Identify Land Cover Research Articles

Assessing Land-Use and Land-Cover Change (LULCC) Between 2009 and 2019 Using Object-Based Image Analysis (OBIA) in Cameron Highlands, Malaysia

Cameron Highlands has witnessed multiple land encroachment activities as well as repeated deforestation leading to extensive land-use and land-cover change (LULCC) over the last 6 decades. This is due to the heightened increase in demand for urban growth as well as agricultural activities and tourism industries. This study aims to identify land cover classes and assess their changes in Cameron Highlands between 2009 and 2019. Geographical information system (GIS) and remote sensing techniques were used to process and analyse Landsat 7 and Landsat 8 imageries with the Object-based Image Analysis (OBIA) classification technique. The result showed that the study area experienced large changes in both agriculture and urban, as well as primary forests. LULC change for agriculture nearly tripled in 10 years from 4.93% in 2009 to 12.63% in 2019. Moreover, urban development increased from 7.48% in 2009 and 9.12% in 2019. However, the expense of this is at the cost of primary forests declining by 59.44 km2 (8.87%) of total land area from 2009 until 2019. The overall accuracy of the classification achieved 94.6%. This study would be able to provide suggestions and a basis for LULCC as well as forest monitoring in other mountainous regions of the world.

Machine Learning Classification Ensemble of Multitemporal Sentinel-2 Images: The Case of a Mixed Mediterranean Ecosystem

Land cover type classification still remains an active research topic while new sensors and methods become available. Applications such as environmental monitoring, natural resource management, and change detection require more accurate, detailed, and constantly updated land-cover type mapping. These needs are fulfilled by newer sensors with high spatial and spectral resolution along with modern data processing algorithms. Sentinel-2 sensor provides data with high spatial, spectral, and temporal resolution for the in classification of highly fragmented landscape. This study applies six traditional data classifiers and nine ensemble methods on multitemporal Sentinel-2 image datasets for identifying land cover types in the heterogeneous Mediterranean landscape of Lesvos Island, Greece. Support vector machine, random forest, artificial neural network, decision tree, linear discriminant analysis, and k-nearest neighbor classifiers are applied and compared with nine ensemble classifiers on the basis of different voting methods. kappa statistic, F1-score, and Matthews correlation coefficient metrics were used in the assembly of the voting methods. Support vector machine outperformed the base classifiers with kappa of 0.91. Support vector machine also outperformed the ensemble classifiers in an unseen dataset. Five voting methods performed better than the rest of the classifiers. A diversity study based on four different metrics revealed that an ensemble can be avoided if a base classifier shows an identifiable superiority. Therefore, ensemble approaches should include a careful selection of base-classifiers based on a diversity analysis.

Forest cover and open land drive the distribution and dynamics of the breeding sites for urban-dwelling Northern Goshawks

Land cover change accompanying large-scale and rapid urbanization has caused major losses to biodiversity, with urban planning that incorporates biodiversity conservation being urgently needed. Understanding how land cover impacts the distribution and dynamics of breeding sites of biodiversity indicator species could facilitate the creation of long-term, landscape-scale urban planning to enhance biodiversity. Here, we identified land cover factors that influence the distribution and dynamics of the breeding sites of goshawks, a biodiversity indicator species, in the urban areas of Kanagawa, Japan. We analyzed how breeding site presence and absence data were correlated with land cover factors using a logistic regression model. To evaluate the dynamics in breeding site distribution, a meta-population dynamics model was used to analyze the relationship between breeding site occupancy history and land cover factors over 6 years. Forest cover, open land, and highly vegetated urban cover positively affected the distribution of goshawk breeding sites. However, the relative importance of highly vegetated urban cover was not as high as forest cover or open land. Forest cover negatively affected local extinction probability, while forest cover and open areas positively affected local colonization probability. In comparison, highly vegetated urban cover had low relative importance for the distribution of breeding sites, with no relationship being detected for breeding site distribution dynamics. The importance of forest cover and open land for both the distribution and dynamics of goshawk breeding sites emphasizes the importance of these land cover types for the breeding of urban-dwelling goshawks. Therefore, managers should focus on maintaining forests and open land to conserve goshawk breeding sites in urban areas to facilitate their long-term viability. Highly vegetated urban cover was considered to be less important for the long-term viability of goshawk breeding sites.

