Kailash Chandra Tiwari

Understanding the intricate relationship between land use/land cover (LULC) transformations and land surface temperature (LST) is critical for sustainable urban planning. This study investigates the spatiotemporal dynamics of LULC and LST across Delhi, India, using thermal data from Landsat 7 (2001), Landsat 5 (2011) and Landsat 8 (2021) resampled to 30-m spatial resolution, during the peak summer month of May. The study aims to target three significant aspects: (i) to analyse and present LULC-LST dynamics across Delhi, (ii) to evaluate the implications of LST effects at the district level and (iii) to predict seasonal LST trends in 2041 for North Delhi district using the seasonal auto-regressive integrated moving average (SARIMA) time series model. LULC classification is performed using the random forest (RF) approach, and the LULC-LST relationships are statistically examined using a one-way ANOVA paired with Tukey’s HSD post hoc test, Pearson’s correlation and simple linear regression analysis. The SARIMA model is employed to predict district-level LST for North Delhi in 2041. The results highlight significant LULC-driven LST variations across Delhi, with urban expansion and post-harvest agriculture lands contributing to the temperature increase. Bare land and urban areas exhibited the highest LST, while vegetation and waterbodies consistently recorded lower temperatures. The surface urban heat island intensity was predominantly pronounced in densely built-up areas including the Aerocity region. The district-level analysis reveals substantial spatial heterogeneity, with western districts predominated by agriculture and major urban expansion recording the highest LST, while the southern-central districts experienced lower temperatures due to influence of floodplains, vegetation and ridges. The correlation analysis demonstrated a strong positive association between urban expansion and LST (r = 0.98), while the one-way ANOVA test indicated significant differences in LST across LULC classes (F(4, 148,238) = 9646, p < 0.05, n = 3). The SARIMA-based projections for North Delhi predict escalating temperatures in all seasons by 2041, with a root mean squared error of 2.1. The finding emphasises the need for adaptive urban planning, advocating for strategic integration of vegetation buffers around industrial zones, landfills and along agriculture-urban interfaces to mitigate heat and inform future urban development policies such as the Master Plan Delhi 2041.

The United Nations (UN) emphasizes the pivotal role of sustainable agriculture in addressing persistent starvation and working towards zero hunger by 2030 through global development. Intensive agricultural practices have adversely impacted soil quality, necessitating soil nutrient analysis for enhancing farm productivity and environmental sustainability. Researchers increasingly turn to Artificial Intelligence (AI) techniques to improve crop yield estimation and optimize soil nutrition management. This study reviews 155 papers published from 2014 to 2024, assessing the use of machine learning (ML) and deep learning (DL) in predicting soil nutrients. It highlights the potential of hyperspectral and multispectral sensors, which enable precise nutrient identification through spectral analysis across multiple bands. The study underscores the importance of feature selection techniques to improve model performance by eliminating redundant spectral bands with weak correlations to targeted nutrients. Additionally, the use of spectral indices, derived from mathematical ratios of spectral bands based on absorption spectra, is examined for its effectiveness in accurately predicting soil nutrient levels. By evaluating various performance measures and datasets related to soil nutrient prediction, this paper offers comprehensive insights into the applicability of AI techniques in optimizing soil nutrition management. The insights gained from this review can inform future research and policy decisions to achieve global development goals and promote environmental sustainability.

Over the past half-century, numerous lakes have surfaced in various alpine regions worldwide, posing a significant threat. This phenomenon occurs as lakes expand from melting glaciers, driven by global warming, leading to potential bursting and extensive damage. India, with a concentration of glaciers in Ladakh, Sikkim, Himachal Pradesh, and Uttarakhand, along with a few in Arunachal Pradesh, faces such challenges. The 2013 glacial lake outburst flood (GLOF) in Uttarakhand impacted over 100,000 lives, caused thousands of deaths, and inflicted substantial infrastructural damage. This study delves into the risk assessment of potentially hazardous glacier lakes in the Alaknanda River basin in Uttarakhand, India. Identification of vulnerable or perilous lakes utilizes metrics such as surface area, position relative to the parent glacier, slope, distance from the basin’s outlet, and the presence of downstream lakes. The initial focus involves accurate mapping of glaciers and glacial lakes in the study area, employing multispectral satellite images from Landsat satellite data. For glacier mapping, we utilized the Normalized Difference Snow Index (NDSI) image, Band Ratio, and a slope map of the area. However, the NDSI and Band Ratio methods proved insufficient in differentiating the extent of debris -covered glacier ice due to similar spectral signatures from the surrounding debris. Specifically, NDSI (GREEN/SWIR) and Band Ratio (VIS/NIR) misclassified proglacial lakes in threshold glacier areas. On the other hand, Band Ratio NIR/SWIR accurately classified clean glacier ice with partly glacier ponds. Hence, Band Ratio NIR/SWIR emerges as a more suitable method for identifying clean glacier ice compared to NDSI (GREEN/SWIR). Moreover, Band Ratio NIR/SWIR outperforms NDSI (GREEN/SWIR) in shadow areas. In this work, manual identification of glaciers involved using a slope map in conjunction with the NDSI technique. The text describes the process of glacial lake mapping in the Alaknanda basin using methods such as NDWI, NDSI, and unsupervised classification. The results were overlaid on Google Earth to identify 463 glacial lakes. Manual editing was then performed using Google Earth images. Subsequently, a comprehensive inventory of the glacial lakes in the Alaknanda basin was created, highlighting four potentially dangerous lakes based on specific factors. Further investigation In ArcGIS 10.1, various metrics were computed, including volume and area, along with other parameters like cross-sections from the lake to the outlet. Consequently, the area of the potentially hazardous Lake 196 is 0.231 square km, with a volume of 508.315 cubic meters. The distance of this lake from the outlet of the basin is 1.372 km.

Hyperspectral data present better opportunities to exploit the treasure of spectral and spatial content that lies within their spectral bands. Hyperspectral data are increasingly being considered for exploring levels of urbanization, due to their capability to capture the spectral variability that a modern urban landscape offers. Data and algorithms are two sides of a coin: while the data capture the variations, the algorithms provide suitable methods to extract relevant information. The literature reports a variety of algorithms for extraction of urban information from any given data, with varying accuracies. This article aims to explore the binary-classifier approach to target detection to extract certain features. Roads and roofs are the most common features present in any urban scene. These experiments were conducted on a subset of AVIRIS-NG hyperspectral data from the Udaipur region of India, with roads and roofs as targets. Four categories of target-detection algorithms are identified from a literature survey and our previous experience—distance measures, angle-based measures, information measures, and machine-learning measures—followed by performance evaluation. The article also presents a brief taxonomy of algorithms; explores methods such as the Mahalanobis angle, which has been reported to be effective for extraction of urban targets; and explores newer machine-learning algorithms to increase accuracy. This work is likely to aid in city planning, sustainable development, and various other governmental and nongovernmental efforts related to urbanization.

Regular monitoring and precise mapping of urban environment is required by various applications. In this paper, a new method is proposed, in which different combinations of feature bands have been utilized for extraction of built-up surfaces, sub-surfaces and materials in different levels (Level-1, 2 and 3) using AVIRIS-NG hyperspectral imagery of Jodhpur, Rajasthan region of India. Features identified in this study are based on spectral indices, major principal components and fractional abundances, in which first combination is developed using spectral indices and fractional abundances while second is made using spectral indices and major principal components and finally third using combination of all the aforesaid features. It is observed that the combined form of all the aforementioned features produces better extraction results than the other two while combination of spectral indices and fractional abundances may be more useful than the combined form of spectral indices and major principal components.