Dam construction in the upper reaches of Latin American rivers appears to lead to the disappearance of migratory stocks in the reservoirs and in the river upstream of the structure. The same occurs in reaches where a whole series of dams and reservoirs have been constructed. The development taking place in the region will mean that vast sections of catchment basins will be completely regulated (OEA, 1985).
Whilst induced spawning techniques have been used with many “piracema” or migratory species, this has not led to the development of a technology for large-scale production of fish fry and fingerlings, which would be necessary if a large proportion of the reservoir's stock were to consist of communities of the river's original migratory species - provided, of course, that the fishery authorities wished to maintain these communities in the reservoirs and upper reaches of the river. The proposed ratio between reservoir area to be restocked from hatcheries and surface area of ponds in the hatcheries is an indication of the current stage of development of production and stocking technologies (see Section 8).
Latin America's experience with floating installations or facilities to pass fish upstream is rather limited. Whilst it is possible to estimate the swimming capabilities of some fish using empirical methods, the size of migrating stocks and the number of different species requiring passage pose additional design problems.
Pool and weir-type installations appear to be efficient for passing fish over heights of 8–10 m. Whether or not the pass is efficient depends on: the location of fish pass entrances, its handling capacity, flow velocities at the fish pass entrances and in the pass itself and the time the fish are exposed to each of these velocities. The heights of 20 m and 22 m to be surmounted at the Ilha Grande (Godoy, 1985) and Corpus (Castello, 1982) dams respectively are the maximum recommended in the literature for this type of intallation. The modified and readapted Borland lock system at Salto Grande (Delfino, Baigún and Quirós, 1986) could be an alternative solution for surmounting heights of over 25 m, provided that a high handling capacity is not required (Boiry and Quirós, 1985). Small handling capacity is one of the major drawbacks of the system (Clay, 1961). If a high handling capacity is required overall and per species for dams more than 25 m high, the only available option would appear to be the fish elevator - e.g., the type proposed for Cierre Chapetón in the middle Paraná (Poddubnyi, Espinach Ros and Oldani, 1981). Installation, operating and maintenance costs differ widely for each of the systems mentioned (Boiry and Quirós, 1985).
Welcomme (1985) is of the opinion that the provision of a fish pass installation for the upstream movement of fish would only be justified where the migration is absolutely essential for the maintenance of fish stocks. This would seem to be the case in Latin America. Furthermore, as new dams are constructed the situation is likely to become increasingly difficult. Fish culture for restocking purposes cannot yet provide the answer. If fish culture were developed however, it could help with the recovery of stocks which have already been lost. Now that reaches are completely controlled and fewer and fewer fish manage to pass upstream, fish culture could be used in conjunction with fish pass installations, as a temporary, although fairly long-term, measure. In order to be able to assess the feasibility of constructing and operating a particular fish pass system in a limited spatial perspective, in other words, in a dam which is to be built at a given site in a given river basin, not only must the dam characteristics and the features of the river in that area be taken into account but, the following questions must also be answered:
are there breeding and rearing grounds upstream of the dam site?
do fish pass through the site in their upstream migrations and, if so, what is their distribution in time?
what size of stocks migrate upstream and what is their distribution in time and space?
When these data are obtained, it will be possible to determine the species and sizes of stocks to be passed upstream. There appears to be consensus in the region that this list should include species of the genera Salminus, Prochilodus, Semaprochilodus, Leporinus and Pseudoplatystoma and probably also the genera Brachyplatystoma, Brycon, Colossoma, Luciopimelodus and Paulicea. No data are available in the region on the size of stocks to be passed over dams, which means that it is not known what size of stock should be developed in order to maintain the fishery at the level at which it stood prior to closure of the river. A rough idea could be obtained by taking into account the size of the spawning and breeding grounds in the basin upstream of the reservoir, the fecundity of each migrating species, estimates of mortality due to natural causes, as well as recruitment for the fishery and the migration range upstream from the dam site. Later, data will also have to be obtained on:
the migration routes taken by individuals of each of the species concerned;
daily movements.
Analysis of the data and the characteristics of the dam and the reach of the river concerned will indicate the type(s) of installation to be provided and it will then be possible to draw a bar-chart giving an idea of operation on a daily basis and to provide some details about each design.
Following an analysis of the water speeds downstream of the dam, the river's hydrological regime and the course the fish take to approach the dam, it will be possible to determine where the fish entrances should be located. The results of this analysis may also indicate that additional structures should be provided downstream.
In order to complete the installation design, data on fish “burst” speeds and endurance, or an empirical estimate of these (see Section 8) will be necessary. The final choice will depend on economic considerations, the ease of passage offered and the size of stocks requiring passage as well as the type and efficiency of installations or other mechanisms which have already been provided in dams in the basin.
No thought has been given to providing special structures to help fish - juveniles or adults - descend Latin American rivers. Pool and weir passes and navigation locks have been considered suitable for this purpose (COMIP, 1986). It is generally assumed that the mortality rate for fish descending through turbines and weirs will be low (Dudley, 1975; COMIP, 1986). Reports have nevertheless indicated that it is as important to help the fish move downstream as it is to help them ascend (Bonetto, 1980).
