Since the 20th century, satellites have been a mainstay in remote sensing. However, with the dawn of the 21st century, Unmanned Aerial Vehicles (UAVs) have emerged as pivotal remote sensing instruments across various sectors, notably in agriculture. Despite satellites and UAVs being complementary tools, they are often utilized without full comprehension of their implications. Precision agriculture, a field keen on maximizing productivity and efficiency, necessitates a clear understanding of the benefits and limitations of each technology, especially when applied to specific crops. This becomes paramount in the context of changing global climate patterns, where crops like pistachio are burgeoning in Southern Europe, notably in regions like Spain. Yet, the knowledge gap regarding the application of cutting-edge remote sensing technology in these crops remains significant. This study juxtaposes the use of satellite remote sensing, specifically Sentinel 2, with UAV remote sensing. This comparative analysis is conducted by evaluating agronomical parameters, such as vegetation indices, across two distinct pistachio orchards in Spain over multiple dates. Our findings highlight a discernible correlation between UAV and satellite data. However, satellite imagery consistently produced underestimations in variable values relative to UAV. This suggests that while satellite imagery might be apt for extensive crops, for woody plantations like pistachio orchards, UAVs present a more judicious choice. Leveraging UAVs can mitigate noise and provide more reliable crop-specific data, thereby ensuring more informed agronomical decisions.

Protected by international, European and local acts, Zostera marina L. meadows, a marine flowing plant (angiosperm), are subject to considerable human pressure related to their geographical distribution: abrasion of anchor chains, trawling due to professional or recreational boat fishing, trampling due to foot fishing at low tide. This study takes part of the European Life Impact project to assess the disturbance/stress of human activity on the subtidal seagrass meadow at La Varde (48°40’59 N ; 1°59’13 W) in the commune of Saint-Malo in Brittany.

The aim of this study is to monitor the surface evolution and fragmentation of the seagrass meadow on a fine spatial and temporal scale using a drone. three drone campaigns per year were carried out on 2021 and 2022. Cross-analysis of underwater truths by snorkeling and drone imagery in Red-Green-Blue natural colors were used to extract the overall envelope of the seagrass meadow. Five classes were identified (immersed seagrass, emerged/immersed sediment, emerged rock, and immersed macroalgae) and a machine learning algorithm, namely the maximum likelihood, was processed.

Preliminary results have revealed a loss of 465.18 m² at annual scale, between 2021 and 2022 of the seagrass meadow. Analysis of the results on a seasonal scale highlights a shrinkage of the meadow during the winter period and an expansion during the summer period, i.e. a differential of 382.4 m² between March and September 2021 and 370.64 m² between February 2022 and October 2022. Seagrass meadow fragmentation has measured (fragmented envelope area) and highlight a difference of -12.15 m² between 2021 and 2022. These temporal variations are attributable by the sensitivity of the Zostera marina L. species to variations in sedimentation and turbidity, which are more pronounced in winter due to the higher level of hydrodynamism at this time of year. As regards fragmentation, the meadow recovery can be explained by the protection measures adopted, such as restricting fishing on foot and setting up ecological moorings.

