The Mitigation Game

From Silenced Rivers: The Ecology and Politics of Large Dams by Patrick McCully

Dam builders and operators have been forced over the years to take a number of steps to mitigate the impact of their projects. Some mitigation measures can reduce some of the harmful impacts of a dam, others may be worse than useless. Mitigation is especially dangerous when it misleads the public into believing that dam builders can recreate the characteristics of wild rivers and fisheries and so allows more dams to be built. Mitigation measures generally reduce the amount of electricity and water which can be provided by dams and increases their construction and running costs. Regulations, such as those in the US, which insist that dams include mitigation measures can therefore render many projects uneconomic (especially as the economic viability of most dam projects at present is marginal at best). The effect of mitigation costs on project economics is an illustration of how many, probably most, dams would not be built were they to have to pay the cost of even a small part of the environmental damage they cause.

The most common mitigation measure taken in the US is to release more water from the reservoir than would be the case if the dam were operated only to maximize power or water storage. These 'instream flows' are usually spilled for the benefit of fish downstream but can sometimes be released in large 'flushing flows' intended to wash away harmful accumulations of boulders and gravel. The US Federal Energy Regulatory Commission now requires the operators of many privately owned hydrodams in the US to release instream flows as a condition of renewing their federal dam licenses. The average loss in power generation for relicensed dams having to guarantee instream flows is around eight per cent and in one case was almost a third. The drop in revenues due to lowered power production have forced some dam operators to close down their hydro plants and to give up plans for new projects.

While instream flows can generally be assumed to be beneficial they can also be little more than a palliative. In most countries instream flows are defined according to arbitrary criteria without any ecological basis. In Spain, for example, dams are supposed to release an 'ecological flow' which is ten per cent of the average annual flow — an amount which would in most cases be totally insufficient to retain the ecological characteristics of the regulated rivers. Instream flow requirements usually give little consideration to the importance of natural seasonal flow variations: releases which raise levels during normally dry spells can even do more harm than good. Instream flow requirements also rarely allow for the releases of the occasional exceptionally large flood flows which are an essential part of most fluvial ecosystems. In general, instream flows can mitigate the effects of dams but cannot recreate the essential variability and dynamism of a wild river.

One of the advantages of spilling extra water is that it will tend to increase downstream dissolved oxygen levels. Other measures can also be taken which increase oxygenation such as artificially aerating the water passing through turbines. Increasing dissolved oxygen is generally the cheapest form of mitigation and appears to generally be effective although as with instream flows there are problems in deciding exactly what dissolved oxygen level is the most beneficial and how to trade-off its costs and benefits.

Another form of mitigating the effects of a dam on downstream water quality is to regulate the temperature of releases by fitting the dam with intakes which can withdraw water from different levels of the reservoir. Around a hundred federal dams in the US are able to make so-called 'selective withdrawals'. In 1995, BuRec began work on a 35-storey-high steel selective withdrawal tower in the reservoir behind California's huge Shasta Dam at a projected cost of $80 million. Shasta was built in the 1940s with just one outlet which when the reservoir is low releases water so warm that it is lethal to the few wild salmon remaining downstream. While selective withdrawal can improve thermal conditions below a dam they can rarely replicate the original seasonal variations in river temperatures as at times the reservoir will not have sufficient water at the right temperature.

The Hatchery Debacle

Probably the most controversial form of environmental 'mitigation' is the use of hatcheries for artificially rearing fish whose natural habitat has been destroyed by dams. Since the late 1940s the US government has spent hundreds of millions of dollars on hatcheries to mitigate the impacts of dams on Pacific salmon. The Bonneville Power Authority, which operates most of the big dams on the Colombia, now spends some $350 million annually on 'fish and wildlife investments' — mainly hatcheries. Yet not only has the number of adult salmon plummeted, but hatchery fish are degrading the genetic diversity of the remaining wild salmon and helping push them toward extinction.

