Abstract. Monitoring water quality in inland water bodies is critical for environmental management, yet traditional sampling methods are costly and spatially limited. Remote sensing offers a viable alternative by enabling large-scale assessment of water quality parameters such as chlorophyll-a (Chl-a) concentrations through empirical models linking spectral indices to in situ measurements. However, the accuracy of these models depends on representative sampling strategies that capture spatial and temporal variability. This study evaluates a clustering-based approach to optimize sampling site selection in the Solís Dam, Mexico, using Sentinel-2 imagery. We analyzed a one-year time series of Sentinel-2 data to compute spectral indices related to Chl-a and turbidity. Unsupervised K-means clustering was applied to stratify the reservoir into zones of distinct water quality variability, guiding the placement of 20 sampling sites. Field campaigns during dry and wet seasons (2024) provided Chl-a measurements, which were correlated with spectral indices. The Gi033BDA index showed the strongest correlation (R2 > 0.7, p < 0.01) and was used to develop a linear regression model for Chl-a estimation. Results confirmed that clustering-derived sampling points effectively represented spatial variability, though temporal mismatches (1-day lag) and samples location inaccuracies introduced minor errors. The method demonstrates how pre-stratification using remote sensing can enhance sampling efficiency while maintaining model accuracy. This approach is particularly valuable for large-scale monitoring, reducing reliance on exhaustive field campaigns. Future work should address temporal dynamics and sensor resolution trade-offs for broader applicability.

Peatlands play a critical role in global and regional climate regulation by functioning as long-term carbon sinks, regulating hydrology, and modulating land–atmosphere energy exchange. Intact peat ecosystems store large amounts of organic carbon and stabilize local climate through high water retention and evapotranspiration, whereas peatland degradation disrupts these functions and can transform peatlands into significant sources of greenhouse gas emissions and climate extremes such as drought and fire. Indonesia contains approximately 13.6–40.5 Gt of carbon, around 40% of which is stored on the island of Sumatra. However, tropical peatlands in this region are highly vulnerable to climate anomalies and land-use change. This study investigates the impacts of major climate anomalies—specifically El Niño and positive Indian Ocean Dipole (pIOD) events in 1997/1998, 2015/2016, and 2019—on peatland cover change across South Sumatra, Jambi, Riau, and the Riau Islands. Landsat 5 Thematic Mapper and Landsat 8 Operational Land Imager/Thermal Infrared Sensor imagery were analyzed using a Random Forest machine learning classification approach. Climate anomaly periods were identified using El Niño-Southern Oscillation (ENSO) and IOD indices from the National Oceanic and Atmospheric Administration. To enhance classification accuracy and detect vegetation and hydrological stress, spectral indices including the Normalized Difference Vegetation Index (NDVI), Modified Soil Adjusted Vegetation Index (MSAVI), Normalized Difference Water Index (NDWI), and Normalized Difference Drought Index (NDDI) were integrated. The results show classification accuracies of 89–92%, with kappa values of 0.85–0.90. The 2015/2016 El Niño caused the most severe peatland degradation (>51%), followed by the 1997/1998 El Niño (23–38%), while impacts from the 2019 pIOD were comparatively limited. These findings emphasize the importance of peatlands in climate regulation and highlight the need for climate-informed monitoring and management strategies to mitigate peatland degradation and associated climate risks.

Abstract Background Assessing the severity of forest fires allows us to identify changes that compromise the natural regeneration capacity of vegetation. In this study, we evaluated the severity and recovery of vegetation after a fire using Sentinel-2 satellite images for the Cajamarca department in northeastern Peru. Hot spots were downloaded from the Fire Information for Resource Management System (FIRMS). This allowed us to identify eight groups with an area > 100 hectares, heat intensity > 100 Fire Radiative Power (FRP), and the presence of trees. By applying the Normalized Burn Ratio (NBR) and the Normalized Difference Vegetation Index (NDVI), the levels of extreme, high, medium, and low severity were determined, as well as the recovery of vegetation before and after the fire events. Results and Conclusions The results indicated that 71.02% of the evaluated territory had low severity, 21.95% had medium severity, 6.41% had high severity, and 0.65% had extreme severity, indicating a prevalence of medium to low severity in the study area. The fires that occurred had similar NDVI levels in the pre-fire stage; however, after the fire, a progressive recovery of vegetation was observed in the study area. This highlights the application of spectral indices to assess the impact and regrowth of vegetation after the development of fires.