Ensuring food security in precision agriculture demands early prediction of corn yield in the United States of America at international, regional, and local levels. Accurate yield estimation can play a crucial role in averting famine by offering insights into food availability during the growing season. To address this, we proposed a Concatenate-based 2D-CNN-BILSTM model that integrates Sentinel-1, Sentinel-2, and SoilGRIDs (global gridded soil information) data for corn yield estimation in Iowa State from 2018 to 2021. Sentinel-2 images provided essential bands and indices, such as Blue, Green, Red, and Red Edge 1/2/3/4, NIR, and SWIR 1/2 Bands, along with NDVI, LSWI, DVI, RVI, DWRVI, SAVI, VARIGREEN, and GNDVI. Additionally, VV, VH, difference VV, and VH, and RVI were extracted from Sentinel-1 SAR images. Soil data encompassing silt, clay, sand, cec, and pH were collected at various depths ranging from 0 cm to 200 cm. To extract high-level features for each month, a dedicated 2D-CNN was designed. The 2D-CNN concatenated high-level features from the previous month with low-level features of the subsequent month, serving as input features for the 2D-CNN. To incorporate single-time soil data features, another 2D-CNN was created. Finally, the high-level features from soil, Sentinel-1, and Sentinel-2 data were concatenated and fed into a BILSTM layer to predict corn yield. The performance of the proposed Concatenate-based 2D-CNN-BILSTM model was compared against random forest (RF), Concatenate-based 2D-CNN, and 2D-CNN models using some metrics like RMSE, MAE, MAPE, and the Index of Agreement. The model was trained on data from 2018, 2019, and 2020, and its accuracy was tested using data from 2021. Results revealed the Concatenate-based 2D-CNN-BILSTM model's impressive Index of Agreement of 84.67% and low RMSE of 0.698 t/ha. Furthermore, validation demonstrated high prediction accuracy for the Concatenate-based 2D-CNN model, with an RMSE of 0.799 t/ha and an Index of Agreement of 72.71%. The 2D-CNN model also performed well, yielding an RMSE of 0.834 t/ha and an Index of Agreement of 69.90%. However, the RF model exhibited lower accuracy with an RMSE of 1.073 t/ha and an Index of Agreement of 69.60%. These findings highlight the potential of utilizing advanced deep-learning techniques in conjunction with remote sensing and soil data to enhance crop yield predictions. The integration of Sentinel 1-2 and SoilGRIDs data with the proposed 2D-CNN-BiLSTM model demonstrated significant improvements in accuracy for corn yield prediction. The combination of soil data and features extracted from Sentinel 1-2 resulted in a decrease in RMSE by 16 kg and an increase in the Index of Agreement by 2.59%. This study demonstrates the power of leveraging advanced Machin Learning (ML) methods for achieving accurate and reliable predictions to support sustainable agricultural practices and food security initiatives.

The coastal zone represents varied and highly productive ecosystems such as mangroves, coral reefs, sea grasses and sand dunes. However, as a result of globalization, anthropological activities have increased on the coastal areas putting these ecosystems under high pressure. This, in turn, has lead to the loss of valuable vegetation along the coastal areas of the world. This study was taken up to detect the change occurring in coastal vegetation of Daman district of India. Daman is one of the Union territories of India which have shown a good development in recent years. As a result, area under the mangrove vegetation has changed at and near the coast of the district. Remote sensing approach was utilized in this study to detect the changes occurred in the vegetation between the years 2016 to 2021. Landsat ETM+ data was used to derive NDVI images of the study period using ERDAS imagine 2014. Field work covering the entire study area was carried out for classifying and accuracy assessment of the vegetation categories, i.e. no vegetation, low vegetation, moderate vegetation and dense vegetation. Vegetation maps for both the years were prepared using ArcGIS software. Results indicated that area under the no vegetation decreased during the study period whereas rest all categories, i.e. low vegetation, moderate vegetation and dense vegetation showed increase. The increase in the vegetation can be attributed to efforts taken up by the Daman official authorities for conserving the coastal areas. This will lead to enhanced ecosystem services provided by these ecosystems.

Increased world population growth will demand more high-quality food production, which can only be achieved by applying a sustainable method for increasing crop yields. According to the Food and Agriculture Organization (FAO) report, weed grasses increase the environmental and economic costs of pesticide use by spreading them across farm boundaries, and their competition with agricultural crops reduces quantity and quality output. Among the pests, weed grasses are considered a crucial biotic constraint to food production. In traditional pest control methods in agriculture, most farm fields are spatially variable in grass weed infestation to a certain degree, but general weed management methods for herbicide application are based on the assumption that grass weeds are distributed uniformly in agricultural fields. However, a smart weed localization system for optimized herbicide dose in the agricultural field is a crucial step for smart farming and is still an open problem in pest control methods. While many object detection models appear to understand localization with a huge training data, a few-shot learning strategy potentially improves scene understanding with limited training data. In this study, the purpose of weed grasses localization from drone-based multispectral images is to locate the weed on large-scale images by using pixel-wise classification. Weed grasses localization with the use of uncertainty modeling in few-shot learning for drone-based multispectral images potentially improves multispectral scene understanding with a small training dataset, while many weed detection methods appear to understand single-time localization with a big training dataset. Few-shot learning can perform on unseen tasks after training a few annotated data and considers several tasks to produce a predictive function, and is an inductive transfer system whose main goal is to improve generalization ability for multiple tasks. Weed grasses localization of the trained model can be just blindly assumed accurate but the truth is not for decision making.

