The Mediterranean region is characterized by a continuous decline in water resources, both in terms of quantity and quality, combined with an increase in demand, among which 70-80% relates to agricultural activities. Anticipated climate changes will probably impact resources beyond the adaptation and mitigation capacities of many countries. The fragile secular balance between resources and uses in this difficult water context is also threatened by the rapid upheavals in agriculture in the semi-arid regions of the South side. Despite ambitious public policies encouraging the modernization of practices, large waste of water still occurs because of permissive regulation, lack of knowledge and contradictory vested interests. An action on agricultural water demand thus represents the main leverage of water saving in the region.

Societal and Environmental issues

For the past five decades, the water resource management policy has been based on increasing the water availability by a greater mobilization of resources but major obstacles and contradictions appear now, in both environmental and social domains, with a neat link between them (Gana & El Amrani, 2009; Kuper, 2011; Ayeb, 2011). Mediterranean countries cannot go much further along this direction and are already facing disparities between exploitable water resources and the ever-increasing water demand. Such a deficit cannot be solved through water conservation or non-conventional water resources (e.g. sea‐water desalination and wastewater reuse) alone.

The substitution of traditional (cereals) for more profitable crops (tree crops) modifies the seasonal patterns of water use and add rigidity to the integrated management of the systems. Reduction in surface water supplies may be combined to quality stress on reservoirs water impulse by an increasing use of agriculture intrants. The uncontrolled multiplication of pumping boreholes causes groundwater depletions of one to two meters per year at some locations. In this context, the sustainability of irrigated and rainfed crops in the context of climate change is not ensured. These major upheavals lead to social tensions and may exclude small-scale farmers where the use of surrogate ponds or deeper wells is difficult to conceive, exacerbating competition between sectors and making downstream needs incompatible with ancestral water rights upstream.

The public authorities are massively fostering new techniques which are supposed to spare water, like drip irrigation, although their efficiency regarding this objective is not clear. In addition, the public authorities are well aware of the risks to maintain this critical situation but do not have the scientific elements to assess objectively the current situation, to apprehend the trajectories of resources evolution in the future and to propose new modes of sustainable management.

Scientific issues and state of the art

Addressing these environmental and societal issues requires the implementation of systemic approaches of the agro-systems eco-hydrological functioning in connection with physical reservoirs (storage and renewal capacities) and actors (operators and managers). This involves the development of integrated modeling-observing tools adapted to the complex characteristics of these regions in order to quantify the impact of agricultural water use on the different reservoirs, to characterize the heterogeneities of landscape mosaics (crop systems, climate, soil, infrastructures …), to be able to translate this functioning into production indicators (agro-system services). It requires representing the strategies of the farmers according to the environmental and socio-economic context and the regional governance policies. Four specific transversal scientific issues were identified to address this objective.

Hydrological processes issues

* Each winter, large amounts of water are stored as snow in the highest mountainous areas of the Mediterranean. Due to already observed rise in temperature, the altitude corresponding to the threshold between snowfall and rainfall snowline is rising (Simonneaux et al., 2008), with less water being stored as snow. Melt rates are also increasing and the snow is melting earlier than in the past in lower elevation areas (Marchane et al., 2015). These evolutions could threaten irrigation systems in particular tree crops in summer.

* The impacts on productivity, water consumption, and water use efficiency of the conversion to “cash crops” and of the ongoing intensification are poorly known. Understanding the hydrological functioning of tree cover requires the implementation of modeling tools well suited to the high heterogeneity of the tree cover (trees spacing, several strata, scattered covers…).

* Global change is also marked by a strong evolution of agricultural practices in terms of artificial intrants. A few years ago, small holders, in mediterranean rain-fed systems, were supposed to scarcely use pesticides substances (Schreinemachers & Tipraqsa, 2012). Situation has strongly evolved as herbicides are more and more in use in these contexts (Haggblade et al, 2017). Herbicides and metabolites (particularly glyphosate and AMPA) are known in EU to contaminate surface water (e.g. Meffe & Bustamante, 2014). By contrast, no public data is available in mediterranean southern countries and management cannot cope with this new constrain.

Upstream/downstream issues

* A better understanding of the processes linking upstream and downstream is essential to ensure relevant water planning as water use and supply upstream can largely affect water availability and quality downstream. Hydro-agricultural developments such as dams or hill reservoirs built upstream to trap sediments and to secure access to water resources for local population may have conflicting effects by reducing run-off thus impacting water supply to dams located downstream (Riaux et al., 2014). Likewise, a changes of cropping or reforestation efforts upstream may also affects drastically run-off and infiltration (Mekki et al., 2006).

* Several key processes contributing to groundwater recharge also occurs at the upstream/downstream interface since recharge of basin aquifers through direct infiltration of rainfall is generally limited as observed in Tensift region of Morocco (Boukhari et al, 2015). Anticipated change in the snow/rain partition and uncontrolled water uptake upstream may affect directly the timing and amount of surface water availability downstream and also the groundwater recharge.

Observation issues

* Long term observations are essential both for resource management and for agro-hydrological model development and calibration. Current water resources observation systems provide a good picture of runoff at the catchment outflow locations while upstream processes such as snowpack dynamics, subsurface flows and agricultural usage are poorly monitored. Some integrated, long-term observatories of water resources exist, in particular in the Tensift and the Merguellil regions, but these successful initiatives need to be multiplied and unified around an integrated regional observation system.

* On the ungauged basin of the South Mediterranean, non-conventional observations such as remote sensing and new innovative experimental designs based on low-cost in situ sensors should be promoted. Remote sensing undergoes a true revolution with the recent launch of the Copernicus Sentinel constellation that provide a large quantity of observations to enable monitoring of key variables for water resources management such as snow covered areas, soil moisture, land cover, crop development and stress, water reservoirs surfaces, etc.

Social, political and economic issues

* “The control and regulation of groundwater abstraction is perhaps the most vexing issue of water management worldwide, with very few encouraging ‘success stories’” (Molle et Closas, 2016). Driving forces are numerous and cannot easily be oriented to fulfill the objective of a sustainable water management. Indeed, water use is highly linked with external factors, such as water availability or climate but also to agricultural, demographic or energy policy.

* Governance, which can be seen as the distribution of decision-making power that governs and shapes the circulation and sharing of water, is a key area of concern and investigation. The current status and future trajectories of water resources must be apprehended through an analysis of the structures of power (Zeitoun et al., 2012; Molle, 2012), in terms of economic, social, political and discursive power.

* The analysis of governance has therefore to include state level but also to observe the water user level, to better understand individual water extraction behaviours. These behaviours are partly explained by economic motives (to maximize individual income) but also influenced by non-economic motives, including adherence to social norms, ethical commitment, altruism, reciprocity and inequity aversion (Bowles, 2008). These dimensions have to be taken into account to explore how farmers perceive the functioning of aquifers and the use of groundwater (Bekkar et al., 2009). These behaviours are deeply linked to the local situation, which explains the need to analyze them in each specific case.