Abstract Scots pine ( Pinus sylvestris L.) is widely distributed, phenotypically plastic forest tree species with modest ecological demands, therefore it is a very suitable, drought tolerant species for afforestation at present. This is especially important given Europe’s changing climate, with rising extremes and unpredictable rainfall challenging forest regeneration. Drought resistance of seedlings is essential for their survival during current reforestation efforts, however, its relation to ecotypic variation is yet not well understood. The objective of this study was to investigate the response of seedlings from two Czech Scots pine ecotypes (upland and lowland), exposed to water deficit at the beginning of the vegetative season – a critical period for successful afforestation from the perspective of precipitation availability. During a greenhouse experiment with nursery pre-grown seedlings, terminal shoot length and selected leaf functional traits (leaf mass per area; water and pigment contents; needle anatomy), chlorophyll fluorescence kinetics and seedling reflectance were monitored during ten-week irrigation reduction and after rewatering. The photochemical reflectance index (PRI) and the red edge position (REP) were calculated from spectral reflectance to distinguish differently treated seedlings. The lowland ecotype grew faster under control but suffered stronger growth reduction and higher mortality under drought. In contrast, across all recorded responses, the upland ecotype responds more consistently to changes in water availability, does not reduce terminal growth, accumulates less biomass and exhibits lower mortality. In general, for terminal growth, there was a significant effect of treatment and also an interaction of treatment and ecotype during the recovery period, unlike the drought period. REP was responsive in recovery period for upland ecotype while PRI showed no consistent drought-related pattern. Our results, in agreement with the fluorescence-based indicators, suggest that current-year needles are more suitable for drought stress detection using spectral indices. The upland ecotype showed several functional traits corresponding to better resilience to drought stress compared to the lowland ecotype. Understanding drought stress and recovery responses via effective leaf functional traits will help forest management to select suitable ecotypes for reforestation, ensuring a higher survival under changing climatic conditions.

Air pollution, particularly fine particulate matter (PM2.5), poses significant public health and environmental risks. This study explores the effectiveness of spatiotemporal graph neural networks (ST-GNNs) in forecasting PM2.5 concentrations by integrating remote-sensing hyperspectral indices with traditional meteorological and pollutant data. The model was evaluated using data from Switzerland and the Gauteng province in South Africa, with datasets spanning from January 2016 to December 2021. Key performance metrics, including root mean squared error (RMSE), mean absolute error (MAE), probability of detection (POD), critical success index (CSI), and false alarm rate (FAR), were employed to assess model accuracy. For Switzerland, the integration of spectral indices improved RMSE from 1.4660 to 1.4591, MAE from 1.1147 to 1.1053, CSI from 0.8345 to 0.8387, POD from 0.8961 to 0.8972, and reduced FAR from 0.0760 to 0.0719. In Gauteng, RMSE decreased from 6.3486 to 6.2319, MAE from 4.4891 to 4.4066, CSI from 0.9555 to 0.9560, and POD from 0.9699 to 0.9732, while FAR slightly increased from 0.0154 to 0.0181. Error analysis revealed that while the initial one-day ahead forecast without spectral indices had a marginally lower error, the dataset with spectral indices outperformed from the two-day ahead mark onwards. The error for Swiss monitoring stations stabilized over longer prediction lengths, indicating the robustness of the spectral indices for extended forecasts. The study faced limitations, including the exclusion of the Planetary Boundary Layer (PBL) height and K-index, lack of terrain data for South Africa, and significant missing data in remote sensing indices. Despite these challenges, the results demonstrate that ST-GNNs, enhanced with hyperspectral data, provide a more accurate and reliable tool for PM2.5 forecasting. Future work will focus on expanding the dataset to include additional regions and further refining the model by incorporating additional environmental variables. This approach holds promise for improving air quality management and mitigating health risks associated with air pollution.