The hatchery programme has failed because dams are continuing to destroy salmon habitat, and also because of the inherent limitations of hatcheries. The genetically homogenous hatchery fish mate with their wild relatives and thereby reduce their genetic fitness: the effects on the natural stock include 'decreased survival and stock size, poor stamina and disease resistance, inappropriate territorial and hiding behavior and other poor performance.' The overcrowding in fish farms, furthermore, means they are highly prone to diseases which are then spread to wild populations. A 1995 report by the prestigious US National Research Council warned that current hatchery policies in the Pacific Northwest were 'based on deep ignorance'. 'It isn't enough to focus only on the abundance of salmon', the NRC concluded. 'The long-term survival of salmon depends crucially on a diverse and rich store of genetic variation.' Some fish biologists in the Northwest now believe that all the hatcheries should be shut down.

Despite the expensive failure of hatcheries in the Pacific Northwest and elsewhere in North America, hatcheries are regularly promoted by government fisheries departments and environmental consultants as a means of mitigating the destruction of natural fisheries by dams in other parts of the world. Part of the 'mitigation' for the Pak Mun Dam in Thailand, for example, is the hatchery rearing of some two dozen species of local fish — only around ten per cent of the species found in the undammed river. Mekong fisheries specialist Walter Rainboth of the University of California believes that the hatcheries at Pak Mun are merely 'public relations gimmicks'.

The backers of the Sardar Sarovar Dam claim they will 'mitigate' the loss of the hilsa fishery by stocking the reservoir and ponds in the estuary with hatchery-bred fish. But fisheries scientists have not yet been able to bred hilsa artificially. In fact the rearing of hilsa currently depends on obtaining spawn from wild adults, which would in all probability be eliminated by the desiccation of the river.

Down the River . . .

Aiding smolts on their danger-filled voyage toward the sea is a cornerstone of the authorities' costly but so far largely fruitless plan to restore the Colombia River salmon. One part of this plan is the installation and improvement of screens and bypass systems which prevent young fish from being sucked down turbine intakes. The Army Corps of Engineers is spending $345 million on upgrading fish facilities at its eight dams on the Columbia and lower Snake. The bypass systems, however, do not help the smolts negotiate the warm, predator-infested reservoirs. The preferred technofix for this is 'barging' — in a striking illustration of how far the Columbia has been transformed from a wild into a managed river, smolts are trapped, crammed into barges and motored through reservoirs and dams. While the survival rate of barged smolts is higher than for those left to negotiate the reservoir on their own, mortality due to stress and exposure to diseases in the barges is still high.

Salmon advocates on the Colombia claim that drawing down the reservoirs during the spring and summer migration is the key to helping the fish stocks to recover. The hydropower and navigation interests on the river, however, are strongly resisting pressure to make the dam operators spill water. The drawdowns would certainly not come cheap: the Corps of Engineers estimates that the necessary structural modifications to the eight relevant dams on the Colombia and Snake would cost up to $4.9 billion — and this sum does not include the huge cost to the dam operators of lost revenues from foregone power production and barge fees.

. . . And Back Up Again

While salmon are by far the best known of migratory fish there are many hundreds of other species with very different migration patterns, especially on large floodplain rivers in the tropics. 'Catadromous' fish live most of their lives in rivers but spawn in estuaries or in the sea, the opposite to salmon; 'amphidromous' species spawn and mature in both salt and fresh water; 'potamodromous' fish migrate entirely within freshwaters. Because these fish do not follow the classic anadromous migratory pattern and have mostly been very little studied they are sometimes not even regarded as migratory and so dam builders have often assumed that they do not need to bother building fishpass facilities in rivers without salmon.

Yet even where fishways have been built they are invariably based on the salmon fishpass model and have been impassable for many native species. In southeastern Australia, where many dams were fitted with fishways modelled on those on European and North American rivers, native potamodromous silver perch have declined by more than 90 per cent since the 1940s and are now listed as a threatened species. Dams have totally eradicated migratory grayling and bass from some coastal rivers in the region.

