Monitoring crop growth conditions during the growing season provides an indication of crop health and informs agricultural management. Since physical and chemical properties of soils tend to be spatially heterogeneous, intra-field soil variability is bound to cause intra-field heterogeneity in crop growth rate. Data fusion of soil properties and derived unmanned aerial vehicle (UAV) Vegetation Indices (Vis) can help to improve model performance accuracy for crop growth assessment. The aim of this study was to investigate and understand the contribution of soil properties and unmanned aerial vehicle UAV data to improve modelling accuracy of intra-field crop growth variability for winter wheat. To achieve this aim, the study used soil data, monthly time-series crop height measurements (cm), and Vis acquired through UAV time-series images. For data analysis, two machine learning methods including optimizable Gaussian process regression (GPR) and optimizable least-squares boosting (LSboost) and bagging (Bag) Ensemble regression (ER) were applied in MATLAB software. Results showed that soil properties, particularly Ca, Mg, K, and Clay were more important than VIs in representing actual crop growth. However, when VIs and soil properties were integrated, the GPR model’s coefficient of determination (R²) improved by 0.01 and 0.03, while the RMSE decreased by 0.25 and 0.78 cm for the two farms, respectively. Overall, GPR (R² = 0.68 to 0.75, RMSE = 15.85 to 18.38 cm) performed slightly better than LSboost-Bag-ER (R² = 0.64 to 0.70 and RMSE = 17.26 to 19.34 cm) for both farms. The findings of this study show that, although intra-field crop growth variability is reasonably predicted by soil properties, UAV data, and Vis separately/independently, the synergistic use of these data sources produced better results than the individual datasets.

Inland water bodies play a crucial role in both ecological and sociological contexts. They serve as significant sources of freshwater, meeting various agricultural, domestic, and industrial water demands. The distribution of these water bodies can change over time due to natural or human-induced factors. Monitoring the extent of surface water is vital for understanding extreme events such as floods and droughts. The availability of dense temporal Earth observation data from sensors like Landsat and Sentinel, coupled with advancements in cloud computing, has enabled the analysis of surface water extent over extended periods. In this study, an automated thresholding approach was applied within the Google Earth Engine platform to extract the surface water extent of the Chembarambakkam reservoir in Chennai. Sentinel-2 data spanning from 2016 to 2023 were used to derive two key indices, namely the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI). These indices were then thresholded to determine the presence of water. The performance of two different thresholding techniques, namely Deterministic thresholding and Otsu, a histogram-based thresholding method, was compared to achieve better results. To enhance the accuracy of the deterministic technique, an iterative method was implemented that averaged the mean values of water and non-water areas to establish a new threshold. While the threshold values were generally similar for both techniques, the Otsu algorithm outperformed the deterministic technique in water classification. Furthermore, a frequency of water occurrence image was obtained using the temporal images, providing insights into the surface dynamism of the reservoir. This information is valuable for understanding the temporal changes in the reservoir's water presence. Overall, this study highlights the significance of surface water monitoring using remote sensing and cloud computing techniques. The comparative analysis of thresholding methods emphasizes the superior performance of the Otsu algorithm in classifying water. The frequency of water occurrence image adds an additional layer of understanding regarding the reservoir's surface dynamics.

The aim of the work was to make a comparative analysis of the state of the snow surface in East and West Antarctica, including changes in snow cover characteristics during the past two decades. To do so, we have developed the ASAR (for Antarctic Snow Albedo Retriever) algorithm, which processes satellite data and retrieves an effective snow grain size and a fraction of rocks not covered by snow. The algorithm has been used to process the MODIS data throughout the entire period of its operation (up to now). We have chosen several test areas (30 x 30 km² approximately) to study the state of the snow cover on Enderby Land (East Antarctica), on the coast of the Ross Sea (the Transantarctic Mountains), and the Antarctic Peninsula (West Antarctica). As a result, we have plotted and analyzed time series of the effective snow grain size and rock fraction in these areas across the last 20 years. The study of snow cover trends on a continental scale can contribute to the investigation of environmental changes in Antarctica.

