ELOHA is built on a "hydrologic foundation" of information about water resources in a region.  As shown on the ELOHA flow chart and the Hydrologic Foundation flow chart detail, the information is used to classify river types based on flow characteristics and to assess ecological responses to hydrologic alteration, as well as to evaluate the status of sites relative to environmental flow standards.  To provide an adequate foundation for ELOHA, hydrologic information needs to:

• be spatially comprehensive so as to include both locations where water managers may want to make allocation or other water management decisions, as well as sites where ecological data have been collected;
• represent historical (unaltered), current, and future conditions;
• include the range of ecologically-relevant flow characteristics; and 
• address ground-water and estuarine flows where appropriate. 

An ideal hydrologic foundation is a regional database of daily or monthly streamflow hydrographs for baseline (unaltered), current, and future conditions over a common time period that represents variability in climate (generally 20 years or more).  The database needs to have enough spatial detail to resolve reaches with different streamflow characteristics (e.g., because of an intervening tributary) and small streams that nonetheless provide significant habitats and may have altered hydrology because of ground-water pumping, diversions, or regulation.  Some global and regional streamflow databases are readily accessible on the internet.

Many different approaches have been used to develop hydrologic information in hydro-ecological analyses.  The approaches have comparative advantages and disadvantages for building a hydrologic foundation for ELOHA.  Hydrologic modeling is a comprehensive approach to address all of the criteria for ELOHA's hydrologic foundation.  Hydrologic modeling can extend the periods of streamflow data for gauged analysis nodes and synthesize data for ungauged analysis nodes as needed for unaltered, current, and future conditions.  To achieve these objectives, unaltered streamflow data are needed for model calibration, data on current water management and uses must be incorporated into the model, and the model must be capable of representing changes in climate, land use, and water management. Existing hydrologic model or decision support system for water management may be adapted to build a hydrologic foundation for ELOHA.  Kennard et al (2009) provide guidance on selecting the period of record for estimating hydrologic metrics for ecological studies.

Resources

ELOHA Flow Chart in
  Spanish
   English

Hydrologic Foundation Flow Chart

An over-view of hydrologic modeling explains the main types of models for estimating flows at ungaged streams and gives some examples.

Tools for building a hydrologic foundation lists more than 25 models by type, capability, limitations, and required skills, and links them to case studies.

Global streamflow  databases 

SOME CASE STUDIES

For a complete list of hydrologic foundation models and case studies cited in the ELOHA toolbox, see the Tools for Building a Hydrologic Foundation table.

SLURP Model:  Mekong River Basin

The Semi-Distributed Land-Use Runoff Process (SLURP) is a continuous simulation distributed hydrological model designed to make maximum use of remotely sensed data. SLURP divides a basin into subbasins, which are further divided into areas of different land covers.  For each land cover within each subbasin, the model simulates the vertical water balance, transforming daily precipitation into evapotranspiration and runoff separately, and routes the runoff downstream through the basin. 

Using only global datasets that are readily accessible on the internet, Kite (2000) used SLURP to synthesize daily streamflow hydrographs in the 800,000-km² Mekong River basin.  Topographic data were obtained from the United States Geological Survey (USGS) GTOPO30 public-domain DEM; land cover was obtained from the USGS 1-km digital land cover map of the world; soil parameters were obtained from the FAO Digital Soil Map of the World; and daily climate data (precipitation, temperature, dew point, and wind) were obtained from the US National Climate Data Center's Global Surface Summary of the Day (GSOD) internet database.  Radiation data were estimated from daily precipitation data.  Major dams and diversions also were included in the model.  Results of the model were converted into time-series of areas flooded (see figure), which were used to evaluate fish and irrigation productivity under different water allocation scenarios.

Flooded areas of different land covers surrounding Tonle Sap Lake - a main product of IWMI's SLURP model of the Mekong

SWAT and SWRRB:  United States, Guatemala

(For basins with good basin characterization, but no streamflow gages.)

