PREDICTION OF REGIONAL HEAT WAVES USING A HYBRID LSTM APPROACH Yash Joshi, Srinivasa Ramanujam Kannan ASME International Mechanical Engineering Congress and Exposition Proceedings Imece, 2025 Heat waves, characterized by prolonged periods of excessively high temperatures, pose significant threats to human health, economies, and infrastructure. Despite significant advancements in heat wave prediction models, existing approaches often lack fine spatial resolution and fail to incorporate topographical factors comprehensively. This limitation is particularly evident in regional studies, such as those focusing on Odisha, India, where local topographical variations significantly influence heat wave dynamics. Addressing this gap, the proposed model integrates topographical features to enhance spatial granularity and improve prediction accuracy for regional-scale heat waves in Odisha. Thus, the primary objective of this study is to develop a machine learning model for regional (0.25° × 0.25°) heat wave prediction that integrates both dynamic factors, such as temperature fluctuations and variability, and static topographical influences. This study employs a hybrid machine learning approach to predict regional heat waves at a fine spatial scale. Key dynamic feature of temperature fluctuations was modelled using Long Short-Term Memory networks (LSTMs), while topographical factors like elevation, vegetation cover, 500 mbar geopotential height and proximity to water bodies, wind components were modelled using XGBoost. After obtaining predictions from both models (namely LSTM and XGBoost), a weighted approach (hybrid) was then used to combine them for the final predicted temperature with a lead time of 3 days. This research enables prediction at regional scale, informs policy development, and strengthens public health responses in heat-vulnerable regions like Odisha.
DYNAMICS OF CLEAR SKY INDEX OF SOLAR INSOLATION M. V. Suprabhat Prasad, Srinivasa Ramanujam Kannan ASME International Mechanical Engineering Congress and Exposition Proceedings Imece, 2025 Understanding the variability of solar energy is essential for optimizing energy resource planning and utilization. This study investigates the clear sky index k (the ratio of actual insolation to ideal clear-sky insolation) using satellite-derived real-sky data (INSAT-3D) and ideal-sky data computed through GRASS GIS, across four strategic locations representing the geographical extremes of India. Daily k values were analyzed over several years to understand long-term insolation changes. Seasonal trends showed high fluctuations during monsoons, with R2 (Between real sky and clear sky) values dropping to nearly 0.00 during periods of heavy rainfall (>6 mm/day), and relatively stable in winter, with R2 values rising to 0.8. Mutual Information Content (MIC) analysis revealed strong nonlinear dependencies between precipitation and k, with MIC exceeding 0.9 during monsoon months in Odisha and Andhra Pradesh. Such deviations help identify key climate events, such as monsoon onset and winter shifts, and are critical for assessing solar insolation stability. These findings significantly enhance climate modelling accuracy, solar resource assessment, and energy project planning by accounting for the dominant influence of atmospheric factors. This study fills a critical research gap by offering a pragmatic framework incorporating satellite datasets, statistical analysis, and seasonal pattern identification for assessing solar energy viability across India’s variable climatic regions.
Integrated Anomaly Detection and Early Warning System for Forest Fires in the Odisha Region Hrishita Hiremath, Srinivasa Ramanujam Kannan Atmosphere, 2024 The present study aims to develop a random forest algorithm-based classifier to predict the occurrence of fire events using observed meteorological parameters a day in advance. We considered the skin temperature, the air temperature close to the surface, the humidity close to the surface level, and soil moisture as important meteorological factors influencing forest fire occurrence. Twenty additional parameters were derived based on these four parameters that account for the energy exchanged in sensible and latent forms and the change in parameters in recent trends. We used the mutual information approach to identify critical meteorological parameters that carry significant information about fire occurrence the next day. The top nine parameters were then fed as input to the random forest algorithm to predict fire/no fire the next day. The weighted data sampling and SMOTE techniques were employed to address the class imbalance in the fire data class. Both techniques correctly classified fire incidents well, given the meteorological input from the previous days. This study also showed that as the class imbalance increases to 1:9, the performance based on the precision, recall, F1 score, and accuracy are maximum, showing the model’s ability to perform with class imbalance. Both techniques helped the random forest algorithm forecast fire instances as the data sample size increased.