Reliable mapping of vegetation and surface water from satellite imagery remains challenging, as common spectral indices can saturate at high biomass, show limited sensitivity across ecosystems, and confuse targets with soil, shadows, or built-up surfaces. We present two indices, the Symbolic Regression Vegetation Index (SRVI) and the Symbolic Regression Water Index (SRWI), discovered with a data-driven symbolic regression framework applied to Sentinel-2 Level-2A reflectance and guided by ESA WorldCover labels. Expressions were evolved from physically interpretable building blocks using non-linear combinations of visible, NIR, and SWIR bands. Indices were derived on a spectrally complex Mediterranean site and evaluated on eleven independent regions spanning diverse biomes. Performance was assessed with the Jeffries-Matusita distance, averaged across months to account for phenology, and compared against established vegetation indices (NDVI, EVI, SAVI, MSAVI2, NDRE) and water indices (NDWI, MNDWI, AWEI, TCW, WI2015). SRVI improves separability between vegetation and non-vegetation and shows higher discrimination among vegetation types relative to all benchmarks. SRWI yields more consistent water delineation with reduced confusion with built-up and shadowed surfaces, outperforming standard alternatives on the same datasets. Results indicate that symbolic regression can produce compact, interpretable indices that generalise across regions and seasons, offering practical gains for global vegetation and water mapping.

Abstract Mentha arvensis (menthol mint) is a high-value aromatic crop central to the agricultural economy of the Barabanki-Lucknow belt of Uttar Pradesh, yet its productivity remains constrained by traditional cultivation practices, limited technological adoption, and climate-driven uncertainties. This review synthesizes research advancements from 2021 to 2025, spanning agronomy, remote sensing, IoT-based monitoring, machine learning, digital advisory systems, and policy frameworks. Studies on AI-enabled irrigation, multimodal sensing, spectral vegetation indices, and improved cultivars highlight substantial opportunities for precision-driven interventions in mentha farming. IoT systems, solar-powered sensor networks, edge computing models, and predictive analytics demonstrate their potential to enhance irrigation scheduling, nutrient management, stress detection, and real-time decision support. Economic analyses and policy reports further emphasize mentha’s profitability and its suitability for technology-led transformation. By integrating insights from agriculture, computer science, and rural development, this review presents a unified understanding of digital and scientific innovations relevant to Mentha arvensis and identifies future directions for building scalable, farmer-centric smart farming ecosystems such as MentholGrow. Keywords Mentha arvensis; Menthol mint; Smart Agriculture; IoT-based monitoring; Machine Learning; Remote sensing; Precision farming; Edge computing; Decision-support systems; Aromatic crops; Digital agriculture; Uttar Pradesh; Barabanki-Lucknow belt.

ABSTRACT This study aimed to demonstrate the efficacy of the ordinary kriging interpolation method as a data augmentation technique to enhance the accuracy of models that predict the water quality parameter total dissolved solids (TDS). For this purpose, the total dissolved solids parameter was utilized as a case study. The models employed seven spectral indices derived from Sentinel-2 images as predictor variables, recognized as effective in predicting TDS: Green, Ferdous, SI-3, SI-5, TDS-1, and TDS-5. Three types of models were compared: (1) simple linear regression models trained with the original field samples; (2) simple linear regression models trained with an augmented dataset of 1000 data points, generated from a geostatistical kriging surface derived from the in-situ data; and (3) multilayer neural network models trained with the same augmented dataset of 1000 data points. It was concluded that the ordinary kriging method is an effective data augmentation technique, as there is a statistically significant difference between the models trained with field samples and those trained with augmented data, with the latter demonstrating greater explanatory power. Additionally, in the validation of the prediction surfaces derived from the models trained with the augmented dataset, it was observed that the mean absolute error (MAE) and the root mean square error (RMSE) were below the natural variability of the original in-situ data, thereby confirming the high predictive capacity of the models trained with the augmented dataset. This research is of significant importance, as it innovatively integrates geostatistical techniques, satellite remote sensing, and artificial intelligence to effectively address the critical issue of data scarcity in water quality modelling, taking TDS as a test parameter. The results of this research establish a robust foundation for the application of artificial intelligence in water quality modelling, facilitating precise predictions at specific locations and times.

