Identifying Agricultural Frontiers for Modeling Global Cropland Expansion

The increasing expansion of cropland is major driver of global carbon emissions and biodiversity loss. However, predicting plausible future global distributions of croplands remains challenging. Here, we show that, in general, existing global data aligned with classical economic theories of expansion explain the current (1992) global extent of cropland reasonably well, but not recent expansion (1992–2015). Deviations from models of cropland extent in 1992 (“frontierness”) can be used to improve global models of recent expansion, most likely as these deviations are a proxy for cropland expansion under frontier conditions where classical economic theories of expansion are less applicable. Frontierness is insensitive to the land cover dataset used and is particularly effective in improving models that include mosaic land cover classes and the largely smallholder-driven frontier expansion occurring in such areas. Our findings have important implications as the frontierness approach offers a straightforward way to improve global land use change models.Predicting where cropland is likely to expand globally has important implications for climate change and biodiversity. However, doing so is challenging because of a lack of data on key drivers of current expansion in so-called frontier areas (e.g., governance). Here, we show that using deviations from a model of cropland extent in 1992 built on all available data improves models of recent (1992–2015) cropland expansion over and above using the best available existing data.Cropland area is increasing globally to satisfy the growing population and consumption rates. Cropland expansion often comes at the expense of forests, which are critical for conserving biodiversity and mitigating against climate change. Therefore, it is essential to know where cropland expansion is likely to occur in the future in order to design policies that prevent expansion in areas that are most likely to conflict with forest conservation. However, predicting where expansion is most likely is difficult as few data are available on key predictors related to governance. Here, we devise a novel, two-stage method for predicting expansion of cropland. First, available data are used to model where cropland existed in 1992. We then use maps that show where the first model fails to explain cropland in 1992 to help predict expansion of cropland between 1992 and 2015. We show that this approach is an improvement over simply using existing data to predict recent expansion of cropland.