positive component of the regional development plan [SEIA Vol 1 4-91] . It is strongly questionable whether a new two-lane highway will make an overwhelmingly positive contribution to the local and regional economy. In any event, the SEIA omits to discuss the impact on existing forests, of the proposed widening of the existing roads. We have already pointed out that the SEIA completely neglects the impacts of road accident. It also only evaluates everyday risks to road users and residents as "minor". But, over the possible 60 year life of the mine, the Mandena-Ehola road will pose major health problems, and the hazards and impacts are bound to be magnified should mining proceed to Petriky or St.Luce. 17) Health The SEIA foresees a "potential problem" for community health, with the influx of a large number of construction workers, followed by in-migrating mineworkers during the life of the project [SEIA Vol 11 5-108]. In mitigation, QMM merely promises to "take all the necessary and possible steps to minimise the risks associated with the spread of sexually transmitted diseases among the local population" [SEIA vol 11 ibid], invoking an "awareness and education programme" for the purpose [SEIA vol 11 5-82]. What is not addressed at all is the specific risk to women's safety, as well as health, occasioned by the arrival of large contingents of males, especially from outside the main residential socio-cultural group. Rio Tinto, parent of QMM, was recently indicted by the Mining Ombudsman for Community Aid Abroad (Ozfam Australia) for allowing staff at its Kelian gold mine in Indonesia to molest, and even rape, local Indigenous women, - and covering up the evidence.. It is worth observing that the prospect of these, and other less traumatic, impacts, have led several major mining companies in recent years to adopt the "fly- in, fly-out" system for workers outside the region, thus avoiding the construction of all-male camps, with their unplanned but inevitable adjuncts of bars and brothels. Nowhere does the SEIA consider this as an option. The assumption that the impacts of workforces will have a "minor residual impact" [vol 11 5-82] is, therefore, in our view, wholly unwarranted. 18) Exposure to Radiation
The SEIA promises the company will observe standards laid down by the ICRP in the mining, handling and disposal of the radioactive minerals associated with the ilmenite deposit in Mandena. Rio Tinto's health and occupational standards at most of its mines are recognised to be high, in comparison with many other mining companies. However, ICRP standards are not the highest possible. For example, the maximum radiation dose for a worker is set by the Commission at 50mSv a year. In fact, the current radiation limit, stipulated by the Australian Commonwealth Radiation Protection Code, and applied to mineral sands operations (under the Australian Code of Practice for Mining and Milling of Radioactive Ores) is 20mSv a year occupational exposure. There is no reason at this stage to doubt the company's assertion that actual exposure in all workplaces will not exceed 20mSv (a maximum of 5 mSv is envisaged) [Vol 11 5-87]. But, once again, worst case scenarios, particularly in the handling or transport of monazite, have not been adequately addressed by the SEIA.
19) Rehabilitation, Restoration or "dumping"? Soil, vegetation, water, fauna and flora in the entire project area all risk being negatively affected in manifold ways. QMM expresses confidence that the area can be completely rehabilitated or "restored". This statement is open to severe doubt. First, for several key components ( some already noted) no remediation time-line is presented, or else the company is still awaiting the results of monitoring before identifying the nature of the problem. More fundamentally, since a large part of the mined-out zone (behind the dredge) will be planted with fast growing species, it is inappropriate to talk of "restoration". There is little discussion of the viability of using stored humus and soils for the rehabilitation project, or whether their removal and storage, for up to six months, may drastically alter the soil's substructure and nutrient yield, and its ability to sustain new plantations. A large proportion of these will be fast-growing varieties which have not been used, at least on such a scale, in the region before. RBM (Rio Tinto's mineral sands operation in South Africa) boasts of a good record in restoring mined dune sands. QIT has also received an award of excellence for its contribution to returning bio-diversity to polluted waters in its home province of Quebec. But, claims by RBM to be "restoring" dunes have been challenged widely by a number of South African environmentalists. (One of them, consulted for this review, says the programme is widely known as a "mine dump vegetation experiment"). Whatever the company's record elsewhere, the fact is that south eastern Madagascar presents a significantly different and greater environmental challenge to RBM or inland mineral sands projects in Australia.. In terms of varied habitat, pressures of local population on land and resources, tourism potential, and danger to endemic species, ironically the QMM proposal probably comes closest to the already-vetoed St Lucia project.
20) The port
The SEIA fails to address the major social and environmental implications of both the construction and operation of the port. These include: the quarrying and transportation of at least 1.2 million cubic meters of rock; the laying and maintenance of the access road; migration and temporary settlement of port workforce; likely disruption to fisheries, tourism activities by the dredging of the Anony River; threats to the estuarine fauna and flora caused by releases of ballast water; the removal of existing land-users from the construction site [SEIA Vol 1 4-51].