The Nexus Land-Use model version 1.0, an approach articulating biophysical potentials and economic dynamics to model competition for land-use

Abstract

Abstract.Interactions between food demand, biomass energy and forest preservation are driving both food prices and land-use changes, regionally and globally. This study presents a new model called Nexus Land-Use version 1.0 which describes these interactions through a generic representation of agricultural intensification mechanisms within agricultural lands. The Nexus Land-Use model equations combine biophysics and economics into a single coherent framework to calculate crop yields, food prices, and resulting pasture and cropland areas within 12 regions inter-connected with each other by international trade. The representation of cropland and livestock production systems in each region relies on three components: (i) a biomass production function derived from the crop yield response function to inputs such as industrial fertilisers; (ii) a detailed representation of the livestock production system subdivided into an intensive and an extensive component, and (iii) a spatially explicit distribution of potential (maximal) crop yields prescribed from the Lund-Postdam-Jena global vegetation model for managed Land (LPJmL). The economic principles governing decisions about land-use and intensification are adapted from the Ricardian rent theory, assuming cost minimisation for farmers. In contrast to the other land-use models linking economy and biophysics, crops are aggregated as a representative product in calories and intensification for the representative crop is a non-linear function of chemical inputs. The model equations and parameter values are first described in details. Then, idealised scenarios exploring the impact of forest preservation policies or rising energy price on agricultural intensification are described, and their impacts on pasture and cropland areas are investigated.

Evaluating the impact of rising fertilizer prices on crop yields

“…This model provides a bioeconomic modeling framework, which ensures at the global-level consistency between economic behaviors and spatial biophysical constraints in the manner of MAgPIE (Lotze-Campen et al, 2008) or GLOBIOM (Havlík et al, 2011). NLU simulates changes in the agricultural sector (food price, land rent, profit, crop yield, and the share of cropland areas in total agricultural lands) with a nonlinear response of yield to fertilizer prices, as well as an explicit representation of livestock systems and international trade (see model equations in supplementary material and Souty et al, 2012). Souty et al (2013) examine how well the NLU model can reproduce annual observation-based estimates of cropland versus pasture areas from 1961 to 2006.…”

“…In this model, a representative farmer can substitute cropland for fertilizer input and extensive livestock production for intensive one, with the objective of minimizing its production cost under a production and a land constraint (see Data Appendix for the complete equations and Souty et al (2012) for a discussion on the limits of the representative agent approach). Both types of substitution are linked, as the intensification in crop production (i.e., substituting cropland for fertilizer) (i) makes the feed more profitable, but on the other hand (ii) mitigates the pressure on land and reduces the scarcity rent.…”

“…5). Three categories of pastures are distinguished: intensive, extensive, and residual pastures (for more details, see Souty et al, 2012). Two categories of crops are distinguished in NLU: "dynamic" crops, corresponding to most cereals, oilseeds, sugar beet, and cassava, and "other" crops with sugar cane, palm oil, vegetables and fruits, some fodder crops, and remaining crops.…”

“…For illustrative purposes, we also consider scenarios assuming a stability of the fertilizer price or a continuation of past trends to 2050. Once developed, these scenarios are tested with the global model Nexus Land-Use (NLU; Souty et al, 2012). To isolate the impact of the land-fertilizer substitution, the model is run in two variants.…”

Spatial modeling of agricultural land use change at global scale

“…To provide this information, spatial land use allocation approaches are employed to downscale aggregate land demands for large world regions to individual grid cells. Examples of such global scale land use allocation approaches can be found in the Global Forest Model (Rokityanskiy et al, 2007), IMAGE (Bouwman et al, 2006), MagPie (Lotze-Campen et al, 2010), KLUM (Ronneberger et al, 2005(Ronneberger et al, , 2009, MIT-IGSM (Reilly et al, 2012;Wang, 2008), GLOBIO3 (Alkemade et al, 2009), GLOBIOM (Havlik et al, 2011), Nexus land use model (Souty et al, 2012(Souty et al, , 2013, and the Global Land use Model (GLM) (Hurtt et al, 2011).…”