Reproductive strategies of most plant species in Mediterranean ecosystems exhibit efficient mechanisms of germination and/or resprouting, ensuring the development of formations similar to those affected by fire based on models of plant succession or auto-succession. The assessment of fire effects and recovery becomes a key element in guiding the strategic orientation of burned areas, encompassing comprehensive management, adaptation, mitigation, and more. In this context, the design of operational methodologies based on the processing of UAV-LiDAR data holds great interest due to the quality of the structural information they provide.

This study presents a methodology for quantifying the malleability of resilience in burned areas through statistical analysis of dasometric parameters derived from UAV-LiDAR data. Flights were conducted over burned areas (fires in 1970, 1995, and 2008 in Pinus halepensis L. forests with Quercus coccifera L. located in Montes de Zuera, Aragón, Spain) and their respective controls (i.e., nearby unaffected areas with the same pre-disturbance characteristics). As a matrix-based linear approach is not applicable for analysis (due to uni-temporal images lacking corresponding elements), a methodology was designed that involved the following phases: (1) flights using an unmanned aerial vehicle (DJI Matrice 300 RTK) equipped with a DJI Zenmuse-L1 LiDAR sensor to analyze two forest attributes: tree height (99th percentile) and Profile Area Change (PAC, a multitemporal LiDAR metric introduced by Hu et al., 2019), using “DJI Terra v.3.6.7” and “FUSION-LDV v.4.21” software; (2) field-based floristic and physiognomic inventories conducted concurrently with the flights, utilizing sampling units of 20m² (10x2m); (3) extraction of random samples (10%) in equally sized quadrangular sectors, both burned and control; (4) application of basic statistics, construction of frequency distribution diagrams, and similarity analysis using the Kolmogorov-Smirnov test.

The maximum vegetation height ranges from 9 to 14 m in control areas, and from 4 to 12 m in burned sectors. Significant differences have been identified in the distributions of vegetation height (99th percentile) among the three fires, exhibiting specific maximum absolute differences (D) depending on the fire year (D = 0.26, 0.31, 0.76 for 1970, 1995, and 2008 fires, respectively), resulting in differences of approximately 2, 0.25, and 5 m between P. halepensis plant communities. Regarding the PAC index, the average values are -3.9, -22.5, and 33.7. Positive values in 2008 indicate greater LiDAR pulse penetration in the burned area, consequently leading to lower regeneration. Negative values in 1995 identify greater complexity and density of regenerated vegetation, while values close to 0 for the 1970 fire indicate greater homogeneity between regenerated structures and their control.

The use of LiDAR metrics and uni-temporal sampling between burned sectors and their corresponding controls facilitates an understanding of the resilience of these communities and the identification of different stages in the recovery process of P. halepensis forests. Considering other contextual variables, such as fire severity, post-fire hygrothermal characteristics, or anthropogenic treatments, may provide new insights into characterizing the malleability of burned Aleppo pine forests.

Seagrass meadows, recognised as powerful carbon sinks and crucial players in climate change mitigation, face significant threats from global warming and anthropogenic activities. Traditional in situ monitoring methods, while accurate, are often expensive, time-consuming, and constrained by the extent of coverage. Thus, this research aims to evaluate the potential of freely accessible Sentinel-2 satellite imagery in mapping and monitoring Cymodocea nodosa seagrass meadows in El Médano (Tenerife, Canary Islands), contributing to ecosystem conservation efforts. The study employed an image from October 27, 2022 processed at Level-1C. This leads to significant challenges due to the optical signal’s attenuation caused by the atmosphere and the water column. The atmospheric correction was addressed by employing the Sen2Cor tool within the Sentinel Application Platform (SNAP). For the water column effect, Lyzenga’s method was used. This method is based on Beer-Lambert’s absorption law, which specifies a log-linear relationship between reflectance values and water depth. Following these corrections, supervised classifications were conducted using the Random Forest, K-Nearest Neighbors (KNN), and KDTree-KNN algorithms, supplemented by unsupervised classifications and in situ data. The blue carbon sequestered by the C. nodosa in the study area was also computed using the InVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) software. The Random Forest classifiers produced the highest F1-scores, ranging between 0.96 and 0.99. Results revealed an average area of 237±5 ha occupied by the C. nodosa in the study region, which translates to an average sequestration of 111,349±2,330 Mg of CO₂. Even at the lowest sequestration level, the seagrass meadows of this study area have the potential to offset the CO₂ emissions produced by the industrial combustion plants sector across the Canary Islands. This research represents a significant step in protecting and comprehending these invaluable ecosystems. It effectively underscores the potential of Sentinel-2 satellite data for mapping seagrass meadows and emphasises their crucial role in achieving net zero carbon emissions on our planet.

