"This action created a geodatabase of field size patterns and dynamics for the entire study region (target years: 1990, 1995, 2000, 2005, 2010, 2014), combined it with parcel data for Brazil around 2015 by intersecting fields with parcel boundaries, and estimated the correlation and the functional form between both size estimates. Further, it mapped clusters of large/small parcel and field sizes, as well as spatial outliers (i.e. large parcels near small fields; and vice versa) using bivariate Local Indicators of Spatial Association. Parcel/field size data was merged with a set of location factors (soil quality, topography, climate, population, accessibility, landscape diversity) on municipality level. The most important location factors explaining parcel/field size patterns were identified using Boosted Regression Trees, which was population density for parcel size, and slope for field size. The action confirmed the hypothesis that the parcel-field size relationship is strongest for small parcels and weakens as parcel size increases. It further confirmed that smaller parcels and fields are mainly located in marginal areas with inferior agro-climatic conditions, accessibility, and economic activity.
This action generated a geodatabase of 10 environmental impact indicators (CO2, NH4, and N2O emissions from agriculture, vegetation loss, NPP loss due to land-use change, biodiversity loss, water use for irrigation, N&P excess, soil erosion, and agrochemical application), malnutrition (wasting and overweight), and income poverty. Using a threshold of two standard deviations for environmental indicators, prevalence values of 20% for malnutrition indicators, and the International Poverty Line threshold of 5.5 US$ (PPP 2011) per capita per day for income poverty, it mapped hotspots of food system challenges. It further assessed the spatial co-occurrence of such hotspots and farm size, as well as place-based impacts of hotspots on agricultural area [ha], number of people [#], and nutritional value of agricultural production [Pcal]. About 71% of global agricultural areas were affected by at least one hotspot, about 5% by hotspots of all three dimensions, which were located mainly in East Africa and South-East Asia. Regions where smallholders (< 2ha) dominate carried the burden of multiple food system challenges, dominantly by a co-occurrence of environmental impact and malnutrition hotspots. With increasing farm size, environmental impacts became more important and were the dominant food system challenges in regions characterised by large to very large farms. In these regions, environmental impacts hotspots were related to calories produced form livestock."