Additional research products (ARP3): FIGSIS (fire model). A model to simulate the effect of fire in rangelands and desertification

Authors:

In Mediterranean ecosystems it has been well documented that the productivity of many lands which are used for grazing of animals has deteriorated as a result of a combination of overgrazing and fires.

Both of these are related to human activities but they are also aggravated by extreme climatic phenomena such as drought. In Greece, more than 10% of wildland fires of known causes are started by shepherds in grazing lands in their effort to increase immediate forage production. This percentage is assumed to be much higher if fires of unknown origin are also taken into consideration (Xanthopoulos 1996). The intensity of the problem differs considerably between regions and the ecological results differ correspondingly. In certain cases, the cycle of grazing-fire has already led to complete degradation, while in others the position is better. It is clear that these differences are primarily due t to the choices made by shepherds.

This tool concerns the development of a simulation model of the fire-grazing cycle in rangelands for Greece. Its aim is to examine the relative importance of factors which influence the cycle, to quantify their impacts and to determine thresholds beyond which overexploitation leads to site deterioration.

The model was developed on a PC with the help of a programmable environment named, NetLogo, which is freely available on the Internet (Wilensky 1999). The user initially inserts data such as: rangeland area, mean rainfall, standard deviation of rainfall, vegetation type, number of animals, grazing months, start and end simulation years, year after last fire, EU subsidy, milk, meat and forage prices for the producer and acceptable income for the shepherd.

The model has several sub-modules that represent:

(i) the production of forage in an area of interest based on the type and condition of rangeland, rainfall, and the time after last fire;
(ii) the intake of browsed biomass through grazing by livestock
(iii) the effect of browsing on the condition of the rangeland and
(iv) the financial outcome for the shepherd including the influence of exogenous factors (meat price, milk price, feeding price and EU subsidies). The financial result, compared with the yearly income which is considered as "fair" is considered as a potential motivation for the shepherd to use fire as a tool for improving immediate forage production.

Yearly forage production is calculated, for grasslands using the equation of Papanastasis (1982), for rainfall between 268-1079 mm and has the form:

Υ=(-32.767 + 0.4351Χ)*10

where Y: annual forage production in kg Ha-1 and X: yearly rainfall in mm.

For sclerophyllous vegetation and phrygana, and for areas with mean rainfall from 50-900 mm. forage production is calculated by the Le Houerou and Hoste (1977) equation for the Mediterranean:

Υ = (-10.372 + 0.217 Χ)*10

where Y: annual forage production in kg ha-1and X: yearly rainfall in mm.

When fire is introduced in the simulation, the equations in the table below (table 8) are used for calculation of total biomass production as a function of post fire age of vegetation, for the first few post-fire years. The same authors provide percentages of this biomass, as a function of post-fire year, that constitute forage.

where X: years after fire and Y is the above ground biomass (kg ha-1)

Monthly requirements of livestock are set by default at 60 kg of dry matter but this value can be controlled by the user. After the insertion of the data the model calculates a set of intermediate variables that describe both the forage production and consumption status (including the need for additional forage obtained from the market), the condition of the site and the financial result of the enterprise (income, including European Union subsidies, versus expenses).

An empirical equation, based on interpretation of interviews with shepherds, was developed for estimating the motivation of a shepherd to use fire, recognizing that by doing so he shepherd takes a risk because use of fire is illegal. The idea is that the motivation (NEED to start a fire) increases when the economic result of the enterprise is less than an expected "fair" income (YSALARY). This income is defined by the user. The result of the equation is a NEED value, corresponding to a probability for starting a fire, in the range from 0-100. The equation has the following form:

NEED = 100 - (100 / 1 + e(-(FR - YSALARY) / RC) )

where: RC is the range of income in which the motivation of a shepherd to use fire changes from (nearly) 0 to 100 and FR is the Financial Result of the enterprise calculated as profits minus expenses.

Implementation

The use of the simulation model of fire-grazing circle in rangelands of Greece, allows the examination of various combinations of grazing intensity, exogenous factors and use of fire in a long-time basis, including a stochastic component (rainfall) with the aim to determine the values of stocking rate and grazing capacity which guarantee sustainable development and the thresholds, which can lead to land degradation and desertification. Future improvements in sub-modules will increase the functionality and representation in order for the model to constitute a tool for better rangeland management of Greece in an effort to control desertification trends.