Remote sensing plays a pivotal role in Precision Agriculture by providing relevant information regarding crop diseases and deficiencies, and fruit counting, which can derive yield estimation. Unmanned Aerial Vehicles (UAVs) offer a great opportunity for Remote Sensing in order to acquire high-spatial-resolution data. Nevertheless, challenges arise when it comes to accurate object detection in crops that present leaf-occlusion. Moreover, there is a limitation while capturing information with only one UAV since the flight time is restricted by the battery level. However, using multiple UAV can provide an augmented agricultural-scene understanding to address occlusion problems. This study presents a novel approach to address these challenges by enhancing UAV path planning specifically designed for fruit detection in woody crops trained in vertical trellis, considering the biophysical environment of the field. The experiments were carried out in a vineyard (Vitis vinifera cv. Loureiro) located in the North-West of Spain. The proposed method implements the Ant Colony Optimization (ACO) algorithm to optimize data collection and enable single and multiple UAVs to fly synchronously while ensuring a safety distance between platforms and achieving efficient coverage of the agricultural area. In order to enhance data collection for fruit detection purposes, the methodology incorporates a two-flight strategy. The first flight (with 1 UAV) serves as a survey to comprehend and analyze the crop structure and environmental conditions of the agricultural field. In this step, the field is discretized as waypoints (areas to visit) and forbidden areas, which include also areas without agronomic interest. Further, the second flight (with n UAVs) is executed following the optimal path between waypoints. Also, it enhances image acquisition by considering multiple angles, effectively mitigating the adverse effects of partial leaf-occlusion. The results obtained from this study highlight that ACO is able to generate optimal and safe routes within the field by covering the whole agricultural area while flying one or multiple UAV platforms. Moreover, it shows potential to solve partial leaf-occlusion for fruit identification.

Accurate computations of the radiance in the gaseous absorption bands typically require fine wavelength steps.
In this paper, we present a fast technique that allows us to compute a spectrum of the radiation reflected by the terrestrial atmosphere.
The technique is based on the $k$-correlation distribution model (where $k$ stands for the absorption coefficient). While the classical $k$-distribution model takes into account only the dependency of the radiance on $k$, the presented model takes as a predictor the direct transmittance and the scattering coefficient.
At selected spectral points, the full radiative transfer simulations are performed and the mathematical relation between a predictor and the radiance is established. Then, the radiance is restored on a fine wavelength grid. This approach can be used to enhance the accuracy of the convolved spectrum computations based on precomputed monochromatic lookup tables.
The numerical analysis shows that the method can be applied to cases with aerosol optical thickness not larger than 2.

The employment of satellite data in remote sensing applications has become paramount for Precision Agriculture, particularly in woody crops. Satellite-based products are widely used in various applications to gather valuable information about crop conditions, providing potential valuable insights for a range of applications such as forecasting yield, predicting crop quality, and managing irrigation. However, the accuracy of satellite-derived products for precision agriculture purposes, such as leaf area and Normalized Difference Vegetation Index (NDVI) is often debated. This study aims to benchmark the satellite data accuracy against UAV LiDAR data and multispectral imagery. In this study, leaf area and NDVI were derived from Sentinel-2 images and compared with ground truth data estimated from the UAV LiDAR dataset collected in 2022 close to veraison, as this is a key phenological stage commonly used in remote sensing for precision viticulture. The UAV flights were conducted over a commercial vineyard, Vitis vinifera cv. Loureiro, located on northern Spain, using a DJI M300 multi-rotor platform equipped with a DJI Zenmuse L1 LiDAR sensor and a Micasense Altum camera. The results indicate a discrepancy in the leaf area estimated from satellite imagery, demonstrating its limited utility for assessing leaf area in woody crops conducted in hedgerow systems. Nevertheless, satellite data could discern spatial patterns and variability within the crop. Thus, while satellite-based remote sensing may not serve as the best tool for leaf area estimation in this context, its capacity to detect crop spatial heterogeneity remains valuable for overall field management and the creation of distinct zones for differentiated management. This research prompts a discussion about the suitability of solely relying on satellite imagery for managing agriculture and underscores the necessity to incorporate a range of remote sensing methodologies for efficiently managing vineyards cultivated in hedgerow systems.