For processes that transform or occupy substantial land areas,
the precision of LCA can be improved by accounting for
spatial variation in their input–output relationships. GIS
technology offers the combination of spatial data and
analytical functions to quantify flows from specific locations.
GIS-based inventory modeling of production processes that
transform and/or occupy a substantial amount of land, such as
agricultural production and resource extraction, allows several
refinements in LCA:
1. Land use can be modeled as a nonlinear function of
agricultural or resource output (Fig. 3).
2. Other inputs and outputs related to land use, such as
irrigation water, fertilizer, nitrogen runoff, and soil
carbon, can also be modeled more accurately.
3. Land use of processes can be described in detail and
expressed as elementary input flows of habitat types.
Quantifying impacts on biodiversity from land-hungry
processes, such as fuel crop production, requires spatially
explicit representations of precisely how much land is used and
where the use occurs. In consequential LCA, land use changes
can be modeled through if-then type modeling, scenario
analysis, or more predictive approaches. Predictive
approaches, such as forecasts, would have to be based on
agro-economic modeling that account for biophysical factors,
market conditions (demand for and price of crops), and
landowner behavior. No such forecasts for our study area
exist, and developing this type of model was outside the scope
of this project. To ensure plausibility of land use scenarios,
basic drivers and constraints, such as profitability and land
ownership, should be taken into account, as shown in this
study