Résumé (en anglais)
The socio-economic central theme of this project is composed of several issues concerning land use: its description, its evolution, its impact on natural resources and the impact of public policies on land use. The evolution of the socio-economic context in which the productive and residential choices are done is conditioned by relatively recent events: climatic change, tensions on the real estate market, agricultural policy reforms, new environmental regulations. We are interested in consequences of such phenomena on the sustainable management of natural resources, the development planning and the territorial management. The data at hand to understand these phenomena is always attached to geographical space and this particular nature has not always been accounted for in econometric approaches of these problems. Our general objective is to propose models for these phenomena which properly take into account the fundamental spatial nature of the data and which use the most recent spatial econometric advances. We are interested in the patterns and changes between the three broad categories of land use: agriculture, urban and other uses. The objective is then to analyze the determinants of this evolution and to try to better understand their economic and environmental effects. Such modeling can then be used to evaluate, among other applications, regional emission levels of greenhouse gases due to land use changes between agriculture and forest, since forests constitute a carbon sink (net absorption), while crops and cattle are positive and major contributors to the global emission level of greenhouse gases. Another application that will be considered is the impact of land-use changes on environmental pressure indicators regarding nitrogen emissions (from organic and mineral sources) and water withdrawal. We also intend to model the production decisions by French farmers, in order to identify the determinants which would not be captured by purely economic or agro-pedoclimatic variables, thus explaining the observed bottlenecks on specialization or adoption of new production modes (MAE or CAB) in some regions. We are also interested in modeling residential water demand as a function of the pricing policy of water utilities in French local communities. A first question will be related to the valuation of (positive) amenities from forestry in terms of improving raw water quality at the local level, and through the reduction in water pretreatment costs. As water price will ultimately be determined by management modes chosen by the municipality but also by the local conditions for water utility operation, the second question the project will address is the impact of the geographic proximity of communities in the decision of management modes of local public services for water distribution and for its price. From the statistical point of view, the common denominator of our models is the family of autoregressive spatial models developed in the spatial econometrics literature, which we use for the spatio-temporal models as well as the discrete choice spatial models. The issues that are common to these models concern spillover effects assessment, predictions evaluation, model selection. Another common concern is also to explore alternative models issued from the spatial statistics literature which is more concerned by modeling continuous fields. These are precisely among the directions recently emphasized by the leading authors in spatial econometrics.
Dates : 01/01/2012 – 30/06/2016
Contact: Christine Thomas-Agnan