The research, led by Rajib Pegu of Sankardeva Mahavidyalaya with colleagues from institutions across Assam, Arunachal Pradesh, and New Delhi, harnessed the archival power of the Landsat program, which has been imaging the Earth’s surface since the 1970s. By processing multi-temporal Landsat imagery on the Google Earth Engine platform, a cloud-computing service that has revolutionized large-scale environmental analysis, the team was able to classify land cover and retrieve surface temperatures for multiple epochs spanning 1995 to 2024 without downloading or manually processing petabytes of raw data.

The technical pipeline behind the study reflects the current state of the art in remote sensing. Land use and land cover maps were generated using a Random Forest classifier, a machine learning algorithm that builds hundreds of decision trees from training samples and votes on the most probable class for each pixel. Random forest classification has become a workhorse of land cover mapping because it handles noisy, high-dimensional satellite data robustly and resists overfitting. For temperature retrieval, the researchers turned to the radiative transfer approach, converting the thermal infrared radiation measured by Landsat’s sensors into true land surface temperature, with corrections for atmospheric effects and for the emissivity of different surfaces, which they estimated using the Normalized Difference Vegetation Index, a standard proxy for vegetation density.

Land surface temperature is not the same as the air temperature reported by weather stations. Satellites measure the radiative skin of the ground itself: rooftops, asphalt, bare soil, crop fields, forest canopies, and water. Because these surfaces absorb and release heat very differently, land surface temperature is an extraordinarily sensitive indicator of what a landscape is made of. Dark, impervious materials like asphalt and concrete absorb solar radiation during the day and re-emit it slowly, while vegetated surfaces cool themselves through evapotranspiration, the process by which plants release water vapor and shed heat. Water bodies, with their enormous heat capacity, remain the coolest surfaces of all.

The Guwahati data bore this physics out with striking clarity. Built-up areas recorded the highest mean surface temperature of any land cover class, at 36.11 degrees Celsius, while water bodies registered the lowest at 27.29 degrees Celsius, a gap of nearly nine degrees across the same city on the same days. This thermal stratification is the surface signature of the urban heat island effect, the phenomenon first described by Luke Howard in London in the early nineteenth century and quantified for modern cities by climatologist T. R. Oke in the 1970s. What makes the Guwahati study notable is that it documents this effect unfolding in a mid-sized city of the Global South, where urban growth is fastest and adaptive capacity is often thinnest.

To establish that the warming was real and not an artifact of a few unusually hot years, the team applied the Mann-Kendall trend test, a non-parametric statistical method that detects monotonic trends in time series without assuming any particular distribution, together with Sen’s slope estimator, which provides a robust measure of the magnitude of change. The result was a statistically significant warming trend at the conventional p-value threshold of 0.05, confirming that Guwahati’s surface has been heating steadily across the three-decade record rather than merely fluctuating.

The statistical core of the study came from a pixel-level analysis of the 2024 dataset. The researchers drew 5000 randomly sampled pixels and examined how vegetation greenness, expressed through the NDVI, related to surface temperature. Pearson’s correlation analysis revealed a significant inverse relationship, with a correlation coefficient of negative 0.492, meaning that greener pixels were systematically cooler. Linear regression showed that vegetation alone explained approximately 24.2 percent of the spatial variability in surface temperature, a substantial fraction given how many other factors, from elevation to soil moisture to building geometry, also shape the thermal landscape. A one-way analysis of variance confirmed that temperature differences among land cover classes were highly significant, with a p-value below 0.001.

These numbers carry a message that extends well beyond one Indian city. Guwahati sits at the gateway to Northeast India, a region of extraordinary biodiversity, and its growth mirrors that of hundreds of secondary cities across Asia and Africa where urban expansion is consuming wetlands, farmland, and forest fringes. Previous work on Guwahati had already flagged declining groundwater levels and shifting heat island patterns, and studies from Pune, Jaipur, Delhi, Mumbai, and the East Kolkata Wetlands have documented similar dynamics. The new study adds a rigorous, three-decade quantitative baseline that links landscape change to thermal change in a single coherent framework.

The implications for planning are direct. Because vegetation explains roughly a quarter of the spatial variation in surface temperature, protecting and expanding urban green cover is not merely an aesthetic preference but a measurable cooling strategy. The authors point to the conservation of urban vegetation, wetlands, and other natural landscapes as priorities, and frame their findings as evidence for climate-responsive urban planning aligned with the United Nations Sustainable Development Goals on sustainable cities, climate action, and life on land. In a warming century, the coolest surfaces a city owns are the ones it has not yet paved.

