The SALT Program

Salt Program has as aim to predict the nature and the intensity of the modifications of the functioning and the dynamics of the savannas of West Africa under the influence of the climatic and anthropic changes, present and future.

SALT is held within the framework of the International Geosphere Biosphere Program (IGBP) and is one of the pilot projects of the Core-Project “Global Changes and Terrestrial Ecosystems ”. It also fits in the theme of biodiversity, which is carried out on the animal, and vegetation functional groups like on the interactions between structure, functioning and dynamics. SALT has also vocation to require tools for decision-making aid by identifying the key processes of soil management with weak inputs and by analyzing the ecological consequences of various scenarios of land use. SALT must finally promote the interdisciplinarity as well as the co-operation between African and French teams and by supporting the scientific exchanges and the training of the students and young researchers.

Savannas of West Africa (4 Mkm² approximately) form a continuous block of vegetation mosaics distributed according to a gradient of aridity growing from South to North, from the Gulf of Guinea to the Saharan borders, and a gradient of continentality from West to East. To take into account the variability of the processes and the stresses which savannas undergo, the studies are undertaken on the long term and are based on a network of research sites which cover the two gradients and various modes of land in Senegal, Mali, Côte-d'Ivoire, Burkina-Faso and Niger. Currently SALT counts main sites (approximately 40 X 40 km, see figure 1) and secondary sites identified in order to take into account regional specificity and to allow the extrapolation of the results acquired on the major sites.

SALT is based on an approach associating the mechanisms of mass and energy fluxes with those of the dynamics of the species and the vegetation structure. A detailed attention is thus paid to the interactions between processes. That results in the analysis of the functional processes, which form the mechanistic base of the savanna ecosystem. Then, one integrates these processes in the system between soil, vegetation and atmosphere at the local level, system whose space expression is related to the distribution of its animal and vegetation components and with that of his intrinsic disturbances: fires, termite nests, pastures, etc. One thus underlined the existence of processes which are expressed in a particular way in savanna, as for example recycling of nitrogen in function, on the one hand, of the cover structure and the plant architecture which make it up, and, on the other hand, of the space-time distribution of the sites of the fauna activity on the soil. Another crucial phenomenon is that of the degradation of the soils related to the variations of the water, the carbon and the nitrogen cycles of the soil.

The objective is thus to determine the biophysics relationships of the energy and mass transfers at the interface between the soil, the vegetation and the atmosphere on a site scale and to understand how those determine the evolution of the ecosystems. In this context, the ecological processes studied can be grouped in three major topics :

- the primary production: the nutriment and organic matter cycles (carbon inventories and fluxes)

- the surface characteristics and their link with the hydric state of the soil and the phenomena of wind or rain erosion

- the structure and the dynamics of the vegetation, from the scale of the patch to that of the landscape.

These great processes are planned to observe, understand and model the responses of the ecosystem and the landscape to the disturbances (climate, fires of pasture, cultivation) on a local scale and according to gradients' of climate change and land use. This work should allow the development of generic models of savanna functioning and dynamics at various scales of space and time.

As SALT aims at understanding these mechanisms on increasing space scales from the patch to the region, that involves a specific activity of the program in the field of the regionalization of the processes.

The space extension of the results obtained at the site is done initially on a semi-regional level. The study of the surfaces occupied by the various components characteristic of the savanna (relations grass - tree/soil) makes it possible to calculate an integrated value of various variables like the primary production or the organo - mineral status of the soil.

Many parameters and variables studied can be the subject of a direct spatialization, either because their quality of sampling is good (use of maps), or because they can be perceived using satellite sensors. The radiometric measurements acquired by the space sensors make it possible to find temporal profiles of surface reflectances or temperatures. At the local scale and after ground validation, these sizes inform about the structure, the hydric state and the production of the vegetable cover. Moreover, the satellite data allow the space integration of these variables on the scale of the area and the follow-up of those in time (seasonal and inter - annual variations).

To integrate these processes at the regional scale goes through the establishment of relations between the satellite data and ground measurements with regard to :

- the identification and the hierarchisation of the modes of land cover on the site scale but also on the scale of the whole of the region

- the hydric status of the ecosystems, with the construction of an existing model of simulation of the surface-atmosphere exchanges based on the functional relations between the structure and the phenology on the one hand, and the water and energy balance, on the other hand

- the biomass, phenology and primary production, with a production model associated with a model based on the energy and water balance of the soil where fine spatialization is introduced by coupling of satellite data with high space resolution (SPOT-HRV) and high timescale (NOAA/AVHRR)

- the spatio - temporal variation of burned surfaces and biomasses, with the NOAA-AVHRR data which provide the space-time distribution of fires on a regional scale, and the values of NDVI which, associated with the ground data, make it possible to determine the biomass present at the moment of fire and to consider the components released in the atmosphere

- the atmospheric turbidity with the installation of a series of photometers on main SALT sites, which makes it possible to follow the optical thickness (aerosols), and the water content of the atmosphere, and is used thus as ground control of the measurements of fire emissions and dust transport.

The spatio - temporal expression of these processes is thus analyzed using satellite data resulting from various sensors which allow the study and the follow-up of the surface parameters (vegetation, soil, fire). The results then provide estimates of the parameters and key variables on the local and regional scales. The satellite data are thus used at the same time to regionalize the site results but also to be connected to ecological models like CENTURY, ETEMA, TIGREE and MUSE whose outputs specify the quantities typical of the structure, of the functioning and the dynamics of savannas of West Africa (carbonaceous, radiative, energy and hydric states, surface roughness, etc.).

Work on the interactions between soil, vegetation and atmosphere and their regionalization was and is carried out within the framework of various international collaborations:

- energy, radiative and hydric balance: collaboration with the HAPEX - Sahel experiment (IGBP/BAHC and WCRP/GEWEX)

- biogenic emissions induced by fires : collaboration with DECAFE experiment (IGBP/IGAC)

- transfers of particles with the atmosphre : collaboration with the PHOTONS-AERONET and Wind Erosion experiments.

Many teams take part in the SALT program, which gives him its full interdisciplinary dimension and its control of the field experiments