Supporting Pro-Poor Reforms of Agricultural Systems in Eastern DRC (Africa) with Remotely Sensed Data: A Possible Contribution of Spatial Entropy to Interpret Land Management Practices

Abstract

In the eastern Democratic Republic of Congo, agriculture represents the most important economic sector, and land control can be considered a perpetual source of conflict. Knowledge of the existing production system distribution is fundamental for both informing national land tenure reforms and guiding more effective agricultural development interventions. The present paper focuses on existing agricultural production systems in Katoyi collectivity, Masisi territory, where returning Internally and Externally Displaced People are resettling. We aim to define a repeatable methodology for building evidence-based and updated knowledge concerning the spatial distribution of the two existing production systems: subsistence-oriented agriculture (SOA) and business-oriented agriculture (BOA). To this aim, we used a supervised object-based classification approach on remotely sensed Sentinel-2 imagery to classify land cover. To classify production systems further within the “agriculture” and “pasture” land use classes, binary classification based on an entropy value threshold was performed. An iterative approach was adopted to define the final HNDVI threshold that minimised commission and omission errors and maximised overall accuracy and class separability. The methodology achieved acceptable observed accuracy (OA equal to 80–90% respectively for agricultural and pasture areas) in the assessment. SOA and BOA respectively covered 24.4 and 75.6% of the collectivity area (34,606 ha). The results conclude that land use and entropy analysis can draw an updated picture of existing land distribution among different production systems, supporting better-adapted intervention strategies in development cooperation and pro-poor agrarian land tenure reforms in conflict-ridden landscapes.

Snow Metrics as Proxy to Assess Sarcoptic Mange in Wild Boar: Preliminary Results in Aosta Valley (Italy)

“…Therefore, the results obtained must be considered as preliminary and deserve more in-depth studies and insights by the scientific community in order to achieve solid evidence on samples obtained over several years and in larger areas confirming or confuting the geospatially based results here obtained. Having said this, the use of remote sensing and GIS tools in the veterinary and eco-epidemiological fields can certainly constitute an added value to ordinary diagnostic techniques [59][60][61][62][63][64][65]. The technological transfer offered by spatial analyzes in terms of repercussions on the veterinary sector as highlighted in this study can certainly be positive.…”

Probabilistic Tracking of Annual Cropland Changes over Large, Complex Agricultural Landscapes Using Google Earth Engine

“…As such, the only feasible way to undertake this task is with remote sensing. However, developing accurate, remotely sensed maps of Africa's smallholder-dominated agricultural systems is hard to accomplish [11][12][13][14].…”

Ecology and Epidemiology of Lyme Disease in Western North America

“…Remote sensing, geographic information systems (GIS), and modelling approaches can be useful in delineating the interactions between wildlife, livestock, and humans in the context of disease transmission [126][127][128][129][130]. A common way to apply knowledge of competent tick vectors and reservoir hosts for B. burgdorferi to Lyme disease surveillance efforts is to develop species distribution models that determine the geographical distribution of the tick species and pathogen of interest.…”