The study also demonstrates how the tools of planetary observation have been democratized. A team of researchers at regional Indian institutions, using free Landsat archives and the Google Earth Engine platform, produced an analysis that would once have required a supercomputing facility and a large budget. As urbanization accelerates across the Global South, this kind of accessible, reproducible satellite monitoring offers city governments a way to see, in near real time, the thermal consequences of every hectare of green space lost, and every hectare of concrete gained.

Subject of Research: The relationship between land use land cover change and land surface temperature dynamics in Guwahati, India, analyzed with Landsat imagery and Google Earth Engine

Article Title: Urbanization-driven thermal changes: a remote sensing approach to land surface temperature (LST) and land use land cover (LULC) dynamics in Guwahati, Assam, India

Article References: Pegu, R., Mazumder, T., Borah, H., Bora, K., & Hazarika, P. (2026). Urbanization-driven thermal changes: a remote sensing approach to land surface temperature (LST) and land use land cover (LULC) dynamics in Guwahati, Assam, India. Environmental Science and Pollution Research, 33(28), 14266-14281. https://doi.org/10.1007/s11356-026-38169-x

Image Credits: AI Generated

DOI: 10.1007/s11356-026-38169-x

Keywords: land surface temperature, land use land cover change, urban heat island, Guwahati, remote sensing, Landsat, Google Earth Engine, random forest classification, NDVI, urbanization, Mann-Kendall test, Assam

Amchang occupies a precarious position in every sense. Declared a wildlife sanctuary in 2004, it sits on the southern bank of the Brahmaputra River, hemmed in by Guwahati’s sprawl to the west and rural agricultural land to the south and east. It forms part of the Indo-Burma biodiversity hotspot, one of only 34 such regions recognised worldwide, and shelters endangered species including the Asian elephant, the hoolock gibbon and the leopard. Its vegetation ranges from tropical moist deciduous and semi-evergreen forest to bamboo brakes and grassland, dominated by trees such as sal, teak and Terminalia species. Yet the sanctuary also carries a heavy burden of biological invasion: earlier surveys have recorded 57 invasive plant species from 22 families, with more than 47 percent of the total area invaded by the four most common invaders and roughly 27 percent of that invaded zone severely impacted.

To track three decades of change, the research team led by Trishna Changkakoti of The Assam Royal Global University assembled multi-temporal Landsat imagery at 30-metre resolution, drawing on Landsat 5 TM data from 1995, Landsat 7 ETM+ from 2005, and Landsat 8 OLI/TIRS data for 2015 and 2025. Images were selected from the dry season months of February and March to minimise cloud cover and phenological differences between years. All processing, from pre-processing through classification, was carried out in Google Earth Engine and ArcGIS 10.8, with radiometric correction converting raw digital numbers into surface reflectance values so that the multi-temporal datasets could be compared consistently. The team then layered in three spectral indices designed to sharpen the contrast between land cover types: the Normalised Difference Vegetation Index for vegetation greenness, the Normalised Difference Water Index for surface water, and the Normalised Difference Built-up Index for urban surfaces.

The heart of the method was a Random Forest classifier, a machine learning algorithm that builds hundreds of decision trees from randomly sampled training data and assigns each pixel a class by majority vote. Nine predictor variables were used, combining six Landsat spectral bands with the three indices, and the classifier was configured with 100 decision trees. Reference points gathered from field surveys and high-resolution Google Earth imagery were split 70:30 into training and validation subsets. The resulting maps distinguished seven land use and land cover categories: dense forest, moderately dense forest, open forest, agriculture, built-up land, barren land and water bodies. Accuracy assessment using an independent confusion matrix yielded an overall accuracy of 81.57 percent and a Kappa coefficient of 0.78, figures the authors describe as good, though slightly below the conventional threshold of 0.80 that would indicate near-excellent agreement.

The classified maps tell a story of two opposing processes running in parallel. In 1995, open forest dominated the sanctuary at 45.4 percent of the area, with moderately dense forest at 21.8 percent and dense forest at 10.5 percent. By 2025, dense forest had fallen to just 5.1 percent and open forest to 38.9 percent, while built-up land climbed from 9.8 to 13.6 percent. Yet moderately dense forest expanded substantially, rising by 35.5 percent or 6.11 square kilometres, a pattern the researchers attribute to secondary succession, the regrowth of woody vegetation on land where pressure has eased or where former agricultural plots have been abandoned. This regenerative pulse, however, cannot compensate for the losses elsewhere. Water bodies, which covered 0.33 percent of the sanctuary in 1995, had dwindled to a mere 0.08 percent by 2025, a hydrological decline the authors link to the broader disturbance regime squeezing the reserve.