Accurate global mapping of lunar iron oxide (FeO) abundance is essential for understanding the Moon’s geological evolution and for supporting future in situ resource utilization (ISRU). While hyperspectral data from the Moon Mineralogy Mapper (M3) provide a unique combination of high spectral dimensionality, hectometre-scale spatial resolution, and near-global coverage, existing FeO retrieval approaches struggle to fully exploit the high dimensionality, nonlinear spectral variability, and planetary-scale volume of the Global Mode dataset. To address these limitations, we present an integrated machine learning pipeline for estimating lunar FeO abundance from M3 hyperspectral observations. Unlike traditional methods based on raw reflectance or empirical spectral indices, the proposed framework combines Discrete Wavelet Transform (DWT), deep autoencoder-based feature compression, and ensemble regression to achieve robust and scalable FeO prediction. M3 spectra (83 bands, 475–3000 nm) are transformed using a Daubechies-4 (db4) DWT to extract 42 representative coefficients per pixel, capturing the dominant spectral information while filtering high-frequency noise. These features are further compressed into a six-dimensional latent space via a deep autoencoder and used as input to a Random Forest regressor, which outperforms kernel-based and linear Support Vector Regression (SVR) as well as Lasso regression in predictive accuracy and stability. The proposed model achieves an average prediction error of 1.204 wt.% FeO and demonstrates consistent performance across diverse lunar geological units. Applied to 806 orbital tracks (approximately 3.5×109 pixels), covering more than 95% of the lunar surface, the pipeline produces a global FeO abundance map at 150 m per pixel resolution. These results demonstrate the potential of integrating multiscale wavelet representations with nonlinear feature learning to enable large-scale, geochemically constrained planetary mineral mapping.

Thermal hotspot detection in landfills using high-resolution land surface temperature data: A case study of active and closed sites.

The Influence of Temperature Dynamics, Land Cover, and Spectral Indices on Urban Environment: a Special Emphasis on Chennai

Quantitative estimation of blueberry SSC using fractional order derivative coupled optimized spectral indices

Measuring the properties of disks around Class 0/I protostars is crucial for understanding protostellar assembly and early planet formation. We present high-resolution (~7.5 au) ALMA continuum observations at 1.3 and 3 mm of 16 disks around Class 0/I protostars across multiple star-forming regions (Taurus, Ophiuchus, and Corona Australis) and a variety of multiplicities. Our observations show a wide range of deconvolved disk sizes (~2–100 au) and the presence of circumbinary disks (CBDs) in all binaries with separations <100 au. The measured properties show similarities to Class II disks, including (a) low spectral index values ( α disks = 2.1 −0.3 +0.5 ) that increase with disk radius, (b) 3 mm disk sizes only marginally smaller than at 1.3 mm (<10%), and (c) radial intensity morphologies well described by modified self-similar profiles. However, there are some key differences: (i) the α 1.3-3 mm values increase monotonically with radius but exceed two only at the disk edge; (ii) higher brightness temperatures, T b , comparable to or higher than the predicted midplane temperatures due to irradiation; and (iii) an approximately ten times higher luminosity at a given size compared to the Class II disks. Together, the results confirm significant optical depth in the observed Class 0/I disks, most with T bol < 200 K, at both 1.3 and 3 mm. Assuming fully optically thick disks at these wavelengths can explain the higher luminosities compared with Class II disks, but the most compact (≲40 au) disks also require higher temperatures, suggesting additional heating from viscous accretion. Taking into account the high optical depths, most disk dust masses are estimated in the range 30–900 M ⊕ (or 0.01–0.3 M ⊙ in gas), with some disks potentially reaching marginal gravitational instability. Based on the elevated T b 1.3 mm , the median location of the water iceline is ~3 au, but this location can extend to more than 10–20 au for the hottest disks in the sample. The CBDs exhibit lower optical depths at both wavelengths and hence higher spectral index values ( τ 3 mm ≲ 1, α CBD = 3.0 −0.3 +0.2 ), dust masses of ~10 2 M ⊕ , and dust emissivity indices of β CBD ~ 1.5 (two Class 0 CBDs) and ~1 (one Class I CBD), suggesting substantial grain growth only in the more evolved CBD. The high optical depths inferred from our analysis provide a compelling explanation for the apparent scarcity of dust substructures in the younger Class 0/I disks at ~1 mm despite the mounting evidence of early planet formation.

