Context and objectives
To improve livestock farming systems sustainability, from an economical and environmental point of view, tools necessary to manage nutrients stocks must be developed. They must be able to estimate forage availability, in quantity and quality to adjust the complementation and, in this way, to reduce feeding costs, that represent the greatest part of the proportional cost in livestock farming systems, and foodstuff wasting. Such maximisation of self-producec fodder crops valorisation in livestock farming systems will aslo increase product traceability and differentiation and, in this way, the sustainability of these sytems from a social point of view.
The use of dynamic model seems to be a promising tool to reach these objectives. Such models are based on the biomass formation through the photosynthetic activity of the plant under different environmental conditions (temperature, soil water content, photosynthetically active radiation, ...). Limitations of existing models are that their accuracy are correlated to their complexity. Some of their parameters being difficult to obtain in a continuous way on a great number of sites. The aim of this study is to validate an agro-meteorological model, using some easily available parameters, for Grass growth in quantity and quality, and to use remote sening data to control and adjust output at regular intervals, on the one hand, and to integrate spatial heterogeneity, on the other hand. The accessibility of this tool will be obtained through its integration on an internet site.
Project outcome
Expected scientific results
This study has allowed the development of a Grass Growth Model based on pedo-climatic data. It takes into account soil hydrological properties, thermic stress, and its impact on the delay before regrowth starting and on the speed of this regrowth, and nitrogen supply. A quality module had also been developed. Nevertheless, performances of this model on 2000 data were far from the one obtained following its cross validation on initial data set. So, model parameters must be adjust, through additional field observations, to integrate the diversity of the situation observed in the study area.
Remote sensing observations underlined the necessity to distinguish between meadow and grazed grasslands. Indeed, the mixture of grazed and cut grasslands, in a 1*1km pixel, do not allow the direct translation of Vegetation Index differences in biomass production. It was then, currently, impossible to integrate remote sensing observations in the pedo-climatic model. To reach this target the proportion of grazed and cut grasslands must be defined in each pixel.
Definition that will also allow a better un-mixing of the proportion of the Vegetation Index linked to the different vegetation components. Nevertheless, remote sensing observations could be used in relative ways. Firstly, to define the relative importance of forage stock constitution in the different agricultural areas. To do so, the definition of an adapted Vegetation Index treshold in function of the animal stocking rate had to be done in each agricultural areas. Secondly, to identify grassland exploitation dynamic at the scale of the different agricultural areas, that is to say the periods of grazing, cutting or the periods of climatic stresses.
| Location: |
Region:
|
|
| Related publications: | ||
| Website: | ||