Climate change and urban expansion are together playing a critical role in disrupting urban microclimate. Lebanon is an example country that is suffering from such effects. Particularly, the capital city “Beirut”, which covers only 0.95% of Lebanon’s total area, consumes 12% of the total national energy and has ~36% of the total population. Metropolitan areas continue to expand along adjacent mountains and leading to diminishing green areas. With the absence of plans to monitor this urbanization, it is critical to examine the interplay between urban temperatures and land use patterns to avoid high-intensity urban heat islands (UHI).

In this research study, we use thermal remote sensing technology to analyze the urban heat mapping of the Greater Beirut Area (GBA) at various spatial and temporal scales. Furthermore, we estimate the vegetation cooling impact as a response to rising urban temperatures. The investigation is conducted within a time frame that spans over 3 decades from 1990 to 2020. We use Landsat-5, -7, -8, and MODIS thermal infrared imagery, with a dataset of 524 images of the surface reflectance. For each year, we calculate the normalized difference vegetation index (NDVI) and land surface temperature (LST) statistics. We also create an overlay between each image and GBA topography to measure the vegetation area. Based on LST statistics, we retrieve the urban heat index (UHIndx) in GBA. Then, we examine the correlation between vegetation and temperature. The spatial-temporal analysis is conducted to relate heat mapping in GBA to topography based on altitude and land cover.

Overall results show that the temperature in GBA has increased over 3 decades, with an increase in the vegetation and urban LST by 1.1 °C and 1.26 °C, respectively. According to the land’s altitude, the highest LST values with a mean of 34.36°C are recorded at the lowest altitudes between 0-30 m at the coastal area, even if it is away by a 6 km distance. A substantial drop in LST values is witnessed when the elevation increases to 100s meters, at which rural areas or mountains with vegetation growth exist. Hence, this temperature dependency on the elevation also aligns with the results obtained from the LST statistics according to the land cover. Results show that green areas are cooler than urban areas, and dense forests witness lower temperatures than clear forests. Therefore, local analyses show that vegetation and altitude have a cooling effect, with temperatures dropping in the high and green mountains. Results of NDVI statistics show that urbanization has reduced the vegetation in the GBA with an 11% drop in green areas over the 3 decades. The vegetation cooling impact is demonstrated by negative spatial correlations between LST and NDVI. Finally, the urban heat index is higher in cities than in rural regions.

This study shows that UHI begins to appear in GBA due to the degradation of green spaces that reduce LST peak values. It highlights the need to consider urban heating in the legal urban planning code, which doesn’t strongly consider urban climatology or UHI.

Introduction. One of the main sources of errors in the recognition of forest cover classes from satellite data is the differences in the spectral brightness coefficients of identical objects in different parts of the image. To recognize the same classes of vegetation, it is proposed to process the scene, taking into account information about the diversity of forest growth conditions. The aim of the work is to develop a methodology for mapping of forest cover, taking into account the seasonal dynamics of the spectral-reflective characteristics of vegetation in different growing conditions.

Objects and methods. The object is the vegetation cover of the Sayano-Shushensky biosphere reserve (51 °46' - 52 °37' N, 91 °04' - 92 °26' E, area about 400 000 ha)is located in the mountainous part of southern Siberia. Landsat 8 (OLI) for 2016, DEM SRTM (90 m), forest inventory data for 2016, ground truth data and thematic maps.were used for forest mapping.

Data processing was carried out using software packages ArcGIS 10, ERDAS Imagine 2014, Trimble eCognition 8. For automated classification of spectral features of satellite data and terrain characteristics determined by DEM, both pixel approaches (ISODATA, MAXLIKE) and object-oriented segmentation method (Multiresolution) were used.

Results and Discussion. A conjugate classification of forest growth conditions and vegetation has been developed on the base of DEM-based topographic profiles crossing the territory of the reserve. Uncontrolled classification of DEM features (elevation above sea level and slope) by ISODATA method was carried out to identify land cover classes according to the conjugate classification. To generalize the obtained the classes, segmentation of the DEM (elevation above sea level) was performed. As a result of applying both methods, segments relatively uniform in elevation and texture of the relief were identified. The resulting classes were interpreted as geomorphological complexes (GMC) of forest growth conditions. These areas are homogeneous in terms of the ratio of mesorelief forms, underlying rocks, the range of elevation above sea level, dissection surface degree (they are similar in climatic and ecological regimes) and the predominant type of vegetation: tundra, subalpine and subalpine woodlands and sparse forests, mountain taiga forests, subtaiga forest-steppe complex.