DETECTION OF RAINFALL VARIABILITY ON THE BASIS ORDINAL SCORE Parthasarathi Mishra, Srinivasa Ramanujam Kannan 2024 IEEE India Geoscience and Remote Sensing Symposium Ingarss 2024, 2024 Climate variability in specific rainfall variations has received considerable attention worldwide. The present work considers the daily rainfall data of TRMM Multisatellite Precipitation Analysis (TMPA) version 7 (3B42V7) high resolution $0.25^{\\circ} \\times 0.25^{\\circ}$ gridded data. The rainfall information for the Angul region (CE) in the state of Odisha, India and its surrounding areas (NE, NW, SE, SW) in the winter season (DJF) are extracted from TMPA. To compare the rainfall pattern of the central region (CE) with its surroundings (NE, NW, SE, SW), an ordinal score is calculated between two-time series of rainfall data to determine the similarities and contrasts between rainfall characteristics. The two-time series concerning the CE region do not have a higher ordinal score, indicating that the rainfall patterns in and around the Angul are not statistically similar. Furthermore, the ordinal score of the NE with respect to the CE region is considerably lower till 2008, indicating inhomogeneity of rainfall. Researchers failed to observe such a trend of inhomogeneity in rain.
Impact of Temperature Perturbation on the Atmospheric Instability Parthasarathi Mishra, Srinivasa Ramanujam Kannan 2024 IEEE Mediterranean and Middle East Geoscience and Remote Sensing Symposium M2garss 2024 Proceedings, 2024 The increase in anthropogenic activities due to large-scale urbanization and industrialization causes unprecedented changes in atmospheric conditions. Several studies reported anthropogenic emissions and their impact on air quality. However, fewer attempts are made to investigate the impact of temperature change due to anthropogenic activity on the Atmospheric Boundary Layer (ABL) and atmospheric instability processes. In this present study, the impact of temperature perturbation due to anthropogenic heating on atmospheric instability (Convective Available Potential Energy (CAPE) and Convective Inhibition Energy (CIN)) is studied exclusively by using the WRF model. Two main CAPE seasons are identified in the study. A high value of CAPE is observed in the monsoon season, and a low CAPE is observed in the pre-monsoon season. Even if a lower value of CAPE does not guarantee deep moist convection in pre-monsoon as more CIN is present.
Sensitivity of Microphysical Schemes to Predict Induced Rainfall Parthasarathi Mishra, Srinivasa Ramanujam Kannan 2024 IEEE Mediterranean and Middle East Geoscience and Remote Sensing Symposium M2garss 2024 Proceedings, 2024 In the present study, numerical simulation is carried out in the mesoscale Weather Research and Forecasting (WRF) platform to determine the best set of physics options for better prediction of hydrometers to capture industrial effects. Angul region (20.85ºN, 85.19ºE) in Odisha state is chosen as the study region for this purpose since the region houses a lot of industries. A comparative study is done with different microphysics (MP) options available in the WRF model. From this sensitivity study, it has been found WDM6 based MP scheme is better at predicting industry-induced rainfall.
Anthropogenic moisture emission and its impact in an urban environment: A numerical study Parthasarathi Mishra, Srinivasa Ramanujam Kannan 2023 IEEE India Geoscience and Remote Sensing Symposium Ingarss 2023, 2023 Moisture emissions from anthropogenic sources vary significantly on both spatial and temporal scales, and accurately estimating moisture emissions in a particular region of interest is quite challenging. Most of the earlier studies on anthropogenic emission either focused on the sensible heat component or the combined effect of sensible heat component and moisture components but largely ignored the impact of the moisture component alone. The present investigation mainly emphasizes anthropogenic moisture emission and its role in the local urban environment. The objective of the present work is to incorporate anthropogenic moisture in model simulation in the urban environment and study its effect on relative humidity and rainfall. Moisture perturbation only increases the horizontal diffusion of moisture without any change in rainfall pattern, whereas moisture and sensible heat perturbation significantly lead to a shift in rainfall pattern.
An Iteration-Based Methodology to Cross Compare Volume Matched Indian Ground Radar Reflectivity Observations Against Space Radar Alok Sharma, S. R. Kannan 2023 International Conference on Machine Intelligence for Geoanalytics and Remote Sensing Migars 2023, 2023 The present work describes an iteration methodology to correct ground radar (GR) reflectivity observations with space radar (SR) on the Tropical Rainfall Measuring Mission (TRMM) satellite. The GR observations are first matched at the same resolution in the horizontal and vertical direction using a volume matching methodology. The volume-matched GR reflectivity is corrected against SR reflectivity using an iterative technique.
