Home
Issues List
1 Sep 06, 2023
A DEEP LEARNING APPROACH FOR MONSOON FORECASTING: UNLEASHING THE POWER OF NEURAL NETWORKS

Accurate monsoon forecasting plays a critical role in numerous sectors, including agriculture, water management, and disaster preparedness. This research paper introduces a deep learning approach for monsoon forecasting, harnessing the power of neural networks. We explore the capabilities of neural networks in capturing complex spatiotemporal patterns present in monsoon data, facilitating more precise and reliable predictions. Our proposed approach utilizes a combination of convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to effectively model both spatial and temporal dependencies in monsoon datasets. We conduct extensive experiments and comparative analyses with existing forecasting methods to demonstrate the effectiveness of our approach. The results showcase the potential of deep learning in enhancing monsoon forecasting accuracy, empowering improved decisionmaking, and planning for monsoon-related event...

Authors: Bheema Shanker Neyigapula.

2 Sep 06, 2023
MULTISPECTRAL IMAGE ANALYSIS OF LANDUSE AND LANDCOVER CHANGE AND CHRONOSEQUENCE ASSESSMENT OF SOIL ORGANIC CARBON IN NATURALLY RECLAIMED OVERBURDEN DUMP IN THE RANIGANJ COAL FIELD AREA, WEST BENGAL

The current study focuses on analysing landuse and landcover (LULC) changes using remote sensing and GIS, as well as carbon sequestration in naturally reclaimed overburden dump soil of an open cast coal-mine in Sonepur-Bazari, Raniganj Coalfield area, Bardhaman district, West Bengal by estimating soil organic carbon (SOC). The landuse/landcover changes over 20 years (1999-2019) due to expansion of mining and identification of overburden dump sites have been carried out using Sentinel 2B MSI and Landsat TM 5 data. The SOC in OB dump soil has been measured in chronosequence (one year, three years, five years, ten years, fifteen years, and twenty years) at two depth profiles (< 15 cm and 15-30 cm) and is compared with those of the adjoining wasteland site. The study indicates that the SOC content of the soil of the recent OB dump at <15 cm depth is 9.1 megagram per hectare or 16.70 megagram CO2 per hectare. At the same depth of a 20-year old OB dump the SOC content is 21.6 megagram per hectare (39.64 megagram CO2 per hectare). Soil samples collected from depth between 15 and 30 cm the SOC increases from 8.3 to 19 megagram per hectare (15.23 to 34.86 megagram CO2 per hectare) over twenty years’ time. The SOC content in the adjoining wasteland site is 17.7 megagram per hectare (32.48 Mg CO2 per hectare) and 17.2 megagram per hectare (31.56 megagram CO2 per hectare) at depths < 15 cm and 15 to 30 cm respectively. Therefore, initially the SOC is less in recent OB dump compared to that of wasteland but within 20 years the naturally reclaimed OB dump accumulates SOC and it is 3.9 megagram per hectare (7.16 megagram CO2 per hectare) and 1.8 megagram per hectare (3.30 megagram CO2 per hectare) more than that of the wasteland at <15 cm and at 15-30cm depths respectively. This clearly indicates that SOC sequestration is dependent on biomass growth. Therefore, a scientific OB dump management program will increase the rate of carbon sequestration which will help in combating climate change at a local level....

Authors: Surajit Chakraborty, Rohit Basu Dhar, Pradip K.Sikdar , Suranjan Sinha.