Predictive Analytics for Identifying Land Cover Change Hotspots in the Mekong Region

Understanding land cover change dynamics and potential pathways of change is of critical importance for sustainable resource management, to promote food security and resilience on a range of spatial scales. Data scarcity is a key concern, however, with the availability of free Earth Observation (EO) data, such challenges can be suitably addressed. In this research we have developed a robust machine learning (random forest) approach utilizing EO and Geographic Information System (GIS) data, which enables an innovative means for our simulations to be driven only by historical drivers of change and hotspot prediction based on probability to change. We used the Mekong region as a case study to generate a training and validation sample from historical land cover patterns of change and used this information to train a random forest machine learning model. Data samples were created from the SERVIR-Mekong land cover data series. Data sets were created for 2 categories both containing 8 classes. The 2 categories included—any generic class to change into a specific one and vice versa. Classes included the following: Aquaculture; Barren; Cropland; Flooded Forest; Mangroves; Forest; Plantations; Wetlands; and Urban. The training points were used to sample a series of satellite-derived surface reflectance products and other data layers such as information on slope, distance to road and census data, which represent the drivers of change. The classifier was trained in binary mode and showed a clear separation between change and no change. An independent validation dataset of historical change pixels show that all median change probabilities are greater than 80% and all lower quantiles, except one, are greater than 70%. The 2018 probability change maps show high probabilities for the Plantations and Forest classes in the ‘Generic to Specific’ and ’Specific to generic’ category, respectively. A time-series analysis of change probability shows that forests have become more likely to convert into other classes during the last two decades, across all countries. We successfully demonstrated that historical change patters combined with big data and machine learning technologies are powerful tools for predictive change analytics on a planetary scale.

Temporal-Spatial Simulation of Landscape Variations Using Combined Model of Markov Chain and Automated Cell

Classification and preparation of land use map are the most commonly used methods in remote sensing data. Numerous advanced classification methods have been developed in a recent year. Among those methods, one can name support vector machine. Identifying land cover changes can play an important role in future decisions of regional managers. Therefore, the present study aimed at employing CA–Markov model to present a powerful but simple model for simulation and prediction of Yazd city. Accordingly, first, satellite images of 2000, 2005, 2010, and 2016 were classified by SVM method, and then land use maps of Yazd city were extracted. The results of the detection of changes indicated the expansion of residential areas and the reduction of dry land and vegetation during a 16-year period. Prediction of land use development in 2040 using CA–Markov model with a kappa coefficient of 84.43 shows the high accuracy of this model. Besides, the urban land use area shows 18,432 hectares increase in 2040 compared to 2010, while the area of dry land uses (− 15,570) rocky areas (− 186) and vegetation (− 2658) hectares decrease compared to 2010. Moreover, the results of CA–Markov model indicated the development of Yazd city by degradation of vegetation.

Change detection in city of Hilla during period of 2007-2015 using Remote Sensing Techniques

This study was carried out in order to identify and map the temporal changes in the land covers in Hilla province using Remote Sensing and GIS. Two images were used of land sate taken in Quick Bird Satellite acquired for the ETM sensor taken on 15 April 2015 and 15 April 2007 with spatial accuracy (0.6 m). These were used as reference sources in the selection of some classification samples after matching them with the Landsat 8 above. In this research using topography maps of the study area (Hilla) with scale 1/10000 generated from the General Authority as a reference data and achieved compatibility with the image of space after it was corrected geometrically 1976, 2000 and 2014. In this research using Geographic Information System (GIS) software with projection images WGS84 process (based on the projection of the global map and the global geodetic system) and within the 38th north. Supervised classification was used to identify land cover classes dominated in the study area. Change detection results showed an increase in the urban areas, in the year 2015 the area had become 1380.350-hectare unit, while in the year 2007 the area was 318.562-hectare unit. Similarly, the vegetation had increased between 2007-2015 from (132.531 to 507.296) hectare unit. Results indicated that urban land and salinization process can be considered as the important phenomena which positively affected the agriculture area. So, the increase in the cover of water is an important reason of decrease in the arid land, which was due to the increase of urban land reach to 1061.788-hectare unit. Many changes have happened in the quantity of reclaimed land in these arid lands between the two years (2007 and 2015).

