8 keys to develop a Thesis in Distributed Hydrological Modeling

 

You have finished your undergraduate/postgraduate studies in water resources and want to make a good research thesis? If this is the case, you have several alternatives: research about groundwater, superficial water, floods, inundation areas, flow/ecosystem interaction, even water supply systems. All these areas have specific applied software to evaluate and simulate the actual and future water requirements.

This article is focused on superficial hydrology, specifically in distributed hydrological modeling and has the intention to show the context of a evaluation using this modeling so you can successfully finish your thesis.

 

Why planning your thesis?

The main step to finish your thesis is... start it. In other words, start researching, look for relevant information, and, the most important part, plan.

To understand a subject you are interested in you need from 3 to 4 months of research work. Now, that work is done in the last years of study, so you can reach the thesis phase already having an idea of what you want to do.

When doing this kind of modeling thesis, you need to know how to use the modeling tool, this is why you need to start learning the software earlier. The 6 months you have to finish your thesis is not enough.

 

Why distributed hydrological modeling?

We could call the distributed hydrological modeling as the most advanced tool to calculate the physical processes of water cycle at a basin level. By being distributed considers the spatial variation of precipitation processes and vegetation cover, type of soil, radiation, among others. Some models even consider the type of precipitation (rainfall or snow) and the accumulation of snow on the surface. The models can be calibrated with field observations, such as flow, soil moisture and baseflow.

The distributed hydrological models are a tool to understand the hydrological answer in a basin and for water management. Research with these kind of tools is extremely important to improve the capacities of water resources professionals.

 

The keys to make your thesis

In case you are in the process of making your thesis or you are interested in developing it on this area, we will give you some keys to simplify your research:

1. Start with the literature review

To start understanding and be more familiarized with hydrological modeling, I would recommend you to read 10 papers about it. Almost all papers are in English. You can go to sciencedirect.com and type "hydrological distributed modeling". You will get a result list like this one:

Some articles may be asbstract, but consider that you are new in this. It is recommendable to read the complete article.

2. Find an advisor that helps

What we are going to mention this in general terms. You will probably have a lot of problems in finding an advisor who is an expert in hydrological modeling.

We have talked about the huge differences between the advance of actual software and the software taught in Universities. In Universities they will not teach you the latest software, but the one that the teacher learned when he was a student, as simple as that.

Another issue is that the advisors do not trust that students will finish their thesis, because students drop it halfway. To find an advisor who is interested in your thesis, you have to show that you are capable, that you dominate software, that you have researched, and everything will depend on your personal capacities to make your project accepted.

Thanks to the bureaucracy in Universities this may take some time, and this is the time you will spend researching. If your advisor starts giving you observations, you have less time to research. Remember, 6 months is not a lot.

3. Choose a distributed hydrological model

There are many of them and they also are open source! We would recommend USGS-PRMS; an advanced and applies model developed by the United States Geological Survey (USGS). 

There is also HEC-HMS in its gridded version However, we have not used it in projects yet. Based in our experiences, HEC-HMS is not that good for simulating large data series. The GUI is the enemy of the model capacity. We will discuss this later.

You can use SWAT, which is spatially distributed. It is important to mention that it does not consider soil moisture on its calculus, which limits the evaluation of infiltration and baseflow.

Since you are researching, you can check the TOPMODEL software which is implemented in GRASS, the Variable Infiltration Capacity Model (VIC) and the HBV.

4. Start from the beginning and reach more

To start understanding how the hydrological modeling implemented in the distributed models work, you need to start using the Soil Moisture Accounting (SMA) package in HEC-HMS. There, you will understand how complex are the required variables to run a model (we never said this would be easy).

There are not many tutorials about it, but after checking the information and characteristics of other models you should be able to understand the basic concepts of this model. Since SMA in used in the not-gridded version of HEC-HMS, it helps you to simplify the analysis.

5. Learn programming

Developing a distributed hydrological model means processing a huge amount of data. That is why you need an advanced tool of data processing.

If you are a WIndows user and think that you can process all data in Excel, you are wrong. If you want to use macros in Visual Basic you will probably have limitations. You need an advanced tool to process data, it can be R or Python. We recommend these two for being easy to understand and apply, and have a lot of documentation available.

Since we only use free software, we will not recommend Matlab.

6. Forget about the GUI

As we said, the GUI is the enemy of the capacities of models. A lot of models can be run in Windows but through the DOS. 

Using models this way is not hard, it even reduces the required time of preprocessing/simulating/importing results, but you need to adapt to this.

7. Find a basin with historical data

One of the advantages of distributed hydrological modeling is that you can use a big area of study, so you know you will have flow stations inside the model boundaries. Depending on the country where you live, different websites will offer you flow data. We recommend to have a daily flow records for at least 4 years to have a solid model.

For precipitation data, some software like USGS-PRMS let you use stations located outside the area of study. Depending on the software you are using, you will be able to import stations or process the precipitation data of each subbasin.

8. Use satellite images

Defining the vegetation cover, soil moisture, actual evapotranspiration or even vectors for precipitation distribution can be obtained from satellite images. The use of these images is common in hydrological modeling.

 

We hope that these keys have given you a perspective of distributed hydrological modeling. We are happy that you have reached this part of the article, which shows that you are interested in this area. We wish you the best of luck in your thesis and we hope that in the future you will use this tool to understand and manage the water resource in a sustainable way.

Saul Montoya

Saul Montoya es Ingeniero Civil graduado de la Pontificia Universidad Católica del Perú en Lima con estudios de postgrado en Manejo e Ingeniería de Recursos Hídricos (Programa WAREM) de la Universidad de Stuttgart con mención en Ingeniería de Aguas Subterráneas y Hidroinformática.

Smiley face

Subscribe to our free e-newsletter for tutorials, articles, webminars, courses and more.