394
that may be necessary to cope with such intraregional variability may be so great as to preclude
simple monitoring and assessment. Third, partly
because the U.S. Environmental Protection Agency
draws ecoregion boundaries without regard to watershed drainage divides (Omernik, 1987; Omernik
and Bailey, 1997), many streams cross ecoregion
boundaries, and their biota may manifest (often in
idiosyncratic ways) the ecological influences of all
the ecoregions that they traverse (Corkum, 1989,
1990; Culp and Davies, 1982). The latter problem
becomes more acute if, to cope with unacceptable
intraregional variability, mappers decide to further
subdivide ecoregions (sometimes termed subecoregions) (Bryce and Clarke, 1997). The smaller the
regional units mapped, the larger the proportion of
streams that will be confounded by boundary effects. These difficulties are not necessarily insurmountable, and to the extent that they are based
on patterns of biotic response that are demonstrated
to be general and repeatable, the lEI and related
multimetric, bioindicator approaches hold much
promise for field assessment.
26.2.3 Guild-based Approach
The composition of local assemblages can be
viewed non taxonomically in terms of the traits possessed by the assemblage members. Such a traitor guild-based approach allows biological communities to be compared across biogeographic boundaries, because the traits, not the species, are the unit
of measurement. This coarse-grained approach has
utility insofar as traits can be identified that are sensitive or responsive to measurable habitat variables
that change naturally across, or in response to human modification of, the landscape. For example,
Poff and Allan (1995) characterized fish species in
the upper Midwest in terms of five life history and
other biological traits hypothesized to be sensitive
to hydrologic variability (a surrogate for habitat stability). In an analysis of the fish assemblages in 34
streams, they found that these ecological traits did
correlate in a meaningful way with several measures of hydrologic variability. Interestingly, when
assemblages were defined by strictly taxonomic
criteria without regard to ecological information,
there was no clear correspondence of the resulting
assemblages with hydrologic variability. Establishing such relationships can help in predicting how
assemblages may change in response to changes in
important environmental or habitat variables. Increasingly, species-level data are being assembled
for a number of ecological traits, including trophic
requirements, reproductive behavior and habitat,
Assessment of Biotic Patterns in Freshwater Ecosystems
life-history patterns, pollution tolerance, and response to disturbance (see Poff, 1997, for a review).
This approach, while insensitive to individual
species identity, has promise for assessing assemblage response to habitat change and comparing assemblages across broad scales.
26.2.4 Assessment of Hydrologic
Variability in Rivers
Most larger rivers in the temperate world (Dynesius and Nilsson, 1994) and many smaller ones are
altered by flow regulation, dams, and diversions.
Flow regulation systematically affects many aspects of stream ecosystems and their biota. The life
histories of the species indigenous to any river system are adapted to the naturally prevailing hydrologic regime of the stream (Amoros et al., 1987b;
Poff and Ward, 1989, 1990; Poff et aI., 1997), and
alteration of flow patterns directly affects biota
even when other aspects of the stream system (e.g.,
channel morphology, riparian vegetation) remain
largely unchanged. Over time, flow alterations often cause substantial, deep-seated changes in
stream ecosystems, including changes in geomorphic structure and dynamics, riparian and floodplain vegetation, and the exchange of surface and
groundwater, all of which have additional impacts
on biota (Bravard et aI., 1986; Amoros et aI., 1987a,
b; Decamps et al., 1988; Johnson, 1994; Ligon et
aI., 1995; Stanford et aI., 1996; Poff et aI., 1997;
Ward, 1997).
Recent studies have demonstrated that hydrologic regimes of streams and rivers can and should
be characterized not just in terms of mean states of
water flow across seasons, but also in terms of variability, constancy, and predictability of flow variation at several time scales, from hours to decades
(Amoros et aI., 1987a, b; Poff and Ward, 1989,
1990; Stanford and Ward, 1992; Poff, 1996;
Richter et al., 1996, 1997). Virtually any of these
aspects of flow regime may be altered by river regulation or other hydrologic changes in the catchment. Hydrologic variability and alterations can
best be evaluated where detailed hydrographic
records exist or can be accurately synthesized; but
if these are not available, similar information may
be developed through intraregional comparisons of
river flow pattern (Poff, 1996). Restoration or rehabilitation of streams and rivers must include returning flow regimes to some semblance of their
natural-historical flow regime (Amoros et aI.,
1987a, b; Stanford et aI., 1996; Poff et al., 1997;
Ward, 1997), making such assessments central to
conservation and restoration efforts.
