I work in all the three languages i.e., MATLAB, Python and R on a regular basis. However, when it comes to make suggestion to someone who is about to start his/her research in hydrology, I always recommend Python.

One of the major advantages of Python and R over MATLAB is no licensing fee.

A hydrological modeler may work in a consultancy or may start his/her consultancy. If s/he works in a consultancy or even in an MNC, s/he might be asked to work on Python or R, as the commercial licence of MATLAB is too expensive to afford when equally efficient free options are available. Imagine starting your own consultancy and purchasing commercial license of MATLAB (it will burn the hole in your pocket).

Another advantage of Python is that in case one want to automate any process in ArcGIS/ QGIS one can use it easily, whereas MATLAB and R cannot be used. One may say that this can also be done using modeler in ArcGIS, however, I would say that modeler can better be used to integrate different tools but when it comes to analyse thousands of images and tens of operations, scripting becomes the only solution.

In case somebody is working in Linux system or uses Linux/Windows/Mac on need basis, getting license of MATLAB (or sometime even pirated version) becomes difficult. On the other hand Python and R are available free of cost in all platforms.

If one wants to automate any program or .exe file in Linux terminal or Windows Command Prompt, Python can be used to automate the process by converting the scripts into an executable.

These days, Google Earth Engine, another powerful application, also have the provision of using Python scripts. So, if one has to do computation on powerful Google Cloud, Python is the only choice among the discussed three.

The single advantage of R over Python and Matlab is the availability of sophisticated and advanced statistical packages. These packages have very good and elaborate documentation and are unanimously accepted in the scientific community. However, if one has expertise in Python switching to R would not be a problem.

So, if you ask which one to choose, I would answer, choose Python as first option and MATLAB as last option.

Hope it helps!

Regards

DS Bisht

This question can't be answered easily. What do you mean with "best"? Do you mean the most accurate, the fastest or maybe the easiest to learn?

While Python3 is quite easy to learn and use, it has some drawbacks like it is quite bloated and multithreading is not so easy to implement (ok, this isn't easy in many languages). Fortran is quite fast, but it is very old and not really developed anymore. For hydrological modelling I would suggest you should dig into R and/or PhreeqC: https://www.usgs.gov/software/phreeqc-version-3

Dear Roshan Singh Thagunna,

I believe the question actually lacks more specification. Do you mean an actual programming language to write your own hydrological model? Or a programming language to run an already implemented model? Well in both cases, there is no optimal solution I can think of. Writing your own model might not be suitable given the actual plethora of available models. Unless you have proven that specific processes occur in your study area that none of the existing models can reproduce. Moreover, it might take a lot of time to write it down, debug, run on tests cases...Definitely not what you want.

Also, most of the available models (commercial or open-source) features (generally) nowadays a GUI interface...So you might not need to write down some code to run the model itself. This is highly convenient, especially if you are starting with hydrological modelling.

However, if you are going to perform hydrological modelling, having skills in programming might help a lot: processing files, cleaning data, inputs, process model outputs, draw graphs, etc. As such, I will suggest either Python, Matlab or R. I myself use R and it is actually a marvel, to me. Find below an interesting paper discussing the possible applications of R (and its available libraries) for hydrology:

Article Using R in hydrology: a review of recent developments and fu...

Interestingly, there are some nice packages to learn hydrological modelling with R, implementing the well-known GR conceptual and lumped models!

Poster airGR: a suite of lumped hydrological models in an R-package

Presentation airGRteaching: an R-package designed for teaching hydrology ...

Good luck!

I prefer to use MATLAB. Like extreme value analysis of long-term precipitation series. Or a rainwater harvesting / "non-potable water demand" simulation.

BTW, a similar question was asked at RG two years ago; https://www.researchgate.net/post/What_is_the_best_programming_language_for_environmental_studies_and_modeling_eg_climatology_hydrology_etc

MATLAB, Python, and R

R programming is good

and secondly is python programming is preferable

I have used "R" to select representative general circulation models selection and downscaling of climate data for the transboundary Koshi river basin in China and Nepal. So, I recommend "R".

Article Representative General Circulation Models (GCMs) selection a...

Hmmm, nobody mentioned Fortran so far, interesting. As others mentioned earlier the language is often chosen based on existing tools that will be implemented for a particular problem. Not often you will be starting from scratch. That's where the power of Python, R and Matlab come in hand.

Maybe Python will be a good choice.

If you are looking to start Hydrological modeling from scratch that's going to take a while what ever language you choose. There are a few orgs that have been developing in this area for years eg UK Met office (Hadley centre) and NOAA and Uk Hydrographical office https://www.gov.uk/government/organisations/uk-hydrographic-office and others. For example also see https://openweathermap.org/models

To be relevant your research would need to reference such models so might be an idea to get on a collaborative program and see what's already being done before re-inventing from scratch.

Good news is that weather/climate forecasting used to require supercomputers but can now be done ( at larger scale ) on groups of server class machines. All depends on the scale/accuracy and mathematical stability of your algorithms. Getting reliable input data and controlled prediction will be the software headaches.

Matlab and python

thanks

Dear Roshan,

Generally, MATLAB and R are the best ones for large-scale data handling. However, Python is also one of the good sources as you can easily connect you geospatial datasets with it too.

Cheers

I work in all the three languages i.e., MATLAB, Python and R on a regular basis. However, when it comes to make suggestion to someone who is about to start his/her research in hydrology, I always recommend Python.

One of the major advantages of Python and R over MATLAB is no licensing fee.

A hydrological modeler may work in a consultancy or may start his/her consultancy. If s/he works in a consultancy or even in an MNC, s/he might be asked to work on Python or R, as the commercial licence of MATLAB is too expensive to afford when equally efficient free options are available. Imagine starting your own consultancy and purchasing commercial license of MATLAB (it will burn the hole in your pocket).

Another advantage of Python is that in case one want to automate any process in ArcGIS/ QGIS one can use it easily, whereas MATLAB and R cannot be used. One may say that this can also be done using modeler in ArcGIS, however, I would say that modeler can better be used to integrate different tools but when it comes to analyse thousands of images and tens of operations, scripting becomes the only solution.

In case somebody is working in Linux system or uses Linux/Windows/Mac on need basis, getting license of MATLAB (or sometime even pirated version) becomes difficult. On the other hand Python and R are available free of cost in all platforms.

If one wants to automate any program or .exe file in Linux terminal or Windows Command Prompt, Python can be used to automate the process by converting the scripts into an executable.

These days, Google Earth Engine, another powerful application, also have the provision of using Python scripts. So, if one has to do computation on powerful Google Cloud, Python is the only choice among the discussed three.

The single advantage of R over Python and Matlab is the availability of sophisticated and advanced statistical packages. These packages have very good and elaborate documentation and are unanimously accepted in the scientific community. However, if one has expertise in Python switching to R would not be a problem.

So, if you ask which one to choose, I would answer, choose Python as first option and MATLAB as last option.

Hope it helps!

Regards

DS Bisht

Python first and alternative choice is Matlab

MATLAB and R are useful for hydrological modeling, especially statistical analysis. However, if you are about to start your research, it's better to learn Python.

At some aspect, R software is much dynamic and flexible platform in compare with MATLAB. Also, easy to download (free of cost).