Abstract. Land Surface Temperature (LST) serves as a critical parameter for evaluating urban climate dynamics and surface energy exchanges. This study examines the relationships between LST and four spectral land cover indices—Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-up Index (NDBI), Normalized Difference Water Index (NDWI), and Normalized Difference Bareness Index (NDBaI)—across four Köppen–Geiger climate zones: Kars (Dfb – Humid Continental), Kilis (Csa – Hot-Summer Mediterranean), Cairo (BWh – Hot Desert), and Malanje (Aw – Tropical Savanna). Using Landsat 8 OLI/TIRS Collection 2 Level-2 imagery acquired in September 2023 (and 2020 for Kars), LST and spectral indices were extracted and analyzed through pixel-based Pearson correlation analysis. The results revealed diverse climatic dependence in the LST–index interactions. In Kars, LST showed a strong positive correlation with NDBI (r = 0.63) and a moderate correlation with NDBaI (r = 0.39). In Kilis, NDVI exhibited a moderate negative relationship with LST (r = −0.47), while NDBI correlated weakly (r = 0.22). Cairo displayed weak overall relationships, with LST–NDBI (r = 0.38) and LST–NDVI (r = −0.22) reflecting the dominance of impervious and arid surfaces. Conversely, Malanje demonstrated the strongest vegetation–temperature interaction, where LST–NDVI correlation reached r = −0.75, LST–NDWI r = 0.72, and LST–NDBI r = 0.53. Across all cities, built-up and bare areas consistently increased LST, while vegetation showed cooling effects that intensified in warmer, more humid climates. These findings highlight that the magnitude and direction of LST–land cover correlations are strongly controlled by regional climate regimes, emphasizing the necessity of climate-specific urban heat mitigation strategies.

This study aims to analyze the differences of each satellite product and their responses between the dry and rainy seasons by investigating seasonal floods and spectral indices based on Hierarchical Split-Based Approach (HBSA) from Sentinel-1and Sentienl-2 imagery, respectively, together with Climate Hazards group Infrared Precipitation with Stations (CHIRPS) precipitation data. The spectral indices calculated from Sentinel-2 images at different dates allow to monitor environmental changes, such as the dynamics of floods and land use and land cover. Sentinel-1 data is suitable to map flood extents by applying threshold algorithms to help differentiate water bodies from other land cover types, while CHIRPS rainfall provides essential information for seasonal variations in flooding. Using these three satellite products, this study is to understand the relationships of each satellite responds and which veriable are best suited to describing flooding on a local scale. The precipitation exhibited significant variability throughout both seasons. The rainy season saw the highest precipitation in September, while the dry season also showed in November, May, and April. The lowest precipitation occurred in June (i.e., rainy season) and January to March and December (dry season). The water bodies varied significantly, with the highest in October and November (rainy season), and the lowest in June, July, and February (dry season). The Modified Normalized Difference Water Index (MNDWI) and Normalized Difference Water Index proposed by GAO (NDWI-GAO) values remained consistent during the rainy season compared to the dry season with significant difference, possibly due to their detection of flooded rice fields with irrigation. These results show the relationships between satellite observation data and further study will be needed to validate the possibility of using them to predict flooding for early warnings.

This study aimed to monitor and evaluate the impact of fires on the appearance of the land surface in open and flat agricultural areas using spectral indices and Landsat8 Operational Land Imagery (OLI) satellite images. A representative area, covering 250.34 km², was selected in Nineveh Governorate in Iraq lies between latitudes (43˚ 20′ 0̋-43˚ 32′ 30̋) N and longitude (36˚ 25′ 0̋-36˚ 11′ 0̋) E on June and July 2019 in order to determine the burned area on the land, and use the resulting spectral signature to generalize it to the rest of the study area. Data from the American Landsat satellite, GIS software, and spectral indicators for vegetation covers, such as the Normalized Difference Vegetation Index (NDVI), the Soil-Adjusted Vegetation Index (SAVI), and the Normalized Burning Ratio index (NBR), were used to extract the results. Then, the Land Surface Temperature index (LST) and the difference Normalized Burning Ratio index (dNBR) were used to measure the validity of the results and the severity of the fires, respectively. After using and performing corrective processing of the spectral indices, applying the three indices and determining the difference between them for a short period of time, it was possible to discover the areas of fires and the proportions of their effects. The NDVI results showed that the area affected by very severe degradation increased due to the fires to 39.29%, while the LST increased from 47.70 to 48.39 degree. The study concluded that the area of ​​fire spread can be determined when there is a discrepancy in the pattern of spatial data that are close in date of capture and by using spectral indices.

Modern technologies for intelligent monitoring play a key role in the sustainable management of forest resources and the protection of biodiversity. Satellite images with appropriate resolution for forest research from different types of platforms, combined with spectral indices (vegetation, soil, forest fire, etc.), allow large-scale monitoring of the state of forests, changes in land cover, damage resulting from fires and anthropogenic activity. This paper presents a brief overview of the development of intelligent monitoring technologies and their applications for the study of forest areas, the possibilities and future directions for making informed decisions by institutions in the forest sector. Models for the study of forest areas affected by fires using Sentinel-2 satellite images in the QGIS and Google Earth Engine environment are considered. In relation to climate change, as well as the growing needs for automation, precise analysis and processing of big geospatial data, the consideration of such models is becoming more and more crucial.