that may be necessary to cope with such intraregional variability may be so great as to preclude
simple monitoring and assessment. Third, partly
because the U.S. Environmental Protection Agency
draws ecoregion boundaries without regard to watershed drainage divides (Omernik, 1987; Omernik
and Bailey, 1997), many streams cross ecoregion
boundaries, and their biota may manifest (often in
idiosyncratic ways) the ecological influences of all
the ecoregions that they traverse (Corkum, 1989,
1990; Culp and Davies, 1982). The latter problem
becomes more acute if, to cope with unacceptable
intraregional variability, mappers decide to further
subdivide ecoregions (sometimes termed subecoregions) (Bryce and Clarke, 1997). The smaller the
regional units mapped, the larger the proportion of
streams that will be confounded by boundary effects. These difficulties are not necessarily insurmountable, and to the extent that they are based
on patterns of biotic response that are demonstrated
to be general and repeatable, the lEI and related
multimetric, bioindicator approaches hold much
promise for field assessment.
26.2.3 Guild-based Approach
The composition of local assemblages can be
viewed non taxonomically in terms of the traits possessed by the assemblage members. Such a traitor guild-based approach allows biological communities to be compared across biogeographic boundaries, because the traits, not the species, are the unit
of measurement. This coarse-grained approach has
utility insofar as traits can be identified that are sensitive or responsive to measurable habitat variables
that change naturally across, or in response to human modification of, the landscape. For example,
Poff and Allan (1995) characterized fish species in
the upper Midwest in terms of five life history and
other biological traits hypothesized to be sensitive
to hydrologic variability (a surrogate for habitat stability). In an analysis of the fish assemblages in 34
streams, they found that these ecological traits did
correlate in a meaningful way with several measures of hydrologic variability. Interestingly, when
assemblages were defined by strictly taxonomic
criteria without regard to ecological information,
there was no clear correspondence of the resulting
assemblages with hydrologic variability. Establishing such relationships can help in predicting how
assemblages may change in response to changes in
important environmental or habitat variables. Increasingly, species-level data are being assembled
for a number of ecological traits, including trophic
requirements, reproductive behavior and habitat,
Assessment of Biotic Patterns in Freshwater Ecosystems
life-history patterns, pollution tolerance, and response to disturbance (see Poff, 1997, for a review).
This approach, while insensitive to individual
species identity, has promise for assessing assemblage response to habitat change and comparing assemblages across broad scales.
26.2.4 Assessment of Hydrologic
Variability in Rivers
Most larger rivers in the temperate world (Dynesius and Nilsson, 1994) and many smaller ones are
altered by flow regulation, dams, and diversions.
Flow regulation systematically affects many aspects of stream ecosystems and their biota. The life
histories of the species indigenous to any river system are adapted to the naturally prevailing hydrologic regime of the stream (Amoros et al., 1987b;
Poff and Ward, 1989, 1990; Poff et aI., 1997), and
alteration of flow patterns directly affects biota
even when other aspects of the stream system (e.g.,
channel morphology, riparian vegetation) remain
largely unchanged. Over time, flow alterations often cause substantial, deep-seated changes in
stream ecosystems, including changes in geomorphic structure and dynamics, riparian and floodplain vegetation, and the exchange of surface and
groundwater, all of which have additional impacts
on biota (Bravard et aI., 1986; Amoros et aI., 1987a,
b; Decamps et al., 1988; Johnson, 1994; Ligon et
aI., 1995; Stanford et aI., 1996; Poff et aI., 1997;
Ward, 1997).
Recent studies have demonstrated that hydrologic regimes of streams and rivers can and should
be characterized not just in terms of mean states of
water flow across seasons, but also in terms of variability, constancy, and predictability of flow variation at several time scales, from hours to decades
(Amoros et aI., 1987a, b; Poff and Ward, 1989,
1990; Stanford and Ward, 1992; Poff, 1996;
Richter et al., 1996, 1997). Virtually any of these
aspects of flow regime may be altered by river regulation or other hydrologic changes in the catchment. Hydrologic variability and alterations can
best be evaluated where detailed hydrographic
records exist or can be accurately synthesized; but
if these are not available, similar information may
be developed through intraregional comparisons of
river flow pattern (Poff, 1996). Restoration or rehabilitation of streams and rivers must include returning flow regimes to some semblance of their
natural-historical flow regime (Amoros et aI.,
1987a, b; Stanford et aI., 1996; Poff et al., 1997;
Ward, 1997), making such assessments central to
conservation and restoration efforts.