The U.S. Department of Agriculture developed the Simulator for Water Resources in Rural Basins (SWRRB) to predict the effects of various watershed management schemes on water and sediment yields in large, complex, ungaged rural basins (Williams et al, 1985).  The program models surface runoff, return flow, percolation, evapotranspiration, transmission losses, pond and reservoir storage, sedimentation, and crop growth to generate daily flows over multi-year time periods.  Although SWRRB is intended for use where calibration data are not available, it has been validated on 11 large agricultural watersheds in the United States (Arnold and Williams, 1987) and in a 2,671-hectare watershed in Guatemala (Maldonado et al, 2001).  Input requirements include daily temperature, solar radiation, and rainfall data (or monthly data with daily conversion parameters); soil, crop, pond, reservoir, and irrigation system characteristics; channel routing data; and other basin characteristics.

The SWRRB model has been superseded by the SWAT model, which has an improved GIS-based interface, and the ability to do much more detailed simulations.  The basic model equations are the same as in SWRRB. 

ACRU Model: South Africa

The ACRU (and ACRU2000) model (School of Bioresources Engineering and Environmental Hydrology, 2007; Kiker et al, 2006) emerged from the Agricultural Catchments Research Unit at the University of KwaZulu-Natal in South Africa to integrate water budgeting and runoff with risk analysis.  Using a multi-layer soil water budgeting approach, ACRU simulates hydrology, crop yield, reservoir yield, and irrigation water demand/supply.  An advantage of ACRU, which has spurred its widespread use, is its flexibility of operation, depending on the availability of input data.  For example, potential and actual evaporation, interception losses, soil water retention constants, leaf area index, hydrograph routing, reservoir storage:area relationships, and crop growth periods all may be estimated by various methods according to the level of input data available or the simulation accuracy required.  The model uses daily time steps, although it accepts monthly input data.  For large catchments, ACRU operates as a spatially distributed model linked to a Geographic Information System (GIS).  ACRU was tested by simulating daily streamflows for 137 sub-catchments in the 4,387-km2 Mgeni River watershed in the KwaZulu-Natal province of South Africa over a 15-year period (1979-1993). The model satisfactorily simulated the highly variable flows that characterize southern African hydrology. The model had less success in areas encompassing large reservoir and transfer schemes combined with urban and peri-urban areas (Tarboton and Schulze, 1991; Kiker et al, 2006).

Thorthwaite-Mather Soil-Water Balance: Costa Rica, Indonesia, Kenya, Mexico

In cases where detailed watershed characteristics are difficult to obtain, lumped water balance models can often generate reliable streamflow hydrographs with few data requirements. With just rainfall and potential evapotranspiration data and generalized soil and aquifer characteristics, the lumped Thornthwaite-Mather (T-M) procedure calculates monthly water balances for relatively large watersheds (Thornthwaite and Mather, 1957).  The T-M procedure has successfully been used in its original or modified form to estimate monthly streamflow in numerous countries, including Kenya (Dunne and Leopold, 1978), Indonesia (Peranginangin et al., 2004), Mexico (Mendoza et al. 2003), and Costa Rica (Calvo, 1986).

India

Jha and Smaktin (2008) review the existing methods used in India for estimating flow characteristics at ungauged sites. It focuses on low and high flows, long-term mean flow, and flow duration curves. Since it lists the actual formulae, it can be used as a quick reference guide for selecting a suitable technique for various geographical, regional and/or river basins in India.

Colorado Stream Simulation Model (StateMod), Colorado, USA

StateMod is a program developed by the State of Colorado, USA, to simulate water allocation and accounting for comparing different water management scenarios in a large river basin.  StateMod is capable of simulating daily and monthly water allocation conditions over the 1975-2005 and 1909-2005  time periods, respectively.  To generate baseline conditions for ELOHA (locally known as WFET, the Watershed Flow Evaluation Tool), CDM et al (2009) changed the input files to eliminate any human influences on the river basin, including diversions and reservoir operations.  StateMod is an implementation of the Colorado Water Conservation Board's Colorado Decision Support System.

Photo credits (left to right): Photo © Harold E. Malde (vernal pools at Table Mountain); Photo © Cheryl Rose (cormorant in wetlands habitat).