The impact of rivers inputs
The influence of river basins on the coastal zone and even on the open part of enclosed seas as the Mediterranean has been recognized since the early 1970’s as a key driver in sediment, water, nutrients and pollutants budgets when UNEP launched a series of Regional Seas Programs on the Caribbean, Gulf of Guinea and Mediterranean, among others. These first studies pointed out the great heterogeneity of Mediterranean and Black Sea rivers in terms of water runoff, sediment discharge, and associated nitrogen and phosphorus fluxes, mainly due the great climatic variability in this region and the related differences of land use and socio-economic practices in the corresponding drainage basins.
Also at temporal scales, the evolution of the river inputs is marked by great differences. More recent state-of-the-art evaluations (e.g., Ludwig and Meybeck, 2003) indicate that freshwater inputs by rivers might have considerably decreased during the last fifty years as the result of the combined effect of ongoing climate change and enhanced anthropogenic water use. Also the suspended sediment delivery by rivers was strongly reduced by human activities, in particular through the massive dam construction that appeared in the Mediterranean drainage basin through the second half of the last century. Riverine nutrient fluxes, on the other hand, are closely linked to economic development. Nitrate loads for example could have increased since the seventies by a factor of two or three in the large Mediterranean rivers such as the Po and Rhone, mainly in conjunction with increased fertilizer spreads. But nutrient levels in rivers are also sensitive to policy and societal behaviour, and not all nutrient species followed the same trends. A typical example for this is the dissolved inorganic phosphorous loads which decreased drastically during the eighties and nineties because of the banning of phosphate detergents and the improvement of wastewater purification in the drainage basins.
Mediterranean rivers are therefore potential drivers for long term changes in the marine ecosystems. However, up to now, potential forcing of the marine ecosystems through rivers could not been evaluated on the basin scale of the Mediterranean and its different sub-basins. Long-term data sets on the river inputs are scarce, which complicates the development of budgeting approaches and their use in ecological studies.
Regular monitoring of nutrients and other basic water quality parameters in rivers has been established in many bordering countries of the Mediterranean. Even if many of these data are still difficult to access for the scientific community, there is increasing effort on the political level to release these data into the public domain. But scientific knowledge and questions evolved since and a number of key elements are not monitored. Dissolved silica, for example, is generally considered as a natural compound and not analyzed, although this nutrient can be important for the marine ecosystems depending on the riverine nutrient loads. Time series on dissolved silica fluxes in Mediterranean rivers are therefore almost impossible to find. Also more and more organic and inorganic contaminants (trace metals, PCBs, pharmaceuticals, …) were identified as potentially harmful for the marine biota, but mostly ignored in the national monitoring programs as well. In the context of the contaminants in the Mediterranean Sea (see MERMEX project), it is important to better constrain the spatial distribution and the evolution of anthropogenic compounds in the marine ecosystems. The uncertainty on sources contribution and the transfer mode in between the marine reservoirs are important, especially for emergent substances. Chemical contaminants constitute as well pertinent tracers useful for improving comprehension of physical and biogeochemical mechanisms.
Another major problem is that national monitoring programs are based on regular sampling (e.g. monthly) and therefore not suitable to quantify the particulate matter and associated nutrient and contaminant fluxes by rivers. In the Mediterranean climate, short and violent flash-floods are typical for the hydrological regimes and account for by far most of the total riverine particle transport. If these events are not followed by adapted sampling strategies (high frequency sampling through automatic devices), it is not possible to assess the average fluxes. This is particularly true for small rivers, where the ratio of peak discharge on mean annual discharge, is frequently about one order of magnitude greater than for rivers in Non-Mediterranean areas (Estrela et al., 2000). Small rivers are naturally less important in global material budgets, but they are very typical for the Mediterranean drainage basin. Adding their fluxes together is far from being negligible and may even surpass the material transport from the large rivers. Moreover, as the small Mediterranean river basins are highly reactive to local climatic features, they may be more vulnerable in the context of climate change and modifications in the frequency of extreme climatic events (floods, droughts) can have server consequences on the river fluxes.
The Gulf of Lions in the Western Mediterranean Sea is particularly interesting for the study of riverine material fluxes and their potential impacts in the marine system. The sedimentary system forms one of the largest platforms in the Mediterranean which receives great amounts of terrestrial inputs via rivers. As a consequence, also the biological systems are characterized by high productivity rates. Monitoring programs of the water quality parameters in the freshwater and coastal systems were set up in the years 1980-90 (e.g. RNO, REPHY, REMI, RINBIO) and supplied the framework data on the biogeochemical functioning in this particular environment. More recently, the French CNRS launched an interdisciplinary research network (ZA Orme) focusing on the social en economic forcing of the material transfers (http://medias.obs-mip.fr/orme/).
The Rhone is the largest Mediterranean river in terms of its freshwater discharge and its inclusion in the Moose network is mandatory. This river is already equipped with an automatic sampling station close to its mouth (station SORA maintained by IRSN/ Agence de l’Eau RMC/ COM) which allows regular sampling of nutrients and other basic water quality parameters during low and normal water stages, and high frequency sampling of particulates during floods for improved budget calculations. Knowledge about the nutrient fluxes in the Rhone can be reconstructed for about the last 30 years and following the trends in the future defines one of the fundamental forcing functions of the marine ecosystems in the future.