Beyond simple area statistics, the study deployed landscape ecology metrics to capture how the sanctuary’s structure, not just its extent, has changed. Shannon’s Diversity Index, which measures the variety and evenness of land cover classes, fluctuated across the period, dipping to 1.36 in 2005 before rising again to 1.49 by 2025. Patch count remained constant at seven discrete patches, but edge length told a more troubling tale: after falling from 68,958 metres in 1995 to 58,518 metres in 2005, total edge length surged to 70,749 metres by 2025. Longer edges mean more of the forest interior is exposed to external pressures such as grazing, fuelwood extraction, encroachment and microclimatic stress, a phenomenon ecologists call the edge effect. The authors interpret the combination of stable patch count and rising edge length as evidence of a landscape that is structurally fragmented yet ecologically dynamic, alternating between degradation and partial recovery.

To determine whether these changes represented genuine long-term trends or mere fluctuation, the team applied two complementary statistical tools. The Theil-Sen slope estimator, which calculates the median of all pairwise slopes in a time series, provides a robust, outlier-resistant measure of the magnitude of change, while the Mann-Kendall test assesses the statistical significance of monotonic trends using Z-statistics and p-values. The pixel-level analysis was unambiguous: 2,748 pixels showed significant decreasing trends between 1995 and 2025, more than double the 1,267 pixels showing positive trends, with 739 pixels remaining stable. In other words, despite pockets of regeneration, the sanctuary’s overall trajectory is one of degradation, and the authors argue that rapid intervention is required to manage the losses.

The spatial pattern of change points to specific drivers. The decline in dense forest and water bodies aligns with encroachment for settlement and agricultural expansion along the sanctuary’s western boundary, where it adjoins Guwahati, while the concentration of new built-up land near transport corridors implicates infrastructure development as a principal engine of forest loss at the periphery. The magnitude of the losses also stands out when set against comparable research: regional assessments in Northeast India have typically reported broad forest declines of roughly 9 to 10 percent, whereas Amchang’s 51.7 percent dense-forest loss over three decades far exceeds those figures, underscoring the severity of unchecked urban expansion and weak buffer-zone enforcement. The authors frame their findings within the Sustainable Development Goals, linking sustainable urban planning to SDG 11, forest conservation and climate mitigation to SDG 13, and ecosystem restoration to SDG 15.

What makes the study methodologically distinctive is its integration of three analytical strands, Random Forest classification, landscape fragmentation metrics and non-parametric trend statistics, within a single framework applied to a peri-urban protected area over 30 years. Because the entire workflow relies on freely available Landsat imagery and open-access processing in Google Earth Engine, the authors argue it can be readily transferred to other protected areas in data-poor settings facing similar pressures at the urban edge. The study is not without limitations: the 30-metre resolution of Landsat constrains the detection of fine-scale fragmentation, dry-season imagery cannot capture seasonal forest dynamics, and the classification accuracy, while respectable, leaves residual uncertainty that propagates into the fragmentation and trend analyses. The authors recommend future work with higher-resolution satellite data and deep learning classifiers. For now, the message from the satellites is clear: without stronger buffer-zone compliance and better land use planning around Guwahati, one of Northeast India’s most biodiverse sanctuaries will continue to erode at its edges, one pixel at a time.

Subject of Research: Three-decade geospatial assessment of land use and land cover change in a peri-urban protected area of Assam, India

Article Title: Spatio temporal land use and land cover changes in a peri urban protected area of Kamrup Metropolitan District, Assam, India using geospatial techniques

Article References: Changkakoti, T., Saikia, P., Borthakur, P., & Roy, M. (2026). Spatio temporal land use and land cover changes in a peri urban protected area of Kamrup Metropolitan District, Assam, India using geospatial techniques. Discover Forests, 2(1), Article 75. https://doi.org/10.1007/s44415-026-00134-4

Image Credits: AI Generated

DOI: 10.1007/s44415-026-00134-4

Keywords: land use land cover change, remote sensing, Random Forest classification, forest fragmentation, peri-urban protected area, Amchang Wildlife Sanctuary, Guwahati, Landsat, landscape metrics, Mann-Kendall test, biodiversity hotspot, urbanization