Vegetation monitoring and mapping based on multi-temporal imagery has recently received much attention due to the plethora of medium-high spatial resolution satellites and the improved classification accuracies attained compared to uni-temporal approaches. Efficient image processing strategies are needed to exploit the phenological information present in temporal image sequences and to limit data redundancy and computational complexity. Within this framework, we implement the theory of Hidden Markov Models in crop classification, based on the time-series analysis of phenological states, inferred by a sequence of remote sensing observations. More specifically, we model the dynamics of vegetation over an agricultural area of Greece, characterized by spatio-temporal heterogeneity and small-sized fields, using RapidEye and Landsat ETM+ imagery. In addition, the classification performance of image sequences with variable spatial and temporal characteristics is evaluated and compared. The classification model considering one RapidEye and four pan-sharpened Landsat ETM+ images was found superior, resulting in a conditional kappa from 0.77 to 0.94 per class and an overall accuracy of 89.7%. The results highlight the potential of the method for operational crop mapping in Euro-Mediterranean areas and provide some hints for optimal image acquisition windows regarding major crop types in Greece.

Rural abandonment and associated rapid urbanization, agricultural intensification, and climate change have been key factors transforming terrestrial landscapes, with significant impacts on the environmental quality and the ecosystem services (ES) provided to human welfare. In this study, two understudied rural river basins located in Thrace, North Greece, were selected to assess changes in landscape pattern, composition, and eco-environmental quality and ecosystem services values (ESV). Cloud-based remote sensing (RS) analyses of multitemporal Landsat imagery in the Google Earth Engine (GEE) platform were applied for multitemporal land use/land cover changes (LULCC) quantification between 1984–2021, along with landscape pattern, eco-environmental quality, and ES assessment. Although ecosystem changes observed were not extensive over this period, eco-environmental quality appeared to be affected due to fragmentation. Preserving the ecosystem’s naturalness can enhance cultural ES to avoid further values loss originating from provisioning ES exploitation. This study highlights the strong connection between landscape configuration and eco-environmental quality, emphasizing the strong impact that anthropogenic activities have on the environment. The monitoring of the effects of LULCC on ecosystem health and the economic value of ES is crucial for the introduction of spatial planning and restoration policies.

Plantations of fast-growing forest species such as black locust (Robinia Pseudoacacia) can contribute to energy transformation, mitigate industrial pollution, and restore degraded, marginal land. In this study, the synergistic use of Sentinel-2 and Sentinel-1 time series data is explored for modeling aboveground biomass (AGB) in black locust short-rotation plantations in northeastern Greece. Optimal modeling dates and EO sensor data are also identified through the analysis. Random forest (RF) models were originally developed using monthly Sentinel-2 spectral indices, while, progressively, monthly Sentinel-1 bands were incorporated in the statistical analysis. The highest accuracy was observed for the models generated using Sentinel-2 August composites (R² = 0.52). The inclusion of Sentinel-1 bands in the spectral indices’ models had a negligible effect on modeling accuracy during the leaf-on period. The correlation and comparative performance of the spectral indices in terms of pairwise correlation with AGB varied among the phenophases of the forest plantations. Overall, the field-measured AGB in the forest plantations plots presented a higher correlation with the optical Sentinel-2 images. The synergy of Sentinel-1 and Sentinel-2 data proved to be a non-efficient approach for improving forest biomass RF models throughout the year within the geographical and environmental context of our study.

The images of the Sentinel-2 constellation can help the verification process of farmers’ declarations, providing, among other things, accurate spatial explicit maps of the agricultural land cover. The aim of the study is to design, develop, and evaluate two deep learning (DL) architectures tailored for agricultural land cover and crop type mapping. The focus is on a detailed class scheme encompassing fifteen distinct classes, utilizing Sentinel-2 imagery acquired on a monthly basis throughout the year. The study’s geographical scope covers a diverse rural area in North Greece, situated within southeast Europe. These architectures are a Temporal Convolutional Neural Network (CNN) and a combination of a Recurrent and a 2D Convolutional Neural Network (R-CNN), and their accuracy is compared to the well-established Random Forest (RF) machine learning algorithm. The comparative approach is not restricted to simply presenting the results given by classification metrics, but it also assesses the uncertainty of the classification results using an entropy measure and the spatial distribution of the classification errors. Furthermore, the issue of sampling strategy for the extraction of the training set is highlighted, targeting the efficient handling of both the imbalance of the dataset and the spectral variability of instances among classes. The two developed deep learning architectures performed equally well, presenting an overall accuracy of 90.13% (Temporal CNN) and 90.18% (R-CNN), higher than the 86.31% overall accuracy of the RF approach. Finally, the Temporal CNN method presented a lower entropy value (6.63%), compared both to R-CNN (7.76%) and RF (28.94%) methods, indicating that both DL approaches should be considered for developing operational EO processing workflows.

Freshwater ecosystems provide an array of provisioning, regulating/maintenance, and cultural ecosystem services. Despite their crucial role, freshwater ecosystems are exceptionally vulnerable due to changes driven by both natural and human factors. Water quality is essential for assessing the condition and ecological health of freshwater ecosystems, and its evaluation involves various water quality parameters. Remote sensing has become an efficient approach for retrieving and mapping these parameters, even in optically complex waters such as small rivers. This study specifically focuses on modelling two non-optically active water quality parameters, dissolved oxygen (DO) and electrical conductivity (EC), by integrating 3 m PlanetScope satellite imagery with data from real-time in situ remote monitoring sensors across two small rivers in Thrace, Northeast Greece. We employed three different experimental setups using a support vector regression (SVR) algorithm: ‘Multi-seasonal by Individual Sensor’ (M-I-S) for individual sensor analysis across two seasons, ‘Multi-seasonal—All Sensors’ (M-A-S) integrating data across all seasons and sensors, and ‘Seasonal—All Sensors’ (S-A-S) focusing on per-season sensor data. The models incorporating multiple seasons and all in situ sensors resulted in R² values of 0.549 and 0.657 for DO and EC, respectively. A multi-seasonal approach per in situ sensor resulted in R² values of 0.885 for DO and 0.849 for EC. Meanwhile, the seasonal approach, using all in situ sensors, achieved R² values of 0.805 for DO and 0.911 for EC. These results underscore the significant potential of combining PlanetScope data and machine learning to model these parameters and monitor the condition of ecosystems over small river surfaces.

The successful implementation of the European Commission’s Common Agricultural Policy (CAP) and the insurance coverage in case of a natural disaster requires precise and regular mapping of crop types and detailed delineation of the disasters’ effects by frequent and accurate controls. Free and open access policy to Copernicus Sentinel data offers a big volume of data to the users on a consistent and complete basis. Today, the Sentinels are involved in an increasing number of agriculture applications, but their effective exploitation is still being investigated and the development of efficient tools, aligned to the user’s needs, is yet to be realised. To this end, the DiAS (Disaster and Agriculture Sentinel Applications) project proposes methods for decision support in agriculture using Sentinel data for crop type mapping, as well as mapping of the extend of fire and flood effects in agricultural areas. The DiAS Decision Support System (DSS) is designed in consultation with potential users in participatory approach and aims to provide a prototype tool, which provides assistance to the responsible paying agencies and insurance organizations to make decisions on farmers’ subsidies and compensations. The DiAS DSS prototype and its functionalities are presented in this paper and its use is demonstrated through example applications for two test sites in Greece. The DiAS DSS demonstrates the necessity for the development of similar tools, as this emerges from the user’s requirements, and wishes to stimulate and inspire further research and development.