IDENTIFICATION OF PRIME SOLAR FARM LOCATIONS IN INDIA BASED ON RECENT CLIMATIC TRENDS G Raj, SR Kannan Proceedings of the 28th National and 6th International ISHMT-ASTFE Heat and … , 2025 2025
Detection of Rainfall Variability on the Basis Ordinal Score P Mishra, SR Kannan 2024 IEEE India Geoscience and Remote Sensing Symposium (InGARSS), 1-4 , 2024 2024
Integrated anomaly detection and early warning system for forest fires in the Odisha region H Hiremath, SR Kannan Atmosphere 15 (11), 1284 , 2024 2024 Citations: 6
Impact of Temperature Perturbation on the Atmospheric Instability P Mishra, SR Kannan 2024 IEEE Mediterranean and Middle-East Geoscience and Remote Sensing … , 2024 2024 Citations: 1
Sensitivity of Microphysical Schemes to Predict Induced Rainfall P Mishra, SR Kannan 2024 IEEE Mediterranean and Middle-East Geoscience and Remote Sensing … , 2024 2024 Citations: 1
Anthropogenic moisture emission and its impact in an urban environment: A numerical study P Mishra, SR Kannan 2023 IEEE India Geoscience and Remote Sensing Symposium (InGARSS), 1-4 , 2023 2023 Citations: 2
A methodology to upscale IMD ground radar observations at the same resolution with TRMM PR reflectivity using ANN A Sharma, SR Kannan Remote Sensing Applications: Society and Environment 30, 100940 , 2023 2023 Citations: 3
A study on precipitation characteristics of Kal Baishakhi: A premonsoon thunderstorm event SR Kannan EGU23 , 2023 2023 Citations: 1
An Iteration-Based Methodology to Cross Compare Volume Matched Indian Ground Radar Reflectivity Observations Against Space Radar A Sharma, SR Kannan 2023 International Conference on Machine Intelligence for GeoAnalytics and … , 2023 2023
Sensitivity of temperature perturbation to precipitation: a parametric study P Mishra, SR Kannan 2023 International Conference on Machine Intelligence for GeoAnalytics and … , 2023 2023 Citations: 6
Study of rainfall pattern near industrial region by using an ordinate pattern-based approach P Mishra, SR Kannan AGU Fall Meeting Abstracts 2022, H34E-03 , 2022 2022 Citations: 4
Atmospheric convection caused by temperature dispersion in and around the industrial source and its effect on precipitation rate: Gaussian approach P Mishra, SR Kannan AGU Fall Meeting Abstracts 2022, A31D-06 , 2022 2022 Citations: 4
A Neural Network based Methodology to Retrieve 3D Rainfall from Microwave Imager using Space Radar and Ground Radar observations A Sharma, SRR Kannan AGU Fall Meeting Abstracts 2022, H34H-01 , 2022 2022
An Alignment-based Methodology to cross-compare Indian Ground Radar Observations with Space Radar A Sharma, SR Kannan AGU Fall Meeting Abstracts 2022, A31A-02 , 2022 2022
A Regression and Neural Network-Based Methodology to Improve Vertical Resolution of Matched Indian Ground Radar Reflectivity Observations A Sharma, SR Kannan IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium … , 2022 2022 Citations: 3
A numerical experiment to study the impact of temperature enhancement by anthropogenic heating on local weather at the Angul region of India P Mishra, SR Kannan Journal of Earth System Science 131 (1), 46 , 2022 2022 Citations: 10
The effect of anthropogenic heat and moisture on local weather at industrial heat islands: A numerical experiment P Mishra, SR Kannan, C Radhakrishnan Atmosphere 13 (2), 357 , 2022 2022 Citations: 14
Rain/no-rain classification from combined radar-Radiometer data using machine learning A Anand, SR Kannan Remote Sensing Applications: Society and Environment 25, 100682 , 2022 2022 Citations: 6
Effect of humidity on the performance of rooftop solar chimney H Dahire, SR Kannan, SK Saw Thermal Science and Engineering Progress 27, 101026 , 2022 2022 Citations: 23
A numerical experiment to study the effect of anthropogenic heat and moisture on local weather PS Mishra, SR Kannan 2021 IEEE International India Geoscience and Remote Sensing Symposium … , 2021 2021 Citations: 7
MOST CITED SCHOLAR PUBLICATIONS
Effect of humidity on the performance of rooftop solar chimney H Dahire, SR Kannan, SK Saw Thermal Science and Engineering Progress 27, 101026 , 2022 2022 Citations: 23
