The development and improvement of methods to map agricultural land cover are currently major challenges, especially for radar images. This is due to the speckle noise nature of radar, leading to a less intensive use of radar rather than optical images. The European Space Agency Sentinel-1 constellation, which recently became operational, is a satellite system providing global coverage of Synthetic Aperture Radar (SAR) with a 6-days revisit period at a high spatial resolution of about 20 m. These data are valuable, as they provide spatial information on agricultural crops. The aim of this paper is to provide a better understanding of the capabilities of Sentinel-1 radar images for agricultural land cover mapping through the use of deep learning techniques. The analysis is carried out on multitemporal Sentinel-1 data over an area in Camargue, France. The data set was processed in order to produce an intensity radar data stack from May 2017 to September 2017. We improved this radar time series dataset by exploiting temporal filtering to reduce noise, while retaining as much as possible the fine structures present in the images. We revealed that even with classical machine learning approaches (K nearest neighbors, random forest, and support vector machines), good performance classification could be achieved with F-measure/Accuracy greater than 86% and Kappa coefficient better than 0.82. We found that the results of the two deep recurrent neural network (RNN)-based classifiers clearly outperformed the classical approaches. Finally, our analyses of the Camargue area results show that the same performance was obtained with two different RNN-based classifiers on the Rice class, which is the most dominant crop of this region, with a F-measure metric of 96%. These results thus highlight that in the near future these RNN-based techniques will play an important role in the analysis of remote sensing time series.

In many areas of the world, maintaining grapevine production will require adaptation to climate change. While rigorous evaluations of adaptation strategies provide decision makers with valuable insights, those that are published often overlook major constraints, ignore local adaptive capacity, and suffer from a compartmentalization of disciplines and scales. The objective of our study was to identify current knowledge of evaluation methods and their limitations, reported in the literature. We reviewed 111 papers that evaluate adaptation strategies in the main vineyards worldwide. Evaluation approaches are analyzed through key features (e.g., climate data sources, methodology, evaluation criteria) to discuss their ability to address climate change issues, and to identify promising outcomes for climate change adaptations. We highlight the fact that combining adaptation levers in the short and long term (location, vine training, irrigation, soil, and canopy management, etc.) enables local compromises to be reached between future water availability and grapevine productivity. The main findings of the paper are three-fold: (1) the evaluation of a combination of adaptation strategies provides better solutions for adapting to climate change; (2) multi-scale studies allow local constraints and opportunities to be considered; and (3) only a small number of studies have developed multi-scale and multi-lever approaches to quantify feasibility and effectiveness of adaptation. In addition, we found that climate data sources were not systematically clearly presented, and that climate uncertainty was hardly accounted for. Moreover, only a small number of studies have assessed the economic impacts of adaptation, especially at farm scale. We conclude that the development of methodologies to evaluate adaptation strategies, considering both complementary adaptations and scales, is essential if relevant information is to be provided to the decision-makers of the wine industry.