A Little Something About Me

Sat Kumar Tomer

My name is Sat Kumar Tomer. I am a PhD student in the department of Civil Engineering at Indian Institue of Science, Bangalore. I maintain the AMBHAS project, and I am specialized in hydrological modelling, Python, MATLAB, CSS, and HTML.

Navigation

info

Research Interests

Download my book, Python in Hydrology from Green Tea Press.

The primary focus of my research work is to improve our capability to predict the hydrological variables and fluxes, with the ultimate goal of understanding the possible impact of landuse change, climate change on hydrological fluxes. Often hydrological models are validated against the variables which are either sparse in resolution (e.g. soil moisture) or averaged over larger regions (e.g. runoff). A combination of the distributed hydrological model (DHM) and remote sensing (RS) has the potential to improve the resolution. Data assimilation schemes provide a way to optimally combine the DHM and RS. Distributed hydrological modelling and the use of remote sensed data to improve the state of the distributed hydrological model.

Current Research

Development of a coupled distributed hydrological model

There is a need to accurately and robustly simulate the hydrologic response of watershed where insufficient data exist for detailed model calibration. The gap in the data can be filled using the remote sensed platforms. My research is aimed at development of a coupled distributed hydrological model capable of utilizing the remotely sensed data. The developed hydrogical model will be physically‐based, distributed‐parameter gridded surface‐subsurface hydrological model.

Estimation of surface soil moisture from microwave satellite data

One of the distributed variable which can be used from remotely sensed platform is surface soil moisture. The surface soil moisture can be retrieved using the active and passive microwave satellies. The active microwave satellites provide a fine spatial resolution (~20m) but a coarse temporal resolution (~20 days), while the passive microwave satellites provide a coarse spatial scale (~20 km) but a fine temporal scale (~3 days). The active microwave satellites provide the backscatter coefficient, using which the surface soil moisture should be retrieved. The passive microwave satellites provide the surface soil moisture products, which often contain a large bias. My research is aimed at retrieving the surface soil moisture from the active microwave satellites (e.g. ENVISAT, RADARSAT, RISAT), and improving the estimated surface soil moisture by passive microwave satellites (e.g. SMOS, AMSR-E).

Estimation of evaporation using MODIS

Another important variable which satellite can provide is the evapotranspiration. The evapotranspiration is estimated using the energy based models. These estimated ET is not calibrated/validated for the water balance. Hence, this can not be directly used in the hydrological models. The aim of my research is to utilize the estimated ET from the MODIS to improve the state of hydrological models.

Assimilation of remote sensing data into a distributed hydrological model

With all the best available models for hydrology, their constitutive relationships and inverse modelling, yet we are not able to achieve the predictions close to the real value in the field condition. This may be due to the fact that forcings (boundary conditions), parameters and measured data have some uncertainty. There is a requirement of a method which can combine the model and measurements, given the fact that both contain the uncertainty, and provide the best estimate. This can be done by means of data assimilation in which we calculate error between prediction and measured value and update the model to get the best value at present time and for future also. The aim of my research is to develop data assimilation algorithms suitable for assimilating the remotely sensed data into hydrological models.

Inverse modelling

The models requires the parameters, which are often are not directly measurable especaillay at a bigger scale. The inverse modelling can be used to estimate the parameters at the required scale by means of optimization algorithm. In hydrology, often the problem is ill-posed, and there is no optimal solution. The aim of my reserch is to develop algorithms which can provide an enseble of paramters which can be used in running the model.

Geostatistics

The hydrological process follow some spatial patterns. The aim of my research is to use geostatistics to undestand the underlying spatial patterns in the observation and model predictions by using tool like variogram and Kriging.

Future plan

I envision my future research to span across some interrelated sub-areas of hydrology. The unifying theme of the research will be improving the prediction of the hydrological variables by utilizing all available satellite and field measured data.

In continuation of my current focus in assimilation of remote sensed data in distributed hydrological modelling, my near term plan is to work on the research issues that include the following:

  1. Temporal downscaling of forcings (rainfall and evapotranspiration) from daily to hourly scale.
  2. Spatial downscaling of remotely sensed products.
  3. Combination of the active and passive microwave data to get a improved resolution (spatial and temporal).
  4. Real time predictions of hydrological variables.

Additionally, I have interest to work in the area that studies the dynamics of copuled distributed hydrological modelling at various scale, ranging from a small watershed scale to a river basin scale. The main motivation here is to reveal the underlying vital processed at different scales and exploit them for improved performance of the hydrological models.