Use of machine learning-based classification algorithms in the monitoring of Land Use and Land Cover practices in a hilly terrain.

The current high rate of urbanization in developing countries and its consequences, like traffic congestion, slum development, scarcity of resources, and urban heat islands, raise a need for better Land Use Land Cover (LULC) classification mapping for improved planning. This study mainly deals with two objectives: 1) to explore the applicability of machine learning-based techniques, especially the Random forest (RF) algorithm and Support Vector Machine (SVM) algorithm as the potential classifiers for LULC mapping under different scenarios, and 2) to prepare a better LULC classification model for mountain terrain by using different indices with combination of spectral bands. Due to differences in topography, shadows, spectral confusion from overlapping spectral signatures of different land cover types, and a lack of access for ground verification, classification in mountainous terrain is difficult task compared to plain terrain classification. An enhanced LULC classification model has been designed using two popular machine learning (ML) classifier algorithms, SVM and RF, explicitly for mountainous terrains by taking into consideration of a study area of Gopeshwer town in the Chamoli district of Uttarakhand state, India. Online-based cloud platform Google Earth Engine (GEE) was used for overall processing. Four classification models were built using Sentinel 2B satellite imagery with 20m and 10m resolutions. Two of these models (Model 'i' based on RF algorithm and Model 'ii' based on SVM algorithm) were designed using spectral bands of visible and infrared wavelengths, and the other two (Model 'iii' based on RF algorithm and Model 'iv' based on SVM algorithm) with the addition of indices with spectral bands. The accuracy assessment was done using the confusion matrix based on the output results. Obtained result highlights that the overall accuracy for model 'i' and model 'ii' were 82% and 86% respectively, whereas these were 87.17% and 87.2% for model 'iii' and model 'iv' respectively. Finally, the study compared the performance of each model based on different accuracy metrics for better LULC mapping. It proposes an improved LULC classification model for mountainous terrains, which can contribute to better land management and planning in the study area.

Assessment of snout analysis of Himalayan glaciers: impact studies on Pindari, Kafni, Sundardhunga, and Baljuri base camp glaciers.

There are several causes for the increasing rate of deglaciation, such as global warming, increase in the concentration of black carbon, and extensive use of fossil fuels which causes the change in the overall climate system and shifting glacier ecosystem. This study was conducted on Pindari valley glaciers part of lesser Himalaya in Uttarakhand. This study investigates to (1) monitor and map change in the frontal length or the snout region of a glacier that can be studied with the help of remote sensing techniques and (2) evaluate the decadal and annual retreat rate of the glacier from 1972 to 2018. The study applies both the maximum likelihood classifier and NDSI spectral indices based classification for extracting the glacier region for different periods. This study reveals a significant amount of retreats taking place in the selected glaciers, Pindari, Sundardhunga, Kafni, and Baljuri base camp glaciers, from 1972 to 2018 as 1719.95 m, 1751.21 m, 1057.01 m, and 810.78 m, respectively. The highest amount of change is noticed in Pindari and Sundardhunga glaciers, higher than ~ 1700 m. The study analyses temporal variation of the annual and decadal retreat rate in the Pindari valley glaciers, which would be helpful for the further study of the other glaciers.

Impact of climate on vegetation in Pindari watershed of Western Himalayas, Kumaun, India, using spatiotemporal analysis: 1972-2018.

Vegetation dynamics is an important aspect for determining climate change trends. The present study delineates to examine spatiotemporal changes of vegetation cover in Pindari valley (Kumaun Himalaya) from the 1972 to 2018 timeline. The study includes the calculation of vegetation spectral indices of normalized vegetation index (NDVI), extraction of different vegetation classes, and statistical analysis of the Mann-Kendall (MK) test on historical metrological data (especially precipitation and temperature) of the study site. For the statistical analysis of metrological data, the power data access viewer datasets have been used. The central feature classes of the study are grassland, scrubland, and forest cover. The results revealed that the region's forest cover significantly decreased by 24.74 sq. km from 1972 to 2018, increased in grassland cover by 17.84 sq. km, respectively, and a slight increase in scrubland class by 3.13 sq. km for the study period. The calculated NDVI shows significant changes over the study location; it has been noticed that the maximum values of the NDVI decreased by 0.24, and the minimum values show growth of about 0.047. The analysis indicates that climatic parameters such as precipitation and temperature are the main limiting factors affecting vegetation growth. The annual mean maximum temperature showed a decreasing trend. The estimated results show an increase in annual rainfall and annual minimum temperature, while the decreasing trend is observed in the case of maximum annual temperature. Objectives of the study are (1) spatiotemporal analysis of the vegetation cover, (2) identification of the main causes of change in the vegetation cover, and (3) statistical trend analysis of long-term metrological data. The outcome of the presented research work would be beneficial for the proper management and monitoring of the forest ecosystem.

Application of remote sensing in alpine grasslands cover mapping of western Himalaya, Uttarakhand, India.

Grasslands are the world's most extensive terrestrial ecosystem, which provides a variety of services for humans, such as carbon storage, food production, crop pollination, pest regulation, and are a major feed source for livestock. However, grasslands are today one of the most endangered ecosystems due to land-use change, agricultural intensification, land abandonment, as well as climate change. Grasslands are an integral part of human societies across the globe, which are broadly known as tropical savannah and temperate grasslands. In the Himalayan region, grasslands are found in more than 55% of the area and different climatic conditions lead to different varieties of grasslands like Danthonia grasslands, kobresia sedge meadow, etc. Grasslands deal with the spatial and temporal distribution of heterogeneous landscapes, which support a high diversity of various species. Owing to very rugged terrain and inaccessibility, the information on the extent of alpine grassland and percent grass cover (%) across the meadows is limited. Therefore, the present attempt was made to assess the current status of grassland in the alpine region of Uttarakhand above 3000 m asl. LANDSAT-8 (OLI and TIRS sensors) satellite data were used to delineate the grasslands using normalized difference vegetation indices (NDVIs) of the alpine region with the help of over 179 ground truth points out of which 50 points are testing points and 129 points are training points. Grass covers (%) were also assessed in the whole alpine region of Western Himalaya of Uttarakhand which nearly consists of over 75 meadows by using random plots (1 × 1 m, total 10 per site) in each meadow. Overall, 89.52% accuracy was achieved based on 50 randomly selected testing points. A total of 4949.25 sq. km area is under the different percentage of grass cover in the alpine region of Uttarakhand, Western Himalaya. Danthonia grasslands below 4000 m and Kobresia sedge meadows above 4000 m elevation are dominant in the state. In the alpine region, over 1056 sq. km grassland area have less than 10% grass cover indicating higher degraded and cold desert areas and only 565.69 sq. km area have more than 60% grass cover, which is highly favorable for rich biodiversity and grazing.