Land cover types and changes in land use in Republic of Srpska (Bosnia and Herzegovina) over the period 2000-2018

Republic of Srpska (RS) covers an area of 24,666 km2 or about 48.5% of Bosnia and Herzegovina’s territory. Spatial and environmental changes accelerated due to dynamic historical period upon the RS formation and the following development period. The extent of these changes can be determined by comparing satellite images obtained from different acquisition periods. By processing images, with the support of geographic information systems, it is possible to create a database that aims to analyze spatial processes in a specific area in order to determine quantitative and qualitative parameters. In this way, the trends of spatial development (e.g. excessive logging, land cover damage, water pollution, etc.) and potentially vulnerable components of the environment can be monitored. The data for Bosnia and Herzegovina were collected from CORINE Land Cover (CLC) database for 39 European countries. The CLC is a project launched by the European Environment Agency (EEA) more than thirty years ago with the aim of collecting, coordinating and ensuring the consistency of information on natural resources and the environment. The goal of the paper is to identify land cover types and determine environment changes in the territory of Republic of Srpska over the period 2000–2018 as a consequence of the land use conversion.

Local climate zones classification method from Copernicus land monitoring service datasets: An ArcGIS-based toolbox

Local Climate Zones (LCZ) have become a worldwide standard for identifying land cover classes, according to their climate-relevant morphological parameters. The LCZ's are mostly used to evaluate urban climate performance, particularly the relationship between the urban heat island effect (UHI) and the characteristics of the built-up environment. The World Urban Database and Access Portal Tools (WUDAPT) has provided a supervised LCZ classification method based only on moderate resolution free satellite imagery, mostly Landsat 7 or 8 (30 m pixel size, in the visible spectrum brands); however, its’ results are less accurate for European cities. Conversely, alternative geographic information system (GIS)-based methods developed so far require information that is hardly available to all, such as building footprints or heights. Here, the ArcGIS based LCZ from Copernicus Toolbox (LCZC) provides an alternative classification method that uses only freely accessible information from the Copernicus Land Monitoring Service (CLMS), being possible to replicate it in 800 European urban locations. The method combines Urban Atlas (UA) and Corine Land Cover (CLC) with Tree Cover Density, Dominant Leaf Type and Grassland information, to produce a higher-resolution baseline shapefile that is classified according to each feature's dominant characteristics. The LCZC toolbox output is a LCZ raster map. It has been validated in five European cities: Athens, Barcelona, Lisbon, Marseille, and Naples.•The LCZC toolbox provides an alternative LCZ GIS-based classification, based on freely accessible CLMS datasets.•The use of CLMS shapefile higher-resolution inputs, particularly the UA and CLC datasets, ensures an output LCZ map that has greater detail and higher accuracy.•The availability of CLMS information in 800 European urban areas guarantees that the method can be replicated in those locations.

Spatio-temporal analysis of land cover changes in the evergreen and semi-evergreen rainforests: A case study in Chittagong Hill Tracts, Bangladesh

The main purpose of this study is to describe the spatio-temporal analysis of land use and land cover status and to identify land cover changes, especially of deforestation and degradation in evergreen, semi-evergreen rainforests of Chittagong Hill Tracts from 1988-2018 by using Landsat 8 OLI-TIRS and Landsat 5 TM satellite imagery. The ArcGIS v10.5 and ERDAS Imagine v15 software were used to process satellite imageries and assess quantitative data for land-use change assessment of this study area. The study revealed that the area of forest land and water body decreased by 17.92% and 5.43% respectively from 1988-2018. On the other hand, the area of agricultural land, barren land and settlement increased by 45.66%, 312.08% and 240.01% respectively. If the present condition remains constant, the projection of future land-use/ land cover changes for the next 15 years will predict that more than 7.37% dense forest (2253.83 ha) land will be decreased and 19.60% agricultural will be converted to other land uses. This study suggests that proper policy should be adopted urgently to conserve residual forest coverage and restore it to regain its past appearance.

Analysis of the Spatial and Temporal Evolution of Land Cover and Heat Island Effects in Six Districts of Chongqing's Main City.

The urban heat island effect has always been one of the hottest issues in urban development. In this study, Landsat images from the summers of 2001, 2004, 2009, 2014 and 2018 were used to identify land cover type in six districts of Chongqing’s main city. Land cover was categorized as water, vegetation or impervious surface with the object-oriented method. Land surface temperature (LST) data was calculated with the atmospheric radiation transfer equation method, and was then divided into different heat island intensity grades. Next, the spatial and temporal changes in land cover type and heat island effect were analyzed in the six districts. Center migration analysis and heat island coefficients were used to quantitatively reflect the spatiotemporal evolution relationship between land cover and heat island effect. All six districts exhibited a trend of expanding impervious surface, with a 419.38% increase from 2001 to 2018, and shrinking vegetation, with a 17.81% decrease from 2001 to 2018. Also from 2001 to 2018, Yuzhong District had the most significant heat island effect, with a heat island coefficient 0.35 higher than the mean value of the whole study area. The impervious surface center migrated in different directions in each district. Both the direction and the corresponding velocity of the impervious surface and heat island centers were tightly correlated, with a correlation coefficient of 0.53. Relative heat island coefficients (the difference from the mean) of water ranged from −2.08 to −1.17 in different districts. That of impervious surface ranged from 1.60 to 1.93, and that of vegetation ranged from −0.22 to 1.09. The internal heterogeneity of land cover and heat island effect in Chongqing’s main city was huge. This study quantitatively analyzed the evolution of the heat island effect in the study area to help provide each district with some targeted suggestions for future urban planning.