The assessment of vegetation cover diversity of the reserve was carried out using the classification of the composite of satellite images Landsat-OLI 8 (June, September) for 2016 and layer of GMC of forest growth conditions. To classify the 16-channel composite within each GMC of forest growth conditions, training samples were created for forest vegetation classes and non-forest lands. The classification was performed using the MAXLIKE method.

The set of forest cover classes obtained in different GMC of forest growth conditions was combined into 9 classes according to conjugate classification: Siberian pine forests; Siberian pine forests with Fir/Spruce; Siberian pine forests with Larch; Larch forests with Siberian pine; Larch forests with Fir/Spruce; Larch forests; Larch forests with Scots pine (mixed with Siberian pine/Spruce); Scots pine forests with Larch; Spruce forests with Fir (mixed with Larch/ Siberian pine). Forest cover map was created at a scale of 1:200 000.

Landfills present a significant environmental challenge, with the potential to contaminate surrounding soil and affect vegetation growth if not maintained properly. This study investigated the impact of landfills on vegetation in the vicinity of Naples, Italy. The analysis employed both Sentinel-2 and Landsat-8 satellite imagery. The Landsat-8 images were processed using the Sen2like tool not only to enhance their resolution to 10 meters but also for harmonizing, enabling a comprehensive comparison with Sentinel-2 data. The images used in the study were from four years, i.e., 2019 to 2022, utilizing a range of vegetation health indices, including NDVI, GCI, and NDWI. An algorithm was developed to calculate these indices for all the images captured during the study period.

The study encompassed 17 landfill sites near Naples. These areas are classified based on the consistent presence of the same type of vegetation with similar phenological patterns in their surroundings. To evaluate the influence of landfills on vegetation, the area was divided into five circular buffer zones, with the landfills center as the epicenter each spanning 30 meters, resulting in a total coverage radius of 150 meters. Within the 12 landfills exhibiting consistent vegetation in the surroundings, polygons were drawn within each buffer zone. In contrast, for the remaining 5 landfills, the nearest vegetation within a radius of 150 meters from the landfill was considered, and the phenology across a span of 4 years was compared. These areas were carefully examined and analyzed by incorporating vegetation indices.

After studying the phenologies, it was discovered that vegetation near landfills showed less phenological growth than vegetation farther away. This distinctive pattern is observed in 2 out of the 17 landfills studied, thereby highlighting the impact of landfills on surrounding vegetation. It was observed that in the two affected areas, the vegetation near the landfill exhibited a 20% and 10% reduction in growth compared to the vegetation farther away from the landfill. Continuous monitoring is emphasized as an essential measure, and the use of satellite imagery, as demonstrated in this study, provides a viable approach for effective assessment.

In conclusion, the study establishes a correlation between landfills and their influence on nearby vegetation in the Naples region. Ongoing monitoring of these areas is essential, and satellite imagery offers a valuable tool for monitoring and analyzing the impact of landfills on vegetation. Further research should explore additional factors contributing to the health and sustainability of vegetation within landfill environments.

It is known that changes in temperature could cause changes in the Antarctic Ice Sheet, which would have an immediate and long-term impact on the global mean sea level [1]. For this reason, the monitoring of air temperature (T_a) is of great interest to the scientific community. On the other hand, Antarctica constitutes an area of difficult access, which makes it difficult to obtain in-situ data. Because of this, land surface temperature (LST) remote sensing data have become an important alternative for estimating T_a. In this work we estimate T_a from daytime and nighttime LST data at maritime Antarctic sites in the South Shetland Archipelago using empirical models, based on the addition of spatiotemporal variables [2]. We have used T_a data from the Spanish Antarctic stations and from the PERMASNOW project stations [3]. MOD11A1 and MYD11A1 (Collection 6) MODIS LST products were downloaded from the Google Earth Engine platform [4] and only the highest quality data were selected. Outliers associated with clouds were removed with filters. Two different multilinear regression models were tested: models for each individual station and global models based on the data from all the stations. The simple regression analysis LST against T_a showed that a better fit is always achieved with daytime LST data (R² average = 0.73) than with nighttime LST data (R² average = 0.56). The performance of the models was improved with the addition of spatiotemporal variables as predictive variables, with which we obtained an average R² = 0.75 for daytime data and an average R² = 0.60 for nighttime data. The global models allowed to improve the correlation and reduce the errors with respect to the models obtained using individual stations. Global models provide a precise description of the behavior of the temperature in maritime Antarctica, where it is not possible to install and maintain a dense network of weather stations.

