NonBasicSpecificComments04

• Title: Comments 4
• Date: 10/23/2003
• SWSI - Doc Type: Comments

210

18

BRIAN D. RICHTER ET AL.

Ecological Applications
Vol. 13, No.1

16

g
14

c:
.Q
f!'12
::s
O.
~
Ii:
~

I::
a:s
CD
::!:
1;
::J
C
C
<

Major Dam Construction

10

8

,

J

o
~ ~ ~ ~ ~ I;;
Q) 0) 0) CI en CJ)
..... ...... .,.. ..... ,.. ....

..,..IIif_ ........... ........ -- ...___ ____ ...... _.... .....-------111.

~ tr ~ &; ~ &; ~
v' 0) 'WI 0';1 m (7). 0)
..... .,.. T""'" ,.. ,... ... ,...

~ ~ to sf
Q) CD en 0)
..... ,.. ...... ~

FIG. 3. Using long-term merisurelnents of river flows for the Roanoke River in North Carolina, Richter et al. (19970)
applied their ~4 Range of Variability Approach" nlethod to assess cllanges associated with majoi' danl constructioll in 1956.
Initial ecosystenl flow requireruents for each of 32 paranleterS (such as annual low-flow duration, portrayed here) were then
defined in ternlS of a range of values. For instance, one ecosystenl flow target was to restore low-flow.duration (defined as
the cumulative nutll-ber .of days Cd) during which flows are <="96 mj/s) to correspond more closely to its. historical range of
va.rIabiHty. This targ~t specified that 50% of Jnean. annual low..ftow durations wo.uld faU Witllil1 the range ~hown here with
horizontal dashed bars; 2.5% WQuld faU below this range, and 25% would fall above this range, Low..flow conditions are
Il~eded to dl'Y out floodplain soils to enable reprodu~tion and growth of plants.

Estimating ecosystelTI flow requireluents requires in-
put froln an interdisciplinaxy gro~p of scientists fa--
lniliar \vith the habitat reql.lirenlents of native biota
(i.e., species, comlTIunities) and the hydrologic, geo-
morphic, and biogeochemical processes that influence
those habitats and support p~'inlary productivity and
nutlient cycling (Swales and Harris ] 9951 King and
Louw 1998; Icstream Flow Council [available online;
see footn'ote 6]). In South Africa] expert assessnlent
workshops are being convened for the Plu~P9Se: of de-
fining necessary flows to support desired fl.lture con..
ditions of riverine ecosystenls (King et a1. 2000). Dur"';
ing these workshops. interdisciplinary participants
draw upon existing data) research results, ecological
and hydrologIcal nlodels, and professionaljudglnent in
developing initial targets for ecosystem flow require-
. ments (King and Louw .1998). A wide variety of tools
and methods is being used worldwide to prescribe eco-
system flow requirements', and these approaches an~
evolving rapidly (Tha11ne 1996, Arthington ana Zal-
ucld 1998, Bragg and Black 1999~ Railsback 20011
Tharme, DO O[[([]; Instrean1 Fl'ow. Conncil [available
online; see footnote 6J).
Defining ecosystel'n flow requirements presents
nlany difficult cha llenges for scientists. For in'stance.,
the liuk between flows and the viability of a l1ative.
species population may not be well understoodJ and
. certainly not known for .all populations of native riv-:
erine species. Population viability also depends upon
a nwnber of. other eco~ystem conditions that are also

influenced by, or unrelated to, flow variations, thereby
obfuscating relationships between flow' variables and
population viability. AsseSSlllents of ecosystel'll flow
requirements sho uld n~ t be limited to considcn:a tion of
species needs, however. The_ flow needs of individual
species provide only a very limited perspecdve of the
bl'oader rallge offlows needed to conserve heaLthy river
ecosystems. Of great importance is evaluating the flo\v
conditions (and particularly, disturbance events asso-
ciated with droughts and floods) that structure river and
floodplain ecosystems (Hill et al. (991, Richter and
Richter 2000, Trush et al. 2000). A river's natural flow
reginle is a cornerstone for detern1ining ecosystem flow
requirell1ents; ~cosystelU flow prescriptions should at.
ways minlic natural flow characteristics to the extent
possible (Poff ct. at 1997, Tharme and King 1998),
It is very important that assump'tions and hypotheses
about flow-biota rela~ionships, other nonflqw related
variables that affect' biota; or the infiu~l1ce of flow on
other ecosystem conditions such as water quality or
physical habitat, be m8;de explicit when defuling initial
estinlates of ecosystem. flow requirements~ Developing
conceptual ecological models that depict presulned re-
lationships is" an excellent way of communicating hy-
potheses (Richter and Richter 2000). Hypotheses
should be formulated in a. manner that allows them to .
be tested through carefully designed w.ater management.
experiments (Step $), These bypothes.es should also, to
the extent possible, express the range of variation in
selected ecosystel11 indicators that is expected under