Abstract We perform a comprehensive superposed epoch analysis of more than 200 corotating interaction regions (CIRs) using WIND spacecraft observations at 1 au. The stream interfaces are identified by minimum variance analysis, and turbulence properties are evaluated using wavelet transforms over a wide range of temporal scales. The analysis of normalized cross helicity ( σ c ) and normalized residual energy ( σ r ) reveals distinct turbulence behaviors across frequencies. The spectral indices of both magnetic and velocity fluctuations smoothly transition from steeper in the slow wind to shallower in the fast wind, while a localized steepening of the velocity spectra near the interface indicates enhanced dissipation due to compression. Across broad frequency bands, σ c shows a clear dip at the stream interface—signifying increased inward Alfvén wave energy—whereas σ r displays a peak–valley–peak structure mainly driven by large-scale velocity shear. In lower-frequency ranges, velocity shear artificially enhances velocity fluctuation energy, producing strong peaks in σ r , while higher-frequency ranges show a smooth increase of σ r from slow wind to fast wind. Nearly half of the analyzed CIRs are accompanied by a heliospheric current sheet (HCS), with many HCSs closely aligned with the stream interface, suggesting an intrinsic link between the two structures. Our findings offer valuable clues for reconciling discrepancies among earlier observational and simulation studies, and provide new insight into how compression and velocity shear modulate solar wind turbulence near CIRs.

Abstract. Land Surface Temperature (LST) is a critical parameter for urban climate analysis, photovoltaic (PV) planning, and environmental monitoring, yet its effective use is hindered by the coarse spatial resolution of thermal sensors like MODIS. This study introduces a two-stage hierarchical Convolutional Neural Network (CNN) framework that integrates multi-sensor satellite data (MODIS, ASTER, Sentinel-2) to generate fine-scale LST products and quantitatively assess scaling effects across heterogeneous landscapes. In the first stage, MODIS LST (1 km) was downscaled to 360 m, 270 m, and 90 m using Sentinel-2–derived scaling factors and validated against ASTER LST. In the second stage, ASTER LST (90 m) was downscaled to 60 m and 30 m and validated with Landsat-based LST. The framework employed spectral indices, topographic parameters, texture features, and sub-pixel land-cover fractions as scaling factors, capturing both spatial and spectral heterogeneity. Comparative evaluation against XGBoost and DisTrad revealed that CNN consistently achieved the highest determination coefficients (R2 = 0.69–0.87) and the lowest RMSE (1.94–2.34 K) and MAE (1.49–1.8 K) values, confirming its superior capacity to model complex nonlinear thermal relationships. Scaling-effect analysis demonstrated that while accuracy naturally decreases with finer resolutions, the CNN model exhibits strong scale stability and resilience to error propagation, outperforming traditional regression and machine-learning approaches. This hierarchical deep-learning design establishes a new paradigm for multi-sensor LST reconstruction, enabling accurate, scalable, and spatially coherent thermal mapping across diverse terrains. The proposed framework offers a generalizable solution for high-resolution thermal monitoring, PV site optimization, and climate-adaptive urban planning.

Land Cover-Land Use (LC/LU) classification provides data for effective management of environmental and ecological decisions at the landscape scale. In this process, Sentinel-2 Multi Spectral Imager (MSI) satellite images contribute to classification methods by facilitating information extraction with their high spectral resolution. While index-based methods mostly focus on the separation of single classes, landscapes require the separation of multiple classes. This study shows how different spectral indexes derived from Sentinel-2 MSI imagery can be used in large areas with the object-based image classification technique. The Silifke district of Mersin province was selected as a sample area. Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Built-up Area Extraction Index (BAEI), Built-up Area Index (BAI), Band Ratio (BR28, BR38), Normalized Built-up Area Index (NBAI), New Building Index (NBI), Urban Index (UI), Normalized Difference Soil Tillage Index (NDTI), Red Edge Based Normalized Difference Vegetation Index (NDVIre) and Normalized Difference Water Index (MNDWI) were used. While no significant results were obtained with BR28, BR38, NBAI, NBI and UI, 0.8815 kappa coefficient of 0.8815 and overall accuracy rate of %94.11 were obtained with other indexes.