Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Projets de développement → Projets de recherche pour le Développement → 2009 → MODELING RANCHER AND PUBLIC AGENCY DECISION-MAKING TO COORDINATE EFFORTS TO BATTLE INVASIVE WEEDS AND WILDFIRE ON NEVADA’S RANGELANDS

United States Department of Agriculture (USDA) 2009

MODELING RANCHER AND PUBLIC AGENCY DECISION-MAKING TO COORDINATE EFFORTS TO BATTLE INVASIVE WEEDS AND WILDFIRE ON NEVADA’S RANGELANDS

Invasive Seeds Wildfire

United States Department of Agriculture (USDA) Research, Education & Economics Information System (REEIS)

Titre : MODELING RANCHER AND PUBLIC AGENCY DECISION-MAKING

Identification : NEV05176

Pays : Etats Unis

Durée : Jul 1, 2009 à Jun 30, 2012

Domaine : Management and Control of Forest and Range Fires ; Weeds Affecting Plants ; Economics of Agricultural Production and Farm Management ; Desert and semidesert shrub land and shinnery

Partenaire : UNIVERSITY OF NEVADA RENO

Objectif
We will develop models of the decision-making processes for rangeland management, both at the private rancher and at the public land manager levels, to battle the problem of invasive weeds and wildfire in the Great Basin. Decision making models are constructed as stochastic dynamic programming (SDP), where the levels of livestock grazing, applications of rangeland vegetation treatments, and frequency of stochastic wildfire interact to affect the direction and the magnitude of the changes in the ecosystem. Irreversibility of an ecosystem flip will be incorporated in the model. The models will be parameterized with information collected at the field level and using the best available knowledge of ecosystem processes. The models will be used to investigate a series of policy questions including : 1. Effectiveness of alternative rangeland improvement strategies, including doing nothing 2. Cost of delaying implementation of rangeland treatments 3. Benefit of reduced uncertainty about ecological parameters.

Descriptif
We model the rangeland management problem in a stochastic dynamic optimization framework. Conceptual models will be numerically implemented and solved with a stochastic dynamic programming (SDP) technique. We develop specifications separately for a private rancher and a social planner in order to accommodate different assumptions about what costs and benefits are explicitly considered by each, and the level of knowledge about fire cycle and vegetation dynamics among the two groups. While public decision makers may have a better access to the latest scientific understandings of the ecosystem, ranchers may be better informed about the specific conditions of their own allotments, and costs to ranching operations. The most general specification of the model is a social planner’s problem, which is most comprehensive in social values taken into account. The rancher’s problem will be derived as modification/simplification of the social planner’s problem. The rancher’ objective is to maximize the net present value of private lands or of public land allotments. Lands used for ranching to generate rancher incomes may also provide other values to society. The social planner is interested in finding optimal grazing level and land treatment schedule, taking into account all social values and both economic and biological parameters governing the system. "Excessive" grazing is modeled to cause increased prevalence of invasive annual grasses. The shifts are reversible if the land has not crossed an ecological threshold, where the system loses resilience or its ability to regenerate native vegetation. Reduced or suspended grazing and various land improvement investments enable recovery of land quality. However, once the land has crossed the threshold, native perennials can no longer out-compete the invasive annuals. In this case, recovery becomes impossible or prohibitively costly, and thus the shift is now irreversible. Changes in the ecological condition of the rangeland are also affected by stochastic wildfires. Before crossing the threshold, fire improves land condition by removing invasive annuals and making more rooms for native perennials to reestablish themselves. After the threshold has been crossed, however, a fire is sufficiently intense to burn the roots of native perennials, allowing invasive annuals expand even further. Stochastic wildfires also introduce into the model uncertainty about current-period ranch income (due to mandatory grazing suspension for range recovery) and about ecological impacts of grazing and land improvement investments. In addition, the costs of fire suppression increase as highly flammable invasive annuals become more prevalent.

Présentation : USDA

Page publiée le 21 décembre 2015, mise à jour le 7 novembre 2017