9386 - Laserfiche WebLink

Home Browse Search

and processes of the river prior to damming, which vary tremendously among rivers. In addition, the ef- fects of dams on flow is only half of the story-the amount and type of sediment captured by the dam and the amount and type of sediment brought into the stream below the dam (e.g., by bank erosion and tribu- taries) is of critical importance in determining the best flow regime for maintaining geomorphic form and processes. Although it may be pos- sible to devise a flow schedule that would transport sediment much as the natural flows did, if the sedi- ment load were to be reduced and/or the size distribution altered, such a flow schedule might result in sig- nificant adjustments of the channel. What is needed is an individual pre- scription for each river. It must in- volve both water and sediment (and, in some circumstances, engineered alterations of the stream). It must be based on a geomorphic/ecological assessment of the effects of the darn on the river ecosystem. The question remains open as to how often and for what types of stream this approach is actually likely to preserve biologically key morphological and hydraulic fea- tures of the physical habitat. In many cases an assessment may give insight into what has happened or is likely to happen but does not offer a solu- tion. An assessment procedure We have found no established pro- cedures or protocols for assessing the ecological effects of downstream geomorphic adjustments to dams. Because we recommend that a geo- morphic analysis be a part of any environmental evaluation of a dam, we offer a possible outline of such an approach. (We should point out that we have not yet applied this approach in its entirety.) The ap- proach has five steps. We assume a worst-case scenario: that a dam is proposed for a river for which we have little or no background geo- morphic or biological information (as may be common in many devel- oping countries) and that we have only two years in which to make predictions and recommendations. The steps are as follows: 40 30 a 0 20 N '?? 10 t ? " ? ??% 1940 1950 1960 1970 1980 1990 Year Figure 7. The estimated number of fish in the McKenzie River chinook salmon spring run decreased on average by ap- proximately 50% from the period 1960 to 1969 (no data were collected from 1961 to 1968 or in 1984). Source: Adapted from Howell et al. 1988. • Characterize and quantitatively describe the channel and watershed. The stream and watershed should be characterized by such features as stream type, bar morphology, de- gree of confinement, and vegeta- tion. Digital elevation data for the entire affected watershed should be obtained and used to estimate the longitudinal profile of the river and the local drainage area at different locations on the river. The size distribution of the bed, bank, and floodplain sediments of the main channel and major tributaries should be determined. Any available his- torical maps or aerial photographs should be used to assess geomorphic processes (e.g., channel migration) and changes in morphology (e.g., channel width). • Monitor water and sediment dis- charge. Unless stations already ex- ist, some monitoring, if only for one or two years, should be attempted. The monitoring is oriented toward gathering data in order to calibrate models rather than for a statistical analysis of the frequency of runoff and sediment transport. Several monitoring stations should be es- tablished, one at the dam site and the other stations downstream in the main channel and on major tribu- taries. • Estimate pre- and post-dam sedi- ment budgets and hydrology. A sedi- ment budget is a quantitative state- ment of the rate of production, trans- port, and discharge of settlement. The sediment budget above and be- low the dam site should be esti- mated. This task requires estimat- ing the sediment discharge to the channel from hillsides, tributaries, and bank erosion upstream and downstream of the dam site. The digital elevation data collected in the first step can be used to create a numerical surface on which quanti- tative sediment budget analyses can be conducted (Dietrich et al. 1993). In addition, regional sediment yields and their controls should be ana- lyzed. • Model effects of dam on down- stream bed elevation and grain size. Depending on the type of river (e.g., sand-bedded versus gravel-bedded) and dam, a model should be as- sembled, usually only one-dimen- sional, that can be used to predict the downstream transport of sedi- ment and sediment storage changes for a specified flow regime in order to predict changes in bed elevation and grain size. Many models al- ready exist, although they need up- dating to specifically address scour and armoring effects. • Predict channel response to the dam using theoretical and empirical models. This step is the most diffi- cult. It is far easier, though still not always straightforward, to assess how an already dammed river has responded than to predict how a river is likely to respond to dam- ming. Under certain combinations of altered sediment and water dis- charge, some changes may be fairly easy to predict (e.g., incision/aggra- dation). However, predicting the entire suite of changes and adjust- ments that will occur is far more difficult, particularly changes in planform (i.e., the morphology of the river as viewed from above). Fluvial geomorphology is not at the point where one can conceptually take apart a river and understand how all of its morphological and process variables interrelate and then put it all back together in a predic- tive model. We are developing a methodology that combines exist- ing theoretical models with an em- March 1995 189