To model hydrology in a watershed, two of the most famous and widely used distributed models are SWAT and HSPF. I have some good understanding of HPSF, but I don't know when and why should one prefer one over the other. HSPF can simulate with a timestep of 1 minute. I have already heard that SWAT is mostly used for agricultural areas. What I want to know that what is the basic difference between these two models in terms of the processes they model?


2 Answers 2


This is a broad question, the best way to compare the models is to go through the manuals and compare the process algorithms to see how the various hydrologic processes are handled. There are other considerations as well, including:

  • what environment/watershed are you modelling? How large, how urbanized, what type of soils/environment/precipitation vs evapotranspiration, how important are snow processes, etc., and compare that to what the models were made for
  • what is the purpose of your modelling? Your choice of model will be different if you are looking at water balance vs flood forecasting vs water quality estimates
  • what data do you have available? Some models are more intensive with data requirements than others, and if you only have daily precipitation and temperature (as is typical) than that will restrict some model types
  • what level of accuracy do you need? The timestep options in the models will make a difference

That being said, generally my understanding from a broad perspective is that: - HSPF is a more detailed hydrologic model with larger data requirements for input forcings, and has also a more comprehensive process library with respect to water quality - SWAT as you said was developed for agricultural regions, has less strict data requirements and also some ability to model water quality parameters (though less detailed than HSPF)

Depending on the purpose of your modelling, you may want to consider:

  • starting with a simpler model, such as GR4J, just to get the ball rolling before investing more time into complex models, particularly if just need simple answers from the model
  • using a flexible model where the processes can be modified on the fly and you are not committed to a particular set of algorithms when you choose the modelling framework, e.g. the Raven hydrologic modelling framework

Hope that helps!


I'll preface by saying I have no hands-on experience with either model; however, HSPF has always seemed cumbersome whereas SWAT seems more user friendly. Again, just my read of the software without any practical experience. But, until that person comes along to answer you're question, here's my two cents:

The best way to answer this questions is to sit down with the model documentation side-by-side and compare [1, 2].

Both HSPF and SWAT are intended for estimating pollutant loading to rivers, so if you're not interested in pollution modeling, there are better/simpler tools out there. If you're interested in typical non-point source pollutants like phosphorous and nitrogen, SWAT might be the better tool since the user manual emphasizes nutrient cycles and would be easier to follow.

From the SWAT Manual, "SWAT was developed to predict the impact of land management practices on water, sediment and agricultural chemical yields in large complex watersheds." Since the model was specifically designed with the intention of modeling agricultural systems, the model contains crop growth and nutrient cycling.

From a quick review of the HSPF manual, it seems that there are modules that would allow you to track a specific pollutant of interest, if you can supply the model with an appropriate relationship. This is discussed in the PQUAL module.

Quoting from the HSPF Manual here: The PQUAL module section simulates water quality constituents or pollutants in the outflows from a pervious land segment using simple relationships with water and/or sediment yield. Any constituent can be simulated by this module section. The user supplies the name, units and parameter values appropriate to each of the constituents that he wishes to simulate. However, more detailed methods of simulating sediment, heat, dissolved oxygen, dissolved carbon dioxide, nitrogen, phosphorus, soluble tracers, and pesticide removal from a pervious land segment are available, in other module sections.

Last - these are just two examples for modeling. If you're interested in general sediment transport, there are a whole host of programs out there. You'd be better served by defining your problem explicitly as hydrology and hydrology software are broad and far reaching.


1: https://swat.tamu.edu/media/99192/swat2009-theory.pdf

2: https://nepis.epa.gov/Exe/ZyNET.exe/91018NWD.txt?ZyActionD=ZyDocument&Client=EPA&Index=1995%20Thru%201999%7C1976%20Thru%201980%7C2006%20Thru%202010%7C1991%20Thru%201994%7C2000%20Thru%202005%7C1986%20Thru%201990%7C1981%20Thru%201985%7CPrior%20to%201976&Docs=&Query=600980015&Time=&EndTime=&SearchMethod=2&TocRestrict=n&Toc=&TocEntry=&QField=&QFieldYear=&QFieldMonth=&QFieldDay=&UseQField=&IntQFieldOp=0&ExtQFieldOp=0&XmlQuery=&File=D%3A%5CZYFILES%5CINDEX%20DATA%5C76THRU80%5CTXT%5C00000025%5C91018NWD.txt&User=ANONYMOUS&Password=anonymous&SortMethod=-%7Ch&MaximumDocuments=15&FuzzyDegree=0&ImageQuality=r85g16/r85g16/x150y150g16/i500&Display=hpfr&DefSeekPage=x&SearchBack=ZyActionL&Back=ZyActionS&BackDesc=Results%20page&MaximumPages=1&ZyEntry=1&SeekPage=x


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.