Epupa Dam and Reservoir Options - A Review of the Aquatic Ecological Aspects of the Draft Feasibility Report

By: 
Kate Snaddon, Bryan Davies
Date: 
Thursday, January 22, 1998

SUMMARY OF COMMENTS

The following comments are based on the draft feasibility report, which was made available by Burmeister Partners, Namibia. The feasibility report examines the two options for the dam, reservoir and hydroelectric power plant: the Epupa and Baynes sites. My comments are limited to the aquatic ecological aspects of the project, and more specifically to effects on the riverine biota and habitat, water quality and quantity.

The main concerns can be summarised as follows:

Flow regulation
The regulation of flow in the Cunene River, downstream of the Epupa and Baynes dam sites, is recommended in the absence of an instream flow requirements (IFR) study. A minimum flow of 20m³/s has been recommended during the construction phase, but no supporting evidence for this flow is provided. Flows will vary daily during operation of the hydroelectric power station, but the effects of this variation are not investigated in the report.

Riverine Biota
The Cunene River has not previously been described in terms of its aquatic ecology. However, the report does not provide detail on the invertebrate communities which inhabit the river, and how these communities will be affected by either dam option. Few assumptions can be made in the absence of such data, and predictions of effects of impoundment are thus tentative.

Loss of Habitat and Biodiversity
Epupa reservoir will completely eliminate the spray-zone habitat of the Falls Area, and reduce riffle areas. This will lead to the loss of biodiversity, which cannot completely be mitigated.

The River Mouth
The likelihood of saltwater intrusion into the river mouth of the Cunene River cannot be ignored without sufficient evidence. Thus it cannot be assumed that the recommended minimum flow of 20m³/s during the construction phase will be sufficient to prevent this.

Water Quality
An increase in phosphorus concentrations in both impoundments could lead to blue-green algal blooms. Any mixing of inflowing water with the epilimnion, where light and temperature conditions allow primary production, will allow this phenomenon to occur. The report assumes no mixing, but this is a tentative assumption.

In conclusion, the report is inadequate for the assessment of the dam options, and no conclusions can be drawn as to whether either dam option should go forward.

GENERAL COMMENTS

A first general comment must be made about the lack of completion of this study. Although this was not the aim of this particular review, it is of great concern that the requirement for a comprehensive public consultation process and the development of a social mitigation program was not included in this report. This is a significant component of the study, and in its absence the report is of little value, to all interested and affected parties. In Section 1.4 of both Executive Summary reports, it is indeed written: "The omission of this element renders this an incomplete EA.." The report is thus unacceptable in terms of the Integrated Environmental Management methodology, which was adopted in South Africa as the accepted environmental impact assessment technique.

Furthermore, the two studies, Epupa and Baynes, are virtually identical in terms of the aquatic ecological effects of impoundment at the two sites. The production of two reports was thus a repetition of results, that could have been less-confusingly presented as one, concise report.

SPECIFIC COMMENTS

Flow regulation
The regulation of flow in the Cunene River, downstream of the Epupa and Baynes dam sites, is recommended in the absence of an instream flow requirements (IFR) study. A minimum flow of 20m³/s has been recommended during the construction phase, but no supporting evidence for this flow is provided. Flows will vary daily during operation of the hydroelectric power station, but the effects of this variation are not investigated in the report.

It is of great concern that regulation of flow in the Cunene River downstream of the Epupa and Baynes dams is recommended in the absence of a complete instream flow requirements (IFR) study. Such a study examines a river system in terms of its biota, water chemistry and discharge, and determines the flow requirements of the system in order to maintain its ecological functioning. The results of an IFR study provide a total quantity of flow that it required annually, as a percentage of mean annual runoff (MAR), and divides this total into blocks of flow throughout the year, according to seasonal requirements of the system. For example, this method determines the frequency and magnitude of required floods, as well as low flow periods.

It is unacceptable that any flow recommendations can be made in the absence of such a study.

For example, both studies claim in their executive summaries that "Maintenance of a minimum flow of approximately 20m³/s (during the construction phase) is deemed sufficient to create a functioning aquatic habitat in the river." The report fails to provide supporting information for this assumption. In addition, the construction phase of either dam option is more than five years, within which damage can be done to the river due to the alteration of the flow regime.

