Search

Weeds are unwanted vegetation that compete with main crops for essential resources like light, water, and nutrients, leading to significant reductions in food crop yield and economic losses. Addressing this issue is crucial, particularly during the Kharif cropping season when cloud cover interferes with remote sensing capabilities. This study is an attempt to investigate the potential of satellite-based temporal analysis in weed detection from agricultural fields. The research focused on rice cultivation at the Research cum Instructional farms of Indira Gandhi Krishi Vishwavidyalaya, Raipur, Chhattisgarh. The study explored the utility of satellite imagery for assessing crop health, demonstrating how weed infestation influences vegetative indices. The study utilized satellite images from PlanetScope and Sentinel-2 to examine the temporal variation in vegetation indices across two treatments: pure rice and rice with weeds. NDVI analysis revealed a significant decline in treatments affected by weeds (upto 41% less), suggesting that time-series satellite data can serve as an early indicator of weed infestation in standing rice crops. These findings were further verified by backscatter values from the Sentinel-1 dataset, which indicated a reduction in backscatter (upto 18% less) due to the suboptimal growth conditions in weed-infested treatments compared to weed-free rice. While the technology has shown efficacy at a preliminary stage, there is significant potential for its broader application and scalability in operational contexts.

Urban Canopy Parameters (UCPs) are crucial for urban microclimate modeling; however, the scarce availability of precise UCP data in developing regions limits their application for urban climates. This study investigated the use of multi-platform remote sensing data viz. very high-resolution satellite (VHRS) optical stereo and Unmanned Aerial Vehicle (UAV) datasets for the computation of UCPs in high-density urban scenarios in India, with varied development characteristics. The results demonstrated high accuracy in terms of building height and footprint extraction from both datasets, key inputs for UCP computation. However, UCPs from UAV data have displayed relatively high accuracy for building footprints (86%), building height (RMSE ~ 0.05 m), and land use/land cover classification (90%). Performance evaluation of computed UCPs against a 3D reference geodatabase showed high prediction accuracy for most UCPs, with overall biases, mean absolute error, and root-mean-square error values significantly better than 1 m, with strong correlation (0.8–0.9). It was concluded that VHRS optical stereo and UAV datasets offer a secure, reliable, and accurate solution for UCP computation in urban areas, particularly in developing regions. These findings have significant implications for urban climate research and the sustainable development of rapidly urbanizing areas facing resource and policy constraints.

The satellite-based nighttime lights (NTL) data from the Defense Meteorological Satellite Program’s Operational Linescan System (DMSP-OLS), available in the public domain from 1992 to 2013, are extensively used for socio-economic studies. The improved NTL products from the Visible Infrared Imaging Radiometer Suite’s Day/Night Band (VIIRS-DNB), on-board the Suomi National Polar-Orbiting Partnership spacecraft and National Oceanic and Atmospheric Administration – 20 (NOAA-20) spacecraft’s, are now available since April 2012. This study investigates the potential of machine-learning algorithms for inter-calibrating them (i.e., DMSP-OLS and VIIRS-DNB) to produce time-series annual VIIRS-DNB-like NTL datasets for the time when VIIRS-DNB data did not exist, for long-term studies. Uttar Pradesh, one of the most populous and largest States of India, is selected as the study area. Two machine-learning algorithms are utilized: (1) Multi-Layer Perceptron (MLP), having deep neural networks (DNN) architecture, and (2) Random Forest (RF), a widely used method. The DMSP-OLS and VIIRS-DNB data of 2013 (common year of data availability) and ancillary data pertaining to land cover, topography, and road network are used to train the models. The qualitative and quantitative analysis of annual VIIRS-DNB-like NTL images simulated from annual DMSP-OLS composites of 2004–2012 indicates that RF captures better spatial details at the local-scale and is able to efficiently handle the saturation problem at urban centers; while MLP is found to be superior at regional-scale. Both MLP and RF models significantly reduce the blooming effect around settlements, a common problem observed in DMSP-OLS data. It is inferred that depending on the research objectives, both RF and MLP algorithms can be appropriately utilized for producing VIIRS-DNB-like NTL images from DMSP-OLS annual NTL composites. The research can be further expanded by using other DNN architecture-based algorithms and improved spatio-temporal ancillary datasets over areas with different socio-economic, physiographic, and climatic settings.