In the tropics there only a small number of examples of fish ladders being successfully used by native species. 'Experiences with such structures as fish ladders in Africa,' says UN Food and Agriculture Organization fishery biologist, G.M. Bernacsek, 'have been few and unsatisfactory'. In South America, Yacyretá Dam on the Paraná River has been fitted with fish elevators costing $30 million which, according to the World Bank, were designed 'based upon the consultants' knowledge and experience with fish migrations on the Columbia River'. While only a few of the more than 250 species in the Paraná are well studied, it is known that at least some of its species migrate up and down the river several times during their lives. 'This aspect,' a internal World Bank evaluation of its loans for Yacyretá dryly notes, 'was not considered.' And so the Yacyretá elevators, based on salmon migrations, only transport fish upriver.

World Bank and Thai government officials for years refuted the claims of independent Mekong fisheries experts and local fishing communities that the fish ladder planned for the highly controversial Pak Mun Dam would be largely useless and that the dam would have a devastating impact on the Mun River's highly diverse and productive fishery. Thai electricity utility EGAT even produced a video for national television promoting the experimental ladder as 'helping to conserve biodiversity.' Well before the dam was completed in 1994, however, fish catches in the Mun, the largest tributary of the Mekong, had dropped disastrously. In 1995, the Thai Department of Fisheries admitted that the experimental fish ladder was not working and EGAT agreed that local fisherpeople should be compensated for the loss of their fishery (although the World Bank still claimed that 'there has been no evidence to suggest that the dam will adversely affect fish stocks'). When a journalist from the Wall Street Journal visited Pak Mun in March 1996, 'two, small, dead fish [were] the only signs of life' in the ladder.

Furthermore, there are no bypass facilities at the dam, which is just upstream from the mouth of the Mun, to allow the scores of migratory fish species in the river to descend the river without a potentially lethal trip through the dam's turbines. Plodprasop Suraswadi, director of Thailand's Department of Fisheries, admitted to the Bangkok Nation newspaper in 1995 that there was a problem for fish migrating down the Mun but claimed that this would in fact be a good thing. 'This will pose no severe consequences,' Plodprasop said, 'as it would be beneficial for Thailand not to lose this group of fish to other downstream countries.'

Mitigating for the Cameras

To alleviate public concern over the massive numbers of animals drowned when a large reservoir is filled, dam authorities often plan highly publicised rescue operations. Despite decades of experience that these rescues are of extremely little benefit and repeated criticism from wildlife conservationists, dam builders still insist on mounting them, mostly because, as William Partridge, a senior World Bank environmental employee has cynically remarked about the Yacyretá rescue effort, they make 'good TV.'

Wildlife rescue plans fail to capture all but a tiny proportion of the affected animals, most of which drown or starve to death after being stranded on small islands or at the tops of partly flooded trees. The rescue operation at Thailand's Chiew Larn Dam, for example, captured only an estimated five per cent of the animals in the submergence zone. Furthermore, once the animals which have been captured are released they are often lethally stressed, frequently injured, and usually have no replacement habitat in which to live — if a suitable habitat is available it will already be inhabited by competitor animals. Rogério Gribel of the Amazonian research institute INPA says that 'saved' or not, 'all the animals from the flooded area should be considered dead.'

The EIA Industry

Our experience with environmental impact assessment is that when you predict major environmental impacts, the likelihood is that you will get major environmental impacts. The only problem is that you don't ever get quite the impacts you expect.

Professor Frank Grad
Columbia University Law School, 1992

Since the end of the 1960s a growing number of countries and international development agencies have followed the lead of the US in insisting that an environmental impact assessment (EIA) be written before any major infrastructure project be built. A thorough assessment of a proposed dam's possible environmental impacts should indeed be required before any project goes ahead. Unfortunately, governments and dam builders have invariably turned the EIA process into a bureaucratic formality, merely another regulatory hurdle which developers must jump before they can get their project approved. Governments and funders rarely treat EIAs as objective studies to be used to inform an open debate on whether or not a project is desirable, but as rubber stamps for projects they have already decided to build.

