Review of Remote Sensing Methods to Map Coffee Production Systems

Abstract

The coffee sector is working towards sector-wide commitments for sustainable production. Yet, knowledge of where coffee is cultivated and its environmental impact remains limited, in part due to the challenges of mapping coffee using satellite remote sensing. We recognize the urgency to capitalize on recent technological advances to improve remote sensing methods and generate more accurate, reliable, and scalable approaches to coffee mapping. In this study, we provide a systematic review of satellite-based approaches to mapping coffee extent, which produced 43 articles in the peer-reviewed and gray literature. We outline key considerations for employing effective approaches, focused on the need to balance data affordability and quality, classification complexity and accuracy, and generalizability and site-specificity. We discuss research opportunities for improved approaches by leveraging the recent expansion of diverse satellite sensors and constellations, optical/Synthetic Aperture Radar data fusion approaches, and advances in cloud computing and deep learning algorithms. We highlight the need for differentiating between production systems and the need for research in important coffee-growing geographies. By reviewing the range of techniques successfully used to map coffee extent, we provide technical recommendations and future directions to enable accurate and scalable coffee maps.

Disentangling the numbers behind agriculture-driven tropical deforestation

“…For these crops, the data are limited or of poor quality (Table 1) and both pan-tropical approaches rely heavily on agricultural statistics. Statistical records are unreliable for cocoa and coffee cultivation (86), with further uncertainties as these crops can be shade-grown, in which case their expansion into natural forest can be difficult to detect using remote sensing, and they are also often grown together with other crops in agroforestry systems (87)(88)(89). Records for staple crops are frequently based on estimates and may underestimate harvested areas in subsistence or smallholder contexts due to minimum harvested area criteria in records (90).…”

Transformation of coffee-growing landscapes across Latin America. A review

“…In the many coffee-growing landscapes where data is missing, incomplete, or out-of-date, national-level surveys or censuses by governments, national coffee institutes, or agricultural extension agencies could be adjusted to provide the required data on household, farm, and landscape characteristics and dynamics. Recent advances in remote sensing and machine learning could also make it easier in the future to track coffee production; identify different land uses within coffee landscapes; monitor changes in landscape structure, composition, and function; and potentially also characterize shade levels within coffee fields (Hunt et al 2020 ). Long-term research on coffee landscape structure, patterns, and socioecological processes could be centered on a network of coffee “sentinel landscapes” representing a minimum, but sufficient, set of socioecological contexts in which a broad range of biophysical, social, and economic characteristics and processes are monitored using a consistent set of methods (Dewi et al 2017 ).…”

Mapping the Distribution of Coffee Plantations from Multi-Resolution, Multi-Temporal, and Multi-Sensor Data Using a Random Forest Algorithm

“…Souza et al [10] mapped coffee plantations by integrating multi-temporal variables extracted from Landsat (30 m) and Rapid-Eye imagery (5 m), achieving significant accuracy compared to when only using Rapid-Eye. Sentinel 2 is a new generation of medium-resolution imagery launched in 2015 and has significant potential for mapping coffee plantations [34]. It has 13 bands that consist of the spatial resolution of 10 m (for the visible and near-infrared light bands) and 20 m (for the near-infrared and shortwave infrared bands) from blue wavelengths (0.458-0.523 µm) to SWIR (2.100-2.280 µm) which are sufficient for calculating a range of indices to successfully map coffee plantations.…”