|
--W"
<br />project objectives following completion [Pedroli et al.,
<br />2002; Bernhardt et al., 2005]. These problems suggest
<br />that the scientific practice of river restoration requires
<br />an understanding of natural systems at or beyond our
<br />current knowledge, and presents a significant chal-
<br />lenge to river scientists.
<br />Second, most restoration projects focus on a single,
<br />isolated reach of river, yet many scientists advocate
<br />the watershed as the most appropriate spatial unit
<br />to use for most river restoration [National Research
<br />Council, 1999]. Restoration undertaken within a
<br />watershed context reflects the importance of key
<br />processes and linkages beyond the channel reach, such
<br />as upstream/downstream connectivity, and hillslope,
<br />floodplain, and hyporheic /groundwater connectivity
<br />[Sear, 1994; Angermeier, 1997; Frissell, 1997; Poff
<br />et al., 1997; Stanford and Ward, 1992; Graf, 2001;
<br />Palmer et al., 2005]. The importance of these link-
<br />ages is without question; water, sediment, organic
<br />matter, nutrients and chemicals move from uplands,
<br />through tributaries, and across floodplains at varying
<br />rates and concentrations. Migratory fish move up-
<br />stream and downstream during different stages in their
<br />lifecycles. These obvious examples of the inextricable
<br />linkages within watersheds are too often ignored in
<br />river restoration. To date, restoration has largely been
<br />done on a piece -meal basis, with little to no monitor-
<br />ing to assess performance, and little integration with
<br />other projects. This reflects the lack of process -based
<br />approaches in current practice as well as the fact that
<br />comprehensive restoration strategies that reestablish
<br />watershed -scale connections and processes are more
<br />difficult to implement because of sociopolitical and
<br />financial constraints.
<br />Third, restoration is too often focused on creating
<br />a desired form that is then artificially constrained.
<br />Because natural variability is an inherent feature of
<br />all river systems, restoration of an acceptable range of
<br />variability of process is more likely to succeed than
<br />restoration aimed at a fixed endpoint that precludes
<br />variability. Restoration of process is also more likely
<br />to address the causes of river ecosystem degradation,
<br />whereas restoration toward a fixed endpoint addresses
<br />only symptoms. The widespread clearing of the exotic
<br />riparian shrub tamarisk from western rivers has been
<br />supported by the public, politicians, and managers
<br />because tamarisk is perceived to be the cause of the
<br />problem rather than one of the many symptoms of
<br />altered rivers. Tamarisk removal has been sold as a
<br />means of restoring diversity of native communities
<br />and salvaging water through decreasing evapotranspi-
<br />ration, yet no scientific study has been able to quan-
<br />tify the yield on these investments.
<br />To persist as healthy ecosystems, rivers must be able
<br />to adjust to and absorb change at the time scales
<br />over which change occurs. An ideal ecologically
<br />successful restoration creates hydrological, geomor-
<br />phological, and ecological conditions that allow the
<br />targeted river to be self - sustainable in its new context
<br />[Palmer et al., 2005]. One of the implications of this
<br />understanding of river dynamics is that monitoring
<br />and evaluation of conditions before and after restora-
<br />tion must recognize the variability inherent even in
<br />"stable" rivers. Restoration that focuses on process
<br />rather than form will more effectively address most
<br />restoration goals. Process is more crucial than form in
<br />goals such as a) improving water quality by changing
<br />infiltration -runoff paths and b) stabilizing banks and
<br />increasing pool volume by allowing riparian vegeta-
<br />tion to remain along river banks. Restoration projects
<br />that attempt to create a static or fixed form, such as
<br />meanders with riprapped banks, commonly fail [Kon-
<br />dolf et al., 2003]. Rivers possess physical integrity,
<br />an aspect of ecological integrity, when their processes
<br />and forms maintain active connections with each
<br />other in the present hydrologic regime [Graf, 2001].
<br />Advancing the science and practice
<br />Rivers are highly valued by the public; everyone
<br />interacts with and pays attention to rivers [Tunstall et
<br />al., 2000]. As the practice of river restoration contin-
<br />ues to grow, the need to develop a sound scientific
<br />basis is obvious, as evidenced by the number of
<br />working groups and policy initiatives devoted to this
<br />topic within the federal government (e.g. USGS inter-
<br />agency River Science Network), non - governmental
<br />organizations (e.g. The Nature Conservancy, Ameri-
<br />can Rivers, local watershed groups), and academia
<br />(e.g. the National River Restoration Science Synthesis
<br />project [Palmer et al., 2003] and the National Center
<br />for Earth- Surface Dynamics).
<br />Achieving restoration goals will be limited by a
<br />variety of scientific and non - scientific factors [Anger -
<br />meier, 1997; Hennessy, 1998]. Scientific limitations
<br />include unavailable information on critical ecosystem
<br />conditions or processes, and inadequate synthesis of
<br />available information during model development.
<br />Non - scientific limitations include infeasibility of
<br />
|