International environmental consulting is now a big and very profitable business. According to the British Consultants Bureau, UK consultants earned $2.5 billion on overseas contracts in 1994 — and the second biggest sector of the market after project management was writing EIAs. The environmental assessments for large internationally funded dam projects are invariably written by consultants from a relatively small number of companies, some of which, such as German consultants Lahmeyer International, are also directly involved in dam building. Others, such as Norwegian firm Norconsult, are subsidiaries of dam builders. There is an obvious conflict of interest when the company assessing the environmental viability of a project is also likely to get the contract to build it.

Even apparently independent environmental consultancies with no direct link to dam builders also have a strong self-interest in underplaying the environmental impacts of projects and exaggerating their benefits. If their conclusions are not favourable to the dam funders or builders then the consultants will be less likely to get contracts from those agencies or companies in future (the World Bank's guidelines on environmental assessment specify that consultants must be 'acceptable to both the World Bank and the local contracting agencies'). Consultancies, funders and builders often have warm and mutually rewarding relationships. British consultancy Environmental Resources Limited, for example, was awarded more than 11 contracts on World Bank development projects and eight with the UK government's Overseas Development Administration between 1985 and 1992 in South Asia alone.

Furthermore, there is no quality control on the consultants' reports — they are usually not peer reviewed as they would be were they to be published in an academic journal, and much worse they are often treated as state or commercial secrets and hidden from public scrutiny. This inbuilt bias for consultants writing EIAs to conclude what their clients want to hear means that the conclusions of an EIA for a large dam can often be guessed before reading the report: the dam's environmental impacts can be accurately predicted, will be relatively minor, and can be relatively cheaply and easily mitigated. In one form or another these seem to be the conclusions of almost every EIA for an international dam project.

Even when individual sections of an EIA are critical or raise concerns that some effects cannot be predicted these points are invariably toned down in the report's overall conclusions (and criticisms in drafts frequently disappear when they appear in final form). The 1994 feasibility study for a cascade of dams on the Mekong written by Canadian engineering and environmental consultants Acres International and French dam agency Compagnie International de Rhône, for example, states that 'not enough is known' about fisheries ecology in the river 'to predict the effects' of the dams. Yet the consultants predict that the 'environmental impacts of the proposed projects are expected to be . . . not severe'.

One of the clearest examples of a corrupting relationship between a dam building agency and an environmental consultancy is that between Thai utility EGAT and TEAM Consulting Engineering Company, a link which goes back for three decades. In 1978, EGAT commissioned TEAM Consulting to write the EIA for the Nam Choan Dam. Their final report was never publicly released but was used by EGAT to claim that the project would not have serious impacts on the two wildlife sanctuaries that it would partially flood. The Wildlife Ecology section of the EIA, however, was obtained by Belinda Stewart Cox, a British biologist doing research on the birdlife in the sanctuaries.

The TEAM consultants were unable to enter the submergence zone because it was held by Communist insurgents so they surveyed an area downstream which they presumed had similar habitats and then extrapolated this to the reservoir. Although the study contained no maps or location description Stewart Cox concluded from the species listed and omitted that TEAM had probably not studied riverine forest at all. TEAM's report failed to mention the ecologically valuable nature of the sanctuary areas which would be flooded; the effect of the reservoir on fragmenting animal populations; and the effect on aquatic species of converting a river to a reservoir. TEAM claimed that only six of the mammals it listed were classified as rare; Stewart Cox says that 35 were protected under Thai law.

TEAM also said that the reservoir would 'create favourable conditions for most bird species' because 'water birds find it easier to catch fish'. Yet according to Stewart Cox, only two out of 113 listed bird species would be likely to catch fish in the reservoir. Similarly TEAM stated that otters — which favour shallow, shady rivers — would benefit from the reservoir. Stewart Cox concluded that overall the TEAM report was 'inadequate, inaccurate, sloppy, misleading and, in some instances, apparently fraudulent. It is, in every respect, an inadmissible and unprofessional document.'

