Faculty Publications

Now showing 1 - 3 of 3

Future transition in climate extremes over Western Ghats of India based on CMIP6 models
(Springer Science and Business Media Deutschland GmbH, 2023) Shetty, S.; Umesh, P.; Shetty, A.
The effect of climate change on the tropical river catchments in the Western Ghats of India is studied using the Coupled Model Intercomparison Project-6 data (CMIP-6). Multi-model ensembles of rainfall and temperature are constructed using the Random Forest ensemble technique for bias-corrected GCMs in the near future (2014–2050) and far future (2051–2100) horizons. For the two catchments each in the southern, central, and northern Ghats, the trend in minimum and maximum temperatures, precipitation, and other indices are calculated. By 2100, dry sub-humid and humid catchments will see a higher increase in mean annual temperature than per-humid central catchments. In future decades, the warm days and nights increase by 45–50% and 40–70%, respectively, with twofold warming in the winter season. Under a climate change scenario, annual rainfall increases in Vamanapuram, Ulhas, and Purna, while Chaliyar, Netravati, and Aghanashini catchments experience a decrease in rainfall in the far future with an increase in pre-monsoon rainfall. The southern catchments are anticipated to have contrasting variations in the rainfall extremes; northern catchments face a substantial increase in very wet to extremely wet days and medium to heavy rainfall. In all catchments (excluding Vamanapuram), cumulative wet days increase with a decrease in cumulative dry days. After the mid-twenty-first century, humid to per-humid catchments encompass an increase in cool nights, whereas it disappears in dry sub-humid catchments of the Ghat. Interestingly, warming tendencies begin to slow down after 2050. This investigation can assist in comprehending the regional climate extremes in the Western Ghats to formulate better climate risk planning and adaptation strategies. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.
The effectiveness of machine learning-based multi-model ensemble predictions of CMIP6 in Western Ghats of India
(John Wiley and Sons Ltd, 2023) Shetty, S.; Umesh, P.; Shetty, A.
The popularity of cutting-edge machine learning ensemble approaches has solved many climate change research and prediction issues. The six top-performing GCMs obtained from Technique for Order Preference by Similarity to an Ideal Solution were ensembled using seven machine learning ensemble methods such as Random Forest Regressor (RFR), Support Vector Regressor (SVR), Linear Regression (LR), Adaptive Boosting Regressor (AdaBoost), Extreme Gradient Boosting Regressor (XGBR), Extra Tree Regressor (ETR), Multi-Layer Perceptron neural network (MLP) and simple Arithmetic Mean (AM) over the diverse geo-climatic basins. Precipitation is best simulated by EC-Earth3 and BCC-CSM2-MR. Maximum temperature by MPI-ESM1-2-HR, EC-Earth3-Veg, INM-CM5-0 and MPI-ESM1-2-LR. Minimum temperature by INM-CM5-0 and MPI-ESM1-2-LR model. The MME of XGBR and RFR stand out for their superior performance across all six basins, with exceptional performance over the per-humid basins, while AdaBoost, SVR and the AM underperform. Examining the interseasonal variability of the simulated MMEs over the basins highlights the reliability of these MME models. The anticipated change in maximum and minimum temperature in the SSP245 and SSP585 in the future horizon corroborates the undeniable rise in temperature by all the MMEs with a dramatic change in future temperature in AM and AdaBoost in precipitation with a factor of two rises in the far future over the recent past. Though climate change is expected to increase precipitation, atmospheric stabilization over the Ghats will affect the spatiotemporal distribution of precipitation. We recommend a comprehensive testing and validation approach to generate ensembles in regional investigations involving complicated and diverse precipitation mechanisms. © 2023 Royal Meteorological Society.
Enhancing soil organic carbon estimation accuracy: Integrating spatial vegetation dynamics and temporal analysis with Sentinel 2 imagery
(Elsevier B.V., 2024) Mruthyunjaya, P.; Shetty, A.; Umesh, P.
This article introduces an improved method for estimating Soil Organic Carbon (SOC) using Sentinel 2 images, with a specific emphasis on the Dakshina Kannada area in India. By examining 364 soil samples, SOC estimation models were constructed using Random forests (RF) and Partial Least Squares Regression (PLSR), focusing on the impact of nearby vegetation pixels. The approach consisted of classifying soil samples by the presence of plant pixels at distances of 0, 10, and 20 m, and evaluating the influence of dry vegetation by the use of the Normalised Burn Ratio 2 (NBR2). The findings demonstrated a significant improvement in the precision of the model (by up to 20 %) when vegetation pixels within a 20-meter radius of the sample locations were omitted. The research also included a temporal analysis utilizing Sentinel-2 images from April 2017 to May 2023. This analysis showed strong relationships between the amount of exposed soil and the accuracy of predicting soil organic carbon (SOC) levels. These results emphasize the need to take into account both the spatial dynamics of vegetation and the temporal variations in bare soil covering to get an accurate estimate of soil organic carbon (SOC). This study improves the accuracy and dependability of SOC evaluations by including geographical and temporal aspects, providing useful insights for agricultural and ecological applications. © 2024 The Author(s)

Faculty Publications

Browse

Filters

Settings

Sort By

Results per page

Search Results