API Publ 4736:2006 pdf download

API Publ 4736:2006 pdf download.ldentification of Key Assumptions and Models for the Development of Total Maximum Daily Loads.
The objective of watershed models is to simulate the entrainment of pollutants (including sediment) from ground surfaces by precipitation and subsequent surface water runoff and pollutant transport to surface waters. Watershed models may be used to generate inputs to receiving water quality models or may include their own receiving water modeling capability. Ground water models are not usually considered as watershed models, but in specific instances where ground water contributions k) surface waters may be important pollutant sources, ground water models may also serve as watershed models. Some watershed models also include the ability to simulate ground water pollutant contributions in a watershed.
These models are referred to as watershed models because they are intended to predict runoff volumes and pollutant loadings from the surface areas that are tributary to a receiving water body’5. The geographical size of the watershed can range from very small tributaries (which may he referred to as sub-watersheds) to entire river basins.’6
These models have the following attributes in common:
• They use land use data, soils data, and surface slopes as the basic foundation for estimating pollutant loadings:
and
• They calculate the pollutant loadings as a function of the volume of pollutant runoff that is estimated from precipitation-runoff correlations.
The models differ in the degree of detail in predicting the entrainment and transport of pollutants into surface waters. The time scales used for pollutant loading predictions can be in years (e.g.. annual averages), days, or even minutes. The temporal scale of the model should be based on the needs of the study — if shorttcrm variations in water quality are important. watershed analysis time scale must be short enough to be useful to the TMDL analysis. Conversely, in other cases (e.g.. nutrient loadings) predictions on a seasonal or annual basis will be adequate for a
TM DL.
The following sections describe generic categories of watershed models available for TMDLs. ranging from the simplest loading equations to complex watershed models capable of simulating large watersheds, multiple pollutants, and receiving water impacts.
Loading equations
The simplest methods for pollutant load predictions are equations relating pollutant loadings to land use. EPA’s (onspt’ndiun refers to these as ‘simple models.” These models use empirical equations to correlate pollutant loadings to land-use and runoff volumes. Table 4-I lists a few examples of loading equations used for watershed pollutant load predictions.
Comprehensive watershed modeling is needed for TMDLs when watershed runoff pollutant loadings contribute significantly to surface water quality impairment. The required temporal scale of water quality predictions also determines the complexity of simulation models. For example. if surface water quality variations associated with single storm events are important, then a comprehensive, detailed watershed model is needed. These models will generate short-term waste load estimates used in dynamic receiving water quality models (or the waste load model may perform this impact analysis) to predict responses to short-term events.
Conversely, if evaluation of seasonal or annual variations in surface water quality is the objective of a TMDL (which will often be the case for nutrient enriched waters). less complex models can he used. It may even be possible to use a watershed loading equation with site-specific data to satisfy the TMDL evaluation needs, In such cases, the receiving water quality modeling can often be performed with a steady-state water quality model using the seasonal or annual waste loads predicted by the watershed model.
Table 4-3 lists some examples of comprehensive watershed pollutant loading models used for prediction of watershed runoff for evaluations that are consistent with the requirements of a
TM DL.
All of these models have extensive site-specific data requirements that must be satisfied for them to function as adequate simulation models. All require calibration and validation beflire they can be considered as suitable for predicting pollutant loadings for TMDLs.