The LULC change vis-à-vis climate change inherently encompassing human dimensions consequently impact hydrological processes. A slight change in it may affect the water yield, as both are explicitly linked through various hydrological processes. The future availability of water resources largely depends upon planning and management of land use in this changing environment. However, the continuous human interactions keep on modifying the land use land cover (LULC) to fulfil the enhanced demand especially due to significant increase in population and development towards better facilities. These changes consequently impact each and every hydrological process vis-à-vis water availability. It has now become a major concern for water resources planners and managers. Moreover, these changes in LULC pattern along with changing climate put forth a huge challenge in front of them. The present study investigates the capabilities of Variable Infiltration Capacity (VIC) model, which has been developed as soil-vegetation-atmosphere transfer schemes, to assess the runoff potential. Initially, the required model inputs were derived and simulated at the national scale. It was found that the model has assessed the runoff potential of major river basins with high accuracy. The successful implication of the model at the national scale offered an opportunity to carry out a detailed hydrological simulation to study the impact of LULC change on hydrological regime of a basin. For the analysis, the Pennar River Basin, located in Southern part of India, was selected. The LULC change detection analysis showed that there is human induced alteration to LULC pattern, as the area under urban landscape is increasing (0.14%) and natural forest is declining (0.7%). These changes resulted in increase of runoff potential by around 45%. The study on Indian River basin clearly indicates that the LULC change influence and alter the hydrological regime of the river basin. Keywords: Land use land cover change, Runoff potential, Hydrological modeling, Variable infiltration capacity model, Impact assessment

Many real-world applications require remotely sensed images at both high spatial and temporal resolutions. This requirement, however, is generally not met by single satellite system. A number of spatiotemporal fusion models have been developed to overcome this constraint. Landsat and Visible Infrared Imaging Radiometer Suite (VIIRS) data have been extensively used for detection and monitoring of active fires at different scales. Fusing the data obtained from these sensors will, therefore, significantly contribute to the satellite-based monitoring of fires. Among the available spatiotemporal fusion methods, the spatial and temporal adaptive reflectance fusion model (STARFM) and enhanced STARFM (ESTARFM) algorithms have been widely used for studying the land surface dynamics in the homogeneous and heterogeneous regions. The present study explores the applicability of STARFM and ESTARFM algorithms for fusing the high spatial resolution Landsat-8 OLI data with high temporal resolution VIIRS data in the context of active surface coal fire monitoring. Further, a modified version of ESTARFM algorithm, referred as modified-ESTARFM, is developed to improve the performance of the fusion model. Jharia coalfield (India), known for widespread occurrences of coal fires, is taken as the study area. The qualitative and quantitative assessments of the predicted (synthetic) Landsat-like images from different algorithms (STARFM, modified-STARFM, ESTARFM, modified-ESTARFM) indicate that the modified-ESTARFM outperforms the other fusion approaches used in this study. Considering the advantages, limitations and performance of the algorithms used, modified-ESTARFM along with STARFM can be used for surface coal fire monitoring. The study will not only contribute to remote sensing based coal fire studies but also to other applications, such as forest fires, crop residue burning, land cover and land use change, vegetation phenology, etc.

Machine learning classifiers are being increasingly used nowadays for Land Use and Land Cover (LULC) mapping from remote sensing images. However, arriving at the right choice of classifier requires understanding the main factors influencing their performance. The present study investigated firstly the effect of training sampling design on the classification results obtained by Random Forest (RF) classifier and, secondly, it compared its performance with other machine learning classifiers for LULC mapping using multi-temporal satellite remote sensing data and the Google Earth Engine (GEE) platform. We evaluated the impact of three sampling methods, namely Stratified Equal Random Sampling (SRS(Eq)), Stratified Proportional Random Sampling (SRS(Prop)), and Stratified Systematic Sampling (SSS) upon the classification results obtained by the RF trained LULC model. Our results showed that the SRS(Prop) method favors major classes while achieving good overall accuracy. The SRS(Eq) method provides good class-level accuracies, even for minority classes, whereas the SSS method performs well for areas with large intra-class variability. Toward evaluating the performance of machine learning classifiers, RF outperformed Classification and Regression Trees (CART), Support Vector Machine (SVM), and Relevance Vector Machine (RVM) with a >95% confidence level. The performance of CART and SVM classifiers were found to be similar. RVM achieved good classification results with a limited number of training samples.