References:

1. Medley, B.; Thomas, E.R. Increased snowfall over the Antarctic Ice Sheet mitigated twentieth-century sea-level rise. Nat. Clim. Chang. 2019, 9, 34–39, doi:10.1038/s41558-018-0356-x.
2. Recondo, C.; Corbea-Pérez, A.; Peón, J.; Pendás, E.; Ramos, M.; Calleja, J.F.; de Pablo, M.Á.; Fernández, S.; Corrales, J.A. Empirical Models for Estimating Air Temperature Using MODIS Land Surface Temperature ( and Spatiotemporal. Remote Sens. 2022, 14, 3206, doi:10.3390/rs14133206.
3. De Pablo, M.A.; Jiménez, J.J.; Ramos, M.; Prieto, M.; Molina, A.; Vieira, G.; Hidalgo, M.A.; Fernández, S.; Recondo, C.; Calleja, J.F.; et al. Frozen Ground and Snow Cover Monitoring in Livingston and Deception Islands, Antarctica: Preliminary Results of the - PERMASNOW Project. Cuad. Investig. Geográfica 2020, 46, 187–222, doi:http://doi.org/10.18172/cig.4381.
4. Gorelick, N.; Hancher, M.; Dixon, M.; Ilyushchenko, S.; Thau, D.; Moore, R. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens. Environ. 2017, 202, 18–27, doi:10.1016/j.rse.2017.06.031.

Human activities have led to the global redistribution of species, causing a worldwide decline in biodiversity, followed by the apparition of non-native species in natural environments, jeopardizing the normal function of the ecosystems and the apparition of invasive pests and new pathogens to crops and forests. The Canarian archipelago landscape, shaped by Canarian palm tree (Phoenix canariensis) groves, is being affected by this phenomenon, with their consequent decline.

The European Union Natura 2000 protection areas designated Phoenix canariensis groves as a priority habitat as an essential endemic Canary Islands plant species, and in this context, new tools to monitor and treat the pathologies that affect this species.

Traditional pathology diagnostic techniques are resource-demanding and poorly reproducible, and it is necessary to develop new monitoring methodologies. This study presents a tool to identify individuals infected by Serenomyces phoenicis and Phoenicococcus marlatti using UAV-derived multispectral images, machine learning, and probabilistic classification techniques. Two different study areas were selected in Tenerife and La Gomera islands, due to their representativity of the health status of Canarian palm groves.

In the first step, image segmentation was used to automatically identify palm tree specimens. In the second step, a pixel-based classification allowed us to assess the relative prevalence of affected leaves at an individual scale for each palm tree. The calculated affection prevalence ratio was later used alongside labelled in situ data depicting healthy and infected individuals, collected by expert technicians’ visual inspection to build a probabilistic classification model, capable of detecting infected specimens. Both the pixel classification performance and the model’s fitness were evaluated using different metrics such as omission and commission errors, accuracy, precision, recall, and F1-score.

An accuracy of more than 0.96 was obtained for the pixel classification of the affected and healthy leaves, and the probabilistic classification model presented good detection ability, reaching an accuracy of 0.87 for infected palm trees.

It is worth considering that the developed algorithms and the infection detection model could allow for the cost-effective identification of infected palm trees by implementing transfer learning procedures in new study areas. This will imply a drastic decrease in data requirements, facilitating future palm groves' extensive monitoring in the archipelago, and significantly reducing phytosanitary treatment costs.