Relationship between land cover and Anophelinae species abundance, composition and diversity in NW Colombia

Anthropic activities, mainly deforestation, have produced rapid transformation of land cover types in the Urabá region at northwest Colombia. Land cover alterations impact the abundance and composition of the Anophelinae community, affecting malaria transmission dynamics. Therefore, this study used landscape metrics to evaluate the relationship of land cover types with Anophelinae species abundance, composition and diversity in the important malaria endemic Urabá region, in NW Colombia. Orthorectified aerial photographs were used to identify land cover types in four localities of the region. Landscape metrics were obtained and diversity indices were estimated for both, land covers and Anophelinae species collected. The impact of land cover type on the presence and abundance of Anophelinae species was evaluated using a canonical correspondence analysis. Diversity indices showed differences in the Anophelinae community and land covers. The variables with more influence in the Anophelinae community composition were locality, bare soil and the interaction between forest and bare soil covers. The most abundant and dominant species Nyssorhynchus nuneztovari (former Anopheles nuneztovari), related with impacted environments was associated with grass, shrub and bare soil land covers. In conclusion, land covers derived from anthropic activities favored the presence and abundance of the main malaria vectors; but, regardless of differences in landscape, unknown specific factors varying among localities lead to a unique configuration in each site that directly shaped anopheline community composition locally. This information is essential for the development of malaria risk maps and for the design of integrated vector control interventions that include the recognition of the landscape features favoring human-vector contact.

Fixing an illusion – an empirical assessment of correction methods for the terrain reversal effect in satellite images

ABSTRACT Identifying land forms and land cover classes are important tasks in image interpretation. Sometimes, a phenomenon called terrain reversal effect (TRE) causes an inverted perception of 3D forms. When this inversion occurs, valleys appear as ridges and vice versa. While the TRE can severely impair the ability to identify 3D land forms, ‘correcting’ for the TRE in imagery can introduce new problems. Importantly, one of most commonly-proposed methods – shaded relief map (SRM) overlay – appears to impair the ability to identify land cover classes. In this paper, we report a comparative empirical evaluation of an SRM overlay solution, and its ‘enhanced’ versions supported by various other cues (stereopsis, motion, labels). In response to the different solutions, we measure the effectiveness, efficiency, confidence and preferences of our participants in land form and land cover identification tasks. All examined methods significantly improve the ability to detect land forms accurately, but they also impair the ability to identify the land cover classes to different degrees. Additionally, participants’ visualization preferences contradict their performance with them, calling for reflection on the visual effects of the applied correction methods. Based on the study, recommendations concerning the correction of the TRE are drawn, and gaps are identified.

ANALISIS PERUBAHAN TUTUPAN LAHAN HUTAN DI IUPHHK-HA PT. GEMA HUTAN LESTARI KABUPATEN BURU PROVINSI MALUKU

The objectives of this study are: identifying land cover, analyzing land cover change, calculating the rate of forest degradation and deforestation, and determining forest management policies. This research was conducted using remote sensing methods and GIS for obtaining land cover change data for several years and conducting mapping. From the research conducted, the results of the classification of forest land cover using Landsat 8 OLI / TIRS imagery in 2013, 2016 and 2018 consisted of forests, shrubs, reeds, swamps, rivers and open land. The highest land cover change that occurred in 2013 up to 2016 covered forests turned into open land amounting to 12261.75 ha. Shrubs into open land also experienced a large change of 12912.25 ha. Forests that have been degraded since 2013-2016 covering an area of ​​17703.07 ha,

Normalization method for multi-sensor high spatial and temporal resolution satellite imagery with radiometric inconsistencies

