An overview of hydrologic models
There are many different types of models available for constructing a hydrologic foundation for ELOHA. Models differ in terms of required inputs, representation of hydrologic process and algorithms for calculating fluxes, and outputs. The Nature Conservancy working with the Pennsylvania Instream Flow Advisory Committee compared different approaches for creating a hydrologic foundation in Appendix 1 of their report on developing instream flow criteria published in 2008.
Statistical models are based on associations between streamflow and basin characteristics. Statistical models can be developed quickly, but have limitations in precision of predictions, efficiency for analyzing many different of potentially significant ecologically-significant flow characteristics, and ability to assess hydrologic alteration and future conditions. Flood statistics for rivers and streams in the United States are available from the USGS National Streamflow Statistics Program. Flood and other flow characteristics are available for a number of states in the US through USGS StreamStats.
The limitation of using a statistical approach to analyze many different flow characteristics has been addressed in Massachusetts by creating synthetic flow duration curves that are combined with streamflow records from index sites to generate synthetic hydrographs. This approach is known as "QPPQ" (Fennessey, 1994) In a recent application of Fennessey's "QPPQ" approach, Waldron and Archfield (2006) found that, except for extreme high and low flows, simulated flows at ungaged sites closely matched measured flows at gaged sites with long-term records under natural conditions when watersheds of gaged and ungaged sites were of similar size. Smakhtin (1999) avoided inappropriately associating particular ungaged and gaged sites by associating ungaged data with "weighted" values of all gaged data.
Process models account for the flux of water through different parts of the hydrologic cycle (precipitation, evapotranspiration, infilitration, ground-water recharge, runoff, and ground-water discharge). These models are generally more complex to construct than statistical models, but can be used to simulate many different types of scenarios (climate change and variability, water use and regulation, land development). A technical background and training are generally required for using process models. A few example of process models are:
USGS Watershed Modeling
US EPA BASINS (HSPF)
Lancaster University TOPMODEL
The development of the hydrologic foundation for ELOHA requires careful consideration of what information will be needed from start to finish for setting environmental flows and what financial, technical, and computational resources are available.