The Upper Parana Atlantic Forest (BAAPA) in Paraguay is one of the most threatened tropical forests in the world. The rapid growth of deforestation has resulted in the loss of 91% of its original cover. Numerous efforts have been made to halt deforestation activities, however farmers’ perception towards the forest and its benefits has not been considered either in studies conducted so far or by policy makers. This research provides the first multi-temporal analysis of the dynamics of the forest within the BAAPA region on the one hand, and assesses the way farmers perceive the forest and how this influences forest conservation at the farm level on the other. Remote sensing data acquired from Landsat images from 1999 to 2016 were used to measure the extent of the forest cover and deforestation rates over 17 years. Farmers’ influence on the dynamics of the forest was evaluated by combining earth observation data and household survey results conducted in the BAAPA region in 2016. Outcomes obtained in this study demonstrate a total loss in forest cover of 7500 km². Deforestation rates in protected areas were determined by management regimes. The combination of household level and remote sensing data demonstrated that forest dynamics at the farm level is influenced by farm type, the level of dependency/use of forest benefits and the level of education of forest owners. An understanding of the social value awarded to the forest is a relevant contribution towards preserving natural resources.

The Paraguayan Chaco is one of the most rapidly deforested areas in Latin America, mainly due to cattle ranching. Continuously forested windbreaks between agricultural areas and forest patches within these areas are mandatory to minimise the impact that the legally permitted logging has on the ecosystem. Due to the large area of the Paraguayan Chaco, comprehensive in situ monitoring of the integrity of these landscape elements is almost impossible. Satellite-based remote sensing offers excellent prerequisites for large-scale land cover analyses. However, traditional methods mostly focus on spectral and texture information while dismissing the geometric context of landscape features. Since the contextual information is very important for the identification of windbreak gaps and central forests, a deep learning-based detection of relevant landscape features in satellite imagery could solve the problem. However, deep learning methods require a large amount of labelled training data, which cannot be collected in sufficient quantity in the investigated area. This study presents a methodology to automatically classify gaps in windbreaks and central forest patches using a convolutional neural network (CNN) entirely trained on synthetic imagery. In a two-step approach, we first used a random forest (RF) classifier to derive a binary forest mask from Sentinel-1 and -2 images for the Paraguayan Chaco in 2020 with a spatial resolution of 10 m. We then trained a CNN on a synthetic data set consisting of purely artificial binary images to classify central forest patches and gaps in windbreaks in the forest mask. For both classes, the CNN achieved an F1 value of over 70%. The presented method is among the first to use synthetically generated training images and class labels to classify natural landscape elements in remote sensing imagery and therewith particularly contributes to the research on the detection of natural objects such as windbreaks.