Distributed systems of small satellites are generating a new type of remote sensing data: multi-spectral data with high spatial and temporal resolution and near-daily global coverage. This data is proving to be valuable for monitoring land cover change. However, in order to achieve widespread application for this purpose, all satellites in the distributed system, or constellation, must acquire imagery with accurate georeferencing and consistent radiometric properties. In this research, we developed a method to automatically co-register and radiometrically normalize a temporally dense series of smallsat images using geocorrected reference imagery. To demonstrate the approach, we normalized a smallsat image time series at both the pixel and the polygon level, smoothing spectral indices through time to detect both abrupt and gradual changes. Using PlanetScope imagery, we tested these methods in a forested region of British Columbia, Canada heavily impacted by forest fires in 2017. By examining the normalized difference vegetation index (NDVI) in the acquired imagery before and after the fires, we found that this method allowed simple identification of burned and unburned areas, which was not readily possible without applying the normalization method. Our result suggests that the developed approach can help fully exploit remote sensing datasets that have high spatial resolution and are acquired with high frequency but potentially contain radiometric inconsistencies in order to quickly identify land cover changes.

A Machine Learning-Based Approach for Land Cover Change Detection Using Remote Sensing and Radiometric Measurements

An approach combining the Hotelling $T^{2}$ control method with a weighted random forest classifier is proposed and used in the context of detecting land cover changes via remote sensing and radiometric measurements. Hotelling $T^{2}$ procedure is introduced to identify features corresponding to changed areas. Nevertheless, $T^{2}$ scheme is not able to separate real from false changes. To tackle this limitation, the weighted random forest algorithm, which is an efficient classification technique for imbalanced problems, has been successfully applied to the features of the detected pixels to recognize the type of change. The feasibility of the proposed procedure is verified using SZTAKI AirChange benchmark data. Results proclaim that the proposed detection scheme succeeds to effectively identify land cover changes. Also, the comparisons with other methods (i.e., neural network, random forest, support vector machine, and $k$ -nearest neighbors) highlight the superiority of the proposed method.

Landcover change analysis of besitang watershed, Langkat-North Sumatera

Watershed is an ecosystem area bordered by topographic pattern and serves as serves as a collector, storage and distributor of water, sediments, pollutants and nutrients in the river system and throughout by a single outlet. The high of population growth led to greater land use change. This research aimed to identify land cover classes and land cover change in Besitang watershed between 1990, 2005 and 2015. This research used Landsat 5 imagery in 1990 and 2005, and Landsat 8 imagery in 2015 with supervised classification using method maximum likelihood classifier. The result showed that there were 12 classes of land cover in the Besitang watershed, they were primary forest, secondary forest, mangrove forest, scrub, rubber plantations, oil palm plantations, dry land agriculture, bare land, settlement, fishpond and water body. The largest area of land cover area is primary forest covering an area of 38,542.43 hectares (39.94%) in 1990, 34,279.16 hectares (35.52%) in 2005, and 34,620.41 hectares (35.88%) in 2015. The largest area of land cover change between 1990 until 2005 is mangrove forest to fishpond with covering an area of 2,364.21 hectares change. While the largest area of land cover change between 2005 until 2015 is mangrove forest to oil palm plantation with covering an area of 1,016.82 hectares changes.

Analyses of Time Series InSAR Signatures for Land Cover Classification: Case Studies over Dense Forestry Areas with L-Band SAR Images.

As demonstrated in prior studies, InSAR holds great potential for land cover classification, especially considering its wide coverage and transparency to climatic conditions. In addition to features such as backscattering coefficient and phase coherence, the temporal migration in InSAR signatures provides information that is capable of discriminating types of land cover in target area. The exploitation of InSAR signatures was expected to provide merits to trace land cover change in extensive areas; however, the extraction of suitable features from InSAR signatures was a challenging task. Combining time series amplitudes and phase coherences through linear and nonlinear compressions, we showed that the InSAR signatures could be extracted and transformed into reliable classification features for interpreting land cover types. The prototype was tested in mountainous areas that were covered with a dense vegetation canopy. It was demonstrated that InSAR time series signature analyses reliably identified land cover types and also recognized tracing of temporal land cover change. Based on the robustness of the developed scheme against the temporal noise components and the availability of advanced spatial and temporal resolution SAR data, classification of finer land cover types and identification of stable scatterers for InSAR time series techniques can be expected. The advanced spatial and temporal resolution of future SAR assets combining the scheme in this study can be applicable for various important applications including global land cover changes monitoring.

Using of Multi-Source and Multi-Temporal Remote Sensing Data Improves Crop-Type Mapping in the Subtropical Agriculture Region.

