Multi-scale land-use disaggregation modelling: Concept and application to EU countries

Abstract

a b s t r a c tChanges of carbon stocks in agricultural soils, emissions of greenhouse gases from agriculture, and the delivery of ecosystem services of agricultural landscapes depend on combinations of land-use, livestock density, farming practices, climate and soil types. Many environmental processes are highly non-linear. If the analysis of the environmental impact is based on data at a relatively coarse-scale (e.g. farm, country, or large administrative regions), conclusions can be misleading. For an accurate assessment of agrienvironmental indicators, data of agricultural activities and their dynamics are needed at high spatial resolution. In this paper, we develop and validate a spatial model for predicting the agricultural land-use areas within the homogenous spatial units (HSUs). For the EU-28 countries, we distinguish about 1.5 Â 10 5 HSUs and we consider 30 possible land-uses to match with the classification used in the Common Agricultural Policy Regionalized Impact (CAPRI) model. The comparison of model predictions with independent observations and with a simple rule-based approach at HSU level demonstrates that the predictions are generally accurate in more than 75 % of HSUs. The frequent crops or land-use are better predicted. For non-frequent crops and/or crops requiring specific cultivation conditions, the model needs further fine-tuning.

Unpacking ecosystem service bundles: Towards predictive mapping of synergies and trade-offs between ecosystem services

“…Recent developments in downscaling or disaggregating datasets hold promise for higher resolution analyses with available datasets (e.g. Keil & Jetz 2014;Lamboni et al 2016).…”

EU‐wide Economic and Environmental Impacts of CAP Greening with High Spatial and Farm‐type Detail

“…CAPRI production outputs, such as crop area and livestock densities at a geographical resolution of NUTS-2 region are distributed on the spatial units using a combination of statistical and geographical information system (GIS) techniques (Kempen et al, 2005;Heckelei et al, 2008;Lamboni et al, 2015). As a last step, the nitrogen (N) budget is calculated for each crop-spatial unit combination or simulation entity (Leip et al, 2011), making sure that crop needs plus over-fertilisation equals the total input by mineral and organic fertilisers, biological N fixation, atmospheric deposition or any other source of N. The database obtained is used to calculate a large array of agri-environmental indicators.…”

Predicting soil properties in the Canadian boreal forest with limited data: Comparison of spatial and non-spatial statistical approaches

“…Environmental covariates are obtained from various sources, including remote sensing products and digital elevation models (McBratney et al, 2003). When detailed expert-based soil maps are available, techniques of spatial disaggregation of polygon information are often used (Bui and Moran, 2001;Lamboni et al, 2016). However, over large regions, and more typically in forested regions, expert-based soil maps are often unavailable and the scorpan model approach (see McBratney et al, 2003), which matches environmental covariates with soil point data (pedons), is commonly used.…”