Furthermore, it is stated that "The environmental consequences of the inevitable daily variation of water levels is more difficult to assess." This concerns in particular the fish and benthic life forms. The report then proposes that such concerns will be addressed only during the construction phase, at which time the mitigation programme will be launched. Subsequent monitoring activities will further gather information on whether the proposed compensation flows are indeed sufficient. This is an unacceptable approach, where flows are determined retrospectively. The case of the Lesotho Highlands Water Project should be remembered here, where a retrospective IFR study produced results that showed that there was insufficient water in the Lesotho Highlands catchments to provide the instream flow requirements of the rivers, and to provide water to South Africa.

The results of the Epupa and Baynes reports are unacceptable in the absence of an IFR study.

Riverine Biota
The Cunene River has not previously been described in terms of its aquatic ecology. However, the report does not provide detail on the invertebrate communities which inhabit the river, and how these communities will be affected by either dam option. Few assumptions can be made in the absence of such data, and predictions of effects of impoundment are thus tentative.

The Cunene River has not previously been described in terms of its aquatic ecology. Thus, the riverine biota and the physical and chemical aspects of the river are not well-known. It is alarming that the sections of the report (such as those describing the Local Setting) that deal with these aspects are lacking in data. As a result, few assumptions can be made in the absence of such data, and predictions of effects of impoundment are thus meaningless. For example, sections in the main report on the invertebrate communities in the river deal only with the freshwater prawn, Macrobrachium vollenhovii. The reproduction of this prawn is described as being "...cued by seasonality and/or natural variability in the Cunene River, and that modifications thereof, flow and water temperature, in particular, may have significant implications on this population." It is very likely that the reproductive success of this species of prawn will be adversely affected by river regulation in the Cunene. In addition, the river must support a wide variety of other invertebrate species, that are not mentioned in the report. Many riverine invertebrates are dependent on flow cues for reproduction, and on the supply of organics and nutrients from upstream, for their survival. Therefore, it can be stated that these species will be affected by both the Epupa and Baynes dams. The lack of detail on these very important riverine inhabitants in the report is of great concern.

An example of the significance of a lack of data on the invertebrate fauna which can lead to negative and far-reaching consequences, is that of the blackfly, of the dipteran family Simuliidae. In South Africa, a number of cases have been documented where river regulation has led to the proliferation of pest species within this dipteran group (e.g. de Moor...). The females of these pest species require a bloodmeal for reproductive success, much like the mosquito, Anopheles spp. In most cases, local stock provide the food source for these invertebrates, leading to stock deaths or decreases in the general health of stock, especially cows. The Simuliidae proliferate in conditions of high flow below dam structures, and often outcompete other fast-flow species. This possibility was not addressed in the report, despite the fact that this could have devastating effects on Himba cattle herds.

Furthermore, it is likely that a warm monomictic lake, with water levels fluctuating regularly for hydropower production, will provide an ideal habitat for mosquitoes such as the malaria-carrier species of Anopheles spp. These species have extremely rapid turnover times, and proliferate in backwaters where temperatures are fairly high. Such backwaters could form suitable pools during drawdowns. It cannot be assumed that fish populations will keep the mosquitoes in check through predation.

Loss of Habitat and Biodiversity
Epupa reservoir will completely eliminate the spray-zone habitat of the Falls Area, and reduce riffle areas. This will lead to the loss of biodiversity, which cannot completely be mitigated.

It has been stated in a World Bank report on choosing sites for Hydroelectric Projects (Ledec et al., 1997) that "If critical natural habitats not occurring elsewhere are lost to a hydroelectric project, the result is an irreversible loss of biodiversity." Such a consequence cannot be completely mitigated.

In the case of Epupa Dam, the report states that "...the reservoir will also completely eliminate the spray-zone habitat of the Falls Area...." Such an impact cannot be mitigated and, especially in the absence of data on the communities supported by this habitat, should be seen in a very serious light.

Furthermore, the report adds, "...Epupa reservoir would reduce the cutoff pools and riffle habitats by 64%...and the loss of..(shallow riffle habitat)...is estimated at 58%." The riffle habitats often support the highest diversity in a river system, thus the loss of these habitats will inevitably lead to a loss of biodiversity.