Crop-type identification is very important in agricultural regions. Most researchers in this area have focused on exploring the ability of synthetic-aperture radar (SAR) sensors to identify crops. This paper uses multi-source (Sentinel-1, Sentinel-2, and Landsat-8) and multi-temporal data to identify crop types. The change detection method was used to analyze spectral and indices information in time series. Significant differences in crop growth status during the growing season were found. Then, three obviously differentiated time features were extracted. Three advanced machine learning algorithms (Support Vector Machine, Artificial Neural Network, and Random Forest, RF) were used to identify the crop types. The results showed that the detection of (Vertical-vertical) VV, (Vertical-horizontal) VH, and Cross Ratio (CR) changes was effective for identifying land cover. Moreover, the red-edge changes were obviously different according to crop growth periods. Sentinel-2 and Landsat-8 showed different normalized difference vegetation index (NDVI) changes also. By using single remote sensing data to classify crops, Sentinel-2 produced the highest overall accuracy (0.91) and Kappa coefficient (0.89). The combination of Sentinel-1, Sentinel-2, and Landsat-8 data provided the best overall accuracy (0.93) and Kappa coefficient (0.91). The RF method had the best performance in terms of identity classification. In addition, the indices feature dominated the classification results. The combination of phenological period information with multi-source remote sensing data can be used to explore a crop area and its status in the growing season. The results of crop classification can be used to analyze the density and distribution of crops. This study can also allow to determine crop growth status, improve crop yield estimation accuracy, and provide a basis for crop management.

Analyzing Land Cover Change Using Remote Sensing and GIS: A Case Study of Gilgit River Basin, North Pakistan

Mountainous areas of northern Pakistan are rich in biodiversity, glaciers and key watershed of Indus Riversystem which provide ecosystem services for their inhabitants. These regions have experienced extensive deforestationand are presently vulnerable by rapid land cover changes, therefore an effective assessment and monitoring is essentialto capture such changes. The aim of this study is to analyze the observed changes in land cover over a period of thirtynine years, divided into three stages (1976-1999, 1999-2008 and 2008-2015). Four images from Landsat 2Multispectral Scanner System (MSS), Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper andLandsat 8 Operation Land Imager data were obtained to detect land cover change. This study used supervisedclassification-maximum likelihood algorithm in ERDAS imagine to identify land cover changes perceived in GilgitRiver Basin, Pakistan. The result showed that the range land, glaciers, water bodies, built-up/agricultural cover are themajor categories that have been altered by the natural and anthropogenic actions. In 1976, built up/agriculture, rangeland, water bodies and glacier cover was 1.13%, 45.3%, 0.66% and 13.2%, respectively. Whereas in 2015, builtup/agriculture, range land, water bodies and glacier cover was 3.25%, 12.7%, 0.91% and 8.2%, respectively. Thesesland cover shifts posed acute threat to watershed resources. Therefore, a comprehensive watershed resourcemanagement is essential or otherwise, these resources will deplete rapidly and no longer be capable of playing their rolein socioeconomic and sustainable environmental development of the area

Analyzing Land Cover Change Using Remote Sensing and GIS: A Case Study of Gilgit River Basin, North Pakistan

Mountainous areas of northern Pakistan are rich in biodiversity, glaciers and key watershed of Indus Riversystem which provide ecosystem services for their inhabitants. These regions have experienced extensive deforestationand are presently vulnerable by rapid land cover changes, therefore an effective assessment and monitoring is essentialto capture such changes. The aim of this study is to analyze the observed changes in land cover over a period of thirtynine years, divided into three stages (1976-1999, 1999-2008 and 2008-2015). Four images from Landsat 2Multispectral Scanner System (MSS), Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper andLandsat 8 Operation Land Imager data were obtained to detect land cover change. This study used supervisedclassification-maximum likelihood algorithm in ERDAS imagine to identify land cover changes perceived in GilgitRiver Basin, Pakistan. The result showed that the range land, glaciers, water bodies, built-up/agricultural cover are themajor categories that have been altered by the natural and anthropogenic actions. In 1976, built up/agriculture, rangeland, water bodies and glacier cover was 1.13%, 45.3%, 0.66% and 13.2%, respectively. Whereas in 2015, builtup/agriculture, range land, water bodies and glacier cover was 3.25%, 12.7%, 0.91% and 8.2%, respectively. Thesesland cover shifts posed acute threat to watershed resources. Therefore, a comprehensive watershed resourcemanagement is essential or otherwise, these resources will deplete rapidly and no longer be capable of playing their rolein socioeconomic and sustainable environmental development of the area