Under the current high anthropogenic pressures, phytoplankton blooms are increasing in waters all over the world. This necessitates monitoring and management actions to prevent its environmentally negative impact. An efficient approach for monitoring water quality can be achieved through remote sensing. Satellite multispectral data acquired from Sentinel-2 were used to assess chlorophyll-a (chl-a) concentration in Karaoun Reservoir, largest water body in Lebanon. Radiometric and atmospheric corrections were applied to downloaded level 1 satellite images. Pre-processing steps of Sentinel-2 images consisted of radiometric calibration on SNAP, resampling bands on ENVI, atmospheric correction on 6S (Second Simulation of the Satellite Signal in the Solar Spectrum) and applying algorithm on ArcGIS.

Comparison between 6S code corrected images and field spectral signatures showed high correlation with R² = 0.74 and R² = 0.82 on two of validation dates indicating a good precision of the 6S code. Several simple linear regression algorithms were tested with all possible Sentinel-2 single bands and band combinations from band 1 to 8. A simple linear algorithm was then developed by comparing in-situ measurements to single bands and band ratio. On a single band level, Band 5 was correlated the most with in situ PC measurements with R²= 0.69. For band combinations, the best fit between bands reflectance and actual PC measurements was found for the band ratio B5/B4 with R²=0.862. Based on these findings, the empirical band ratio model was developed using a Red band 4 of spectral resolution (650-680 nm) with Vegetation Red Edge band 5 (698-713 nm) to estimate chl-a at Karaoun Reservoir. The algorithm gave good estimations for chl-a detection with R² = 0.86 for calibration and R² = 0.8 for validation. The algorithm was then used to map and investigate the spatial-temporal distribution of phytoplankton bloom throughout the reservoir. This multi-disciplinary approach can be applied to other water bodies as advanced monitoring and management approach.

Effective management of reforestation projects faces the challenges posed by deforestation and evolving environmental conditions, necessitating accurate tree health monitoring.

This study introduces a holistic methodology that synergizes high-resolution satellite imagery from Planet and historical data from Sentinel 2 with meteorological insights extracted from ERA5 data. By computing vital vegetation indices (NDVI, NDWI, mSAVI2) and meteorological indices (SPI, KBDI), we establish customized growing conditions, enabling the prediction and continuous monitoring of tree health and stress. The approach integrates time series models for temperature, precipitation, and vegetation indices, augmenting the understanding of growing conditions and facilitating informed site selection for reforestation initiatives. Satellite data is sourced from Copernicus (Sentinel 2 using GEE) and Planet imagery (via QGIS plugin). Copernicus Climate Data Store (ERA5) provides meteorological and climate assimilation data, complemented by reforestation specifics such as tree counts and planting timelines.

Our workflow involves data preparation, including the computation of vegetation and meteorological indices. The process follows several logical steps, which are either done sequentially or (where possible) in parallel. First, by using Facebook Prophet, a baseline reference is constructed starting from 2017, aiding in subsequent automatic forecasting. This baseline is used to compare the predicted NDVI against location-specific averages to gain insights into vegetation dynamics across multiple time frames. Utilizing Planet-NICFI monthly surface reflectance, we aggregate vegetation indices into a 100m grid, facilitating effective site comparisons. NDVI, NDWI, and mSAVI2 contribute to multi-dimensional vegetation assessment, detecting robust vegetation cover, optimal moisture levels, and early crop stages, respectively.

ERA5 temperature and precipitation data are downscaled to derive KBDI and SPI indices, enhancing our understanding of drought severity and precipitation patterns. Time series models, employing the Facebook Prophet library, forecast temperature and precipitation trends at specific sites, enabling the anticipation of future weather-driven indices.

Results demonstrate a versatile framework for regional forecasting adaptable to diverse scenarios. By leveraging the predictive power of NDVI, temperature, and precipitation, we effectively forecast NDVI six months ahead. This predictive capability empowers informed decision-making in reforestation, agriculture, and land management. The integrated approach presents a user-friendly means for stakeholders to gauge tree health and stress, contributing to targeted and sustainable environmental interventions.

In conclusion, this study showcases the potential of integrated satellite imagery and meteorological data for accurate tree health prediction in reforestation endeavours. Our approach emphasizes the value of data synergy and predictive modelling, offering promise for advancing sustainable reforestation practices and promoting ecological resilience.