The storm of protest which Nam Choan provoked among environmentalists and local people forced EGAT to suspend the project. But EGAT did not blame for misleading them as to Nam Choan's possible impact. Instead they rewarded them with another EIA contract, this time for the World Bank-funded Chiew Larn Dam. Here TEAM's 'experts' found 122 wildlife species in the reservoir area — while the Royal Forestry Department's largely futile animal rescue operation found 338 species. Undaunted by TEAM's incompetence, EGAT then contracted them to do an environmental assessment for Pak Mun. TEAM claimed there were 80 fish species in the Mun — later surveys found more than 230 species. Mekong fisheries specialist Walter Rainboth reviewed a leaked copy of the Pak Mun EIA. 'Based on the importance of the project and the capacity for irreversible damage,' Rainboth concluded, 'the report is criminal. If something like this were submitted to Congress in order to solicit funds, its fraudulent nature would deserve criminal indictment.'

The extent to which the original purposes of environmental assessments have been subverted can clearly be seen at the Sardar Sarovar Project. In this case the World Bank and Indian authorities agreed that environmental studies for the world's biggest dam and irrigation project should be done parallel with, rather than before, work on the dam. Repeated criticisms of this approach were defended with the assertion that any environmental impacts would necessarily be less than the project benefits (although the authorities did not know what the environmental conditions prior to construction were, what the scale of the impacts would be, nor how much project benefits might be curtailed by environmental factors such as unsuitable soils in the areas slated for irrigation). The independent commission set up by the World Bank to review Sardar Sarovar concluded that this approach 'subverts any acceptable notion of ecological planning'.

Sardar Sarovar's backers have also claimed that continued monitoring will enable any serious environmental problems to be identified and mitigated. But this attitude fails totally to allow for the fact that many environmental impacts cannot be mitigated once the project is built (and that many others only can be by substantially redesigning the project). It is in fact depressingly common to find the assumption in EIAs that 'monitoring' is the same as mitigation, and that recording environmental damage will somehow stop it.

Consultants invariably write EIAs as if the projects were being built in a world without pressure to maximize profits and cut costs on environmental mitigation. EIAs very rarely discuss whether the mitigation measures they recommend have been implemented — and if implemented effective — for past projects. Nor do they tend to mention what the environmental impacts of other projects have been and whether they were accurately predicted. Even if consultants did wish to discuss the success or otherwise of environmental mitigation, however, their ability to do so would be restrained by the fact that environmental studies usually end before construction is finished. More than 60 per cent of 31 national dam agencies surveyed by the industry journal Water Power & Dam Construction in 1991 stated that they had no formal system for monitoring the impacts of dams in operation — despite the claim in almost every EIA that environmental monitoring will be a key part of mitigation.

The secrecy which frequently surrounds EIAs is the most indefensible part of the EIA industry. The environmental impacts of dams are extremely complex and difficult to predict. Putting a price tag on possible environmental costs and then comparing these with supposed economic benefits is a process fraught with difficulty, assumptions and personal bias. Coming to a decision on whether or not the environmental damage done by a dam will outweigh its benefits is ultimately a subjective and political decision which should be made after an informed debate by the affected people and the general public. Weighing up the cost of the extinction of a species or the desiccation of an estuary against the benefit provided by increased electricity generation should not be the sole remit of a firm of consultants with a vested interest in ensuring that more dams are planned and built.

An argument often used by dam builders and backers in developing countries to defend incomplete and biased environmental surveys is that concern for the environment is a 'first world luxury' which they cannot afford. In fact the opposite is the case. The majority of people in developing countries depend directly on their environment to provide them with subsistence. The environmental destruction caused by dams in developing countries (and to a lesser extent elsewhere) thus carries a major human cost, which falls most heavily on the poorest sections of society. People in developing countries, in fact, can least afford the environmental impacts of large dams.