On the effect of non-raining parameters in retrieval of surface rain rate using TRMM PR and TMI measurements SR Kannan, C Radhakrishnan, D Subramani, B Chakravarthy IEEE Journal of Selected Topics in Applied Earth Observations and Remote … , 2012 2012 Citations: 20
A new ensemble-based data assimilation algorithm to improve track prediction of tropical cyclones D Subramani, R Chandrasekar, SR Kannan, C Balaji Natural hazards 71 (1), 659-682 , 2014 2014 Citations: 18
A new PCA-ANN algorithm for retrieval of rainfall structure in a precipitating atmosphere SR Kannan, R Chandrasekar, C Balaji International Journal of Numerical Methods for Heat & Fluid Flow 21 (8 … , 2011 2011 Citations: 16
The effect of anthropogenic heat and moisture on local weather at industrial heat islands: A numerical experiment P Mishra, SR Kannan, C Radhakrishnan Atmosphere 13 (2), 357 , 2022 2022 Citations: 14
Rainfall estimation from spaceborne and ground based radars using neural networks V Chandrasekar, SR Kannan, H Chen, M Le, A Alqudah Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International … , 2014 2014 Citations: 13
An artificial neural network based fast radiative transfer model for simulating infrared sounder radiances P Krishnan, SR Kannan, C Balaji Journal of Earth System Science, 1-11 , 2012 2012 Citations: 13
View Factors between Disc/Rectangle and Rectangle in Parallel and Perpendicular Planes S Abishek, SR Kannan, SS Katte Journal of thermophysics and heat transfer 21 (1), 236-239 , 2007 2007 Citations: 12
Intercomparison between IMD ground radar and TRMM PR observations using alignment methodology and artificial neural network A Sharma, SR Kannan Journal of Earth System Science 130 (1), 20 , 2021 2021 Citations: 11
A numerical experiment to study the impact of temperature enhancement by anthropogenic heating on local weather at the Angul region of India P Mishra, SR Kannan Journal of Earth System Science 131 (1), 46 , 2022 2022 Citations: 10
Numerical Investigation and correlations for heat diffusion through Planar Ablative Thermal Protection Systems SR Kannan, SS Katte Thermal Science and Engineering Progress 7, 279-287 , 2018 2018 Citations: 8
A numerical experiment to study the effect of anthropogenic heat and moisture on local weather PS Mishra, SR Kannan 2021 IEEE International India Geoscience and Remote Sensing Symposium … , 2021 2021 Citations: 7
Integrated anomaly detection and early warning system for forest fires in the Odisha region H Hiremath, SR Kannan Atmosphere 15 (11), 1284 , 2024 2024 Citations: 6
Sensitivity of temperature perturbation to precipitation: a parametric study P Mishra, SR Kannan 2023 International Conference on Machine Intelligence for GeoAnalytics and … , 2023 2023 Citations: 6
Rain/no-rain classification from combined radar-Radiometer data using machine learning A Anand, SR Kannan Remote Sensing Applications: Society and Environment 25, 100682 , 2022 2022 Citations: 6
Radiative transfer simulations for the MADRAS imager of Megha-Tropiques SR Kannan, C Balaji Journal of earth system science 120 (1), 1-17 , 2011 2011 Citations: 5
Study of rainfall pattern near industrial region by using an ordinate pattern-based approach P Mishra, SR Kannan AGU Fall Meeting Abstracts 2022, H34E-03 , 2022 2022 Citations: 4
Atmospheric convection caused by temperature dispersion in and around the industrial source and its effect on precipitation rate: Gaussian approach P Mishra, SR Kannan AGU Fall Meeting Abstracts 2022, A31D-06 , 2022 2022 Citations: 4
Upscaling IMD Ground Radar Vertical Reflectivity Using TRMM PR Observations and Artificial Neural Network A Sharma, SR Kannan 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 7079 … , 2021 2021 Citations: 4
A methodology to upscale IMD ground radar observations at the same resolution with TRMM PR reflectivity using ANN A Sharma, SR Kannan Remote Sensing Applications: Society and Environment 30, 100940 , 2023 2023 Citations: 3
