Availability and significance of data for the development of a geodatabase on the water physicochemical quality: the example of the Moroccan groundwaters

Water resources are relatively limited in Morocco, as well as in other South-Mediterranean countries, mainly because of the aridity of their climate, marked by recurrent droughts. On the other hand, the water needs are continuously growing and the main solutions adopted during the last 60 years to cover these needs were damming and groundwater pumping, and it is recently that the public departments resorted to the use of treated wastewaters. This water scarcity was progressively amplified by pollutions that reduce the availability of good quality waters. Indeed, the population increase led to an exponential and concomitant expansion of urban, industrial and agricultural activities, which pollute both surface and underground waters, making their use as a big challenge for public sectors. Therefore, both surface and underground waters need, today more than ever, to be sustainably managed. In this context, new diagnostics and syntheses are necessary, but the best way to tackle this challenge should be long-term surveys of the water quality. That means a crucial need to performant information system on this topic, covering the whole country and designed in the context of a decision-making mechanism, both strategic and operational. This work aims to verify the feasibility of such information system, using the example of the groundwater physicochemical parameters. A database has been designed and filled with a high proportion of available physicochemical data, proving that this database is technically feasible. The first synthesis attempts of these data with classical models revealed that they are relatively well adapted to the database structure. However, different obstacles prevent getting significant results. They are mainly due to imprecision and heterogeneity of the data format and the methods used. These obstacles were partly lifted by transforming and completing information, but many existing data will remain unexploited or with weak significance, particularly in spatiotemporal analyses. In addition, this study aims to suggest some standard methods for developing a monitoring that could be qualified as national.