akhilesh.kumar@unsw.edu.au
Akhilesh Kumar
I'm Akhilesh from India. Throughout my graduation in civil engineering, master's in satellite image processing, and research fellowship in hydrology, there has been only one thing constant: Remote Sensing and that's exactly what I will be doing at UNSW but this time with caves, mines, machine learning & groundwater recharge!
Project: Feasibility Assessment of Neural Networks for Groundwater Rceharge Event Prediction at Australia’s National Groundwater Recharge Observation System (NGROS) Sites
Supervised by:Â Andy Baker, Adrian Fisher, Wendy Timms (Deakin University)
Project Description:Â Groundwater modeling traditionally relies on physics-based or conceptual approaches, most of which require extensive data, parameter tuning, and computational resources. These methods struggle with capturing the complex, non-linear, and non-stationary nature of groundwater systems, leading to a shift towards machine and deep learning. These data-driven models, which detect patterns between input variables, perform well with even simple data like meteorological records and topographical features. However, there is no consensus on a learning architecture, input variable(s) or hyperparameter tuning that can work in different hydrogeological environments, especially one as tricky as Australia. Moreover, there has been no study attempting to predict groundwater recharge 'events' using any model. This project aims to fill this gap by conducting the first comprehensive characterization of NGROS sites and developing a machine learning framework to predict recharge events based solely on meteorological and geographical data, addressing a critical need in groundwater management.
- Publications
Journal articles
1. KUMAR, A. AND MEHTA, M., 2024. Investigating the Effect of Relative Spectral Response on the Estimation of Atmospheric Parameters: A Case Study of Landsat 8 (OLI) and Sentinel 2 (MSI). Journal of the Indian Society of Remote Sensing, 52(10), pp.2115-2122.
2. KUMAR, A. & MEHTA, M. 2023a. Evaluation of surface reflectance retrieval over diverse surface types using SREM algorithm in varied aerosol conditions for coarse to medium resolution data from multiple spaceborne sensors. International Journal of Remote Sensing, 44, 3358-3384.
3. KUMAR, A. & MEHTA, M. 2023b. Investigating the applicability of a simple iterative approach for aerosol optical depth (AOD) retrieval over diverse land surface types from Landsat 8 and Sentinel 2 using visible and near-infrared (VNIR) spectral bands. Atmospheric Environment, 314, 120082.
4. MITRA, S. S., KUMAR, A., SANTRA, A. & ROUTH, S. 2023. Investigating impact of CORDEX-based predicted climatic and LCM-based LULC scenarios on hydrologic response of a semi-gauged Indian catchment. Environmental Monitoring and Assessment, 195, 450.
5. SANTRA, A., KUMAR, A., MITRA, S. S. & MITRA, D. 2022. Identification of Built-Up Areas Based on the Consistently High Heat-Radiating Surface in the Kolkata Metropolitan Area. Journal of the Indian Society of Remote Sensing, 50, 1547-1561.
6. SANTRA MITRA, S., KUMAR, A., SANTRA, A., MITRA, D. & ROUTH, S. 2021. Hydrological modeling of catchment specific runoff-response to variable land-use/climatic conditions and trend-based hypothetical scenario generation: a study on a large river basin in Eastern India. Journal of the Indian Society of Remote Sensing.
7. SANTRA MITRA, S., SANTRA, A. & KUMAR, A. 2019. Catchment specific evaluation of Aphrodite’s and TRMM derived gridded precipitation data products for predicting runoff in a semi gauged watershed of Tropical India. Geocarto International, 1-16.
8. YADAV, N. K., SANTRA, A., SAMANTA, A. K., KUMAR, A., MITRA, S. S. & MITRA, D. 2021. Understanding the synergistic relation between land surface temperature and different biophysical parameters in Haldia industrial city of India. Arabian Journal of Geosciences, 14, 2412.
Book chapters
1. MEHTA, M. & KUMAR, A. 2023. Unit-7 Image Statistics. MGY-005 Techniques in Remote Sensing and Digital Image Processing. Indira Gandhi National Open University, New Delhi.
2. MITRA, S. S., KUMAR, A., SANTRA, A. & ROUTH, S. 2021a. Comparative Evaluation of Predicted Hydrologic Response Under Two Extremities of Sustainability Using Transformed Landuse- Landcover and CORDEX-Based Climatic Scenarios. Climate Impacts on Sustainable Natural Resource Management.
3. MITRA, S. S., SANTRA, A., KUMAR, A. & ROUTH, S. 2021b. Long-Term Drought Assessment and Prediction Driven by CORDEX-RCM: A Study on a Hydro-Meteorologically Significant Watershed of West Bengal. Mapping, Monitoring, and Modeling Land and Water Resources. CRC Press.
Conference papers/proceedings
1. KUMAR, A., SANTRA MITRA, S., SANTRA, A. & ROUTH, S. 2020. Precipitation and Runoff Trend Analysis Using Mann Kendall and Sen’s Slope Estimator for Kangsabati River Basin in West Bengal. International Conference on Sustainable Water Resources Management Under Changed Climate. Jadavpur University, Kolkata, India.
2. KUMAR, A., SANTRA MITRA, S., SANTRA, A., ROUTH, S. & SINHA, S. Evaluation of Long-Term Precipitation Trends, and Its Seasonality over a Catchment in Lower Gangetic Basin of West Bengal. ISH- HYDRO 2019 International Conference, 18-20 December 2019 Osmania University, Hyderabad.