The River Mouth
The likelihood of saltwater intrusion into the river mouth of the Cunene River cannot be ignored without sufficient evidence. Thus it cannot be assumed that the recommended minimum flow of 20m³/s during the construction phase will be sufficient to prevent this.

It is stated in the executive summaries of both reports that "The daily regulation of water...is of little importance to the river mouth area." This is proposed to be due to the retardation and buffering effects of the 210km of river between the Epupa Dam and the river mouth, and the 180km distance between Baynes Dam and the mouth. This statement is then contradicted in the following sentence "The minimum flow, however, is more important, since it prevents inflow of salt water to the fresh water habitat of the river mouth." The recommended minimum flow of 20m³/s is indeed part of the daily regulation of water in the river, and in the absence of supporting evidence for this recommended minimum flow, the claim that this quantity of water will maintain the river mouth is unsubstantiated and unacceptable.

Water Quality
An increase in phosphorus concentrations in both impoundments could lead to blue-green algal blooms. Any mixing of inflowing water with the epilimnion, where light and temperature conditions allow primary production, will allow this phenomenon to occur. The report assumes no mixing, but this is a tentative assumption.

The quality of water in the reservoirs could have significant implications for both the reservoirs themselves and the downstream reaches of the river. For example, phosphorus, as stated in the report, will be provided by the inflowing water and the decomposition of flooded vegetation. Increased phosphorus concentrations will lead to increased primary productivity, of, for example, blue-green algal groups such as Microcystis and Anabaena spp. Blue-green algae exude chemicals which may be toxic, such as geosmin (Harding, 1992). Such chemicals will affect fisheries productivity. In the Western Cape in South Africa, for example, the presence of geosmin, released by algae in an impoundment, has concentrated in the flesh of trout in a downstream trout farm (Snaddon & Davies, in press.). The chemical results in a distinctive and unacceptable odour and taste.

The report assumes that the stratification of the reservoirs will result in the trapping of phosphorus below the thermocline, while primary productivity occurs in the epilimnion. This will indeed be true if there is no mixing, and if the inflowing water is of a lower temperature than the reservoir epilimnion. In the event of mixing, such as the inflow of turbulent river water, an algal bloom may occur, with serious consequences for water quality in the reservoir, and thus the fisheries supported by the impoundment.

CONCLUDING REMARKS

Although some of the specialist reports have gone into some detail on the local setting of the Cunene River and its fauna and flora (such as the ornithological report), there are some areas of the report that lack the detail that is necessary for a comprehensive assessment of the dam option and indeed, whether any dam should be constructed. The riverine invertebrate communities have not adequately been assessed and recorded for this river, and the regulation of flow could have devastating effects on the invertebrates, with some consequences for local human populations. Furthermore, in the absence of an IFR study, it cannot be accepted that the recommended flows and mitigation measures that have been suggested in the report, will maintain the ecological functioning of the Cunene River, and its estuary.

Thus, the report is inadequate for the assessment of the dam options, and no conclusions can be drawn as to whether either dam option should go forward.


REFERENCES

Harding, W.R. 1992. Ph.D thesis

Ledec, G., Quintero, J.D. & Mejia, M.C. 1997. Good Dams and Bad Dams: Environmental and Social Criteria for Choosing Hydroelectric Project Sites. Sustainable Development Dissemination Note #1. A report produced for the World Bank. 22pp.

Snaddon, C.D. & Davies, B.R. In press. A preliminary assessment of some of the effects of a small inter-basin water transfer scheme in South Africa on discharge and invertebrate communities. Submitted to River Regulation: Research and Management.



Comments by
Kate Snaddon, Freshwater Ecologist
Freshwater Research Unit, Department of Zoology
Rondebosch 7700, South Africa
Phone: 27-21-650 3638 Fax: 27-21-650 3301
e-mail: ksnaddon@botzoo.uct.ac.za

and Associate Professor Bryan Davies
Freshwater Ecologist and Senior Lecturer
Freshwater Research Unit, Department of Zoology
Rondebosch 7700, South Africa
Phone: 27-21-650 3638 Fax: 27-21-650 3301

Further Reviews: