The challenge is to implement research that can estimate the consequences of climate changes in terms of impact on terrestrial environments and resources. Emphasis should be placed on regions dependent on natural resources and for which demographic pressure is strong. Simulations obtained from climate model projections (using different Representative Concentration Pathways (RCPs)) predict that the Mediterranean basin and its southern periphery are particularly vulnerable to water resources and environmental impact (IPCC, AR5, 2013). An annual rainfall decrease by 30% is found for the projection period 2070-2099 (IPCC, AR5, 2013) associated with a decrease in water resources by 30 to 50% (Milano, 2012). In addition, several studies using regional atmospheric models indicate an increase in the precipitation inter-annual variability with extreme events and a spatial heterogeneous signature, superimposed on a decrease in the total precipitation amount (Giorgi and Lionello, 2008; Raible et al. 2010). Currently, regional climate projections are highly sensitive to the climate model used. In particular, spatial resolution as well as local climate conditions seem to impact significantly on the simulations (Jacob et al. 2014). The Mediterranean region, at the interface between arid and temperate climates with several mountainous areas, is a complex climate system affected by the interactions between mid-latitude and sub-tropical processes. In this context, Morocco, located at the transition between a temperate climate to the North and a tropical climate to the south constitutes a key area for an impact and sensitivity study to global climate changes. The climate is influenced by the Atlantic Ocean, the Mediterranean Sea and the Sahara, together with a very steep orography in the Atlas region. The precipitation distribution is therefore characterised by great spatial variability, and exhibits a marked seasonality, a strong inter-annual variability (Ouda et al. 2005) and in general a pronounced gradient from north to south and west to east. At a broader scale, Morocco is located on the subtropical subsidence path and between the Acores High and the Saharan Low (Agoussine, 2003). Several studies have also identified strong links with inter-annual precipitation variability and NAO index (Knippertz, 2003) as well as remote climate modes (Esper et al. 2007). Continental climate variability at a local/regional scale, if it is to be integrated in climate predictions, needs to be supported by long-term observation. Meteorological stations in Morocco provide climatic data mainly for the last 40 years with only a few stations located in the mountainous region (Tramblay et al. 2012; 2013; Driouech et al. 2010). This climate database is also supported by the IAEA network providing stations for which isotope tracers have been applied to daily/monthly rain and water vapour samples over 2 to 3 years between 2000 and 2004. Besides the poor coverage of instrumented areas, lacustrine systems can provide a climatic data set that offers access to short and long-term time series of climate parameters when knowledge of modern lake water balance is combined with lacustrine sedimentary-climate records. Lake sediment records ideally provide high resolution climate/environmental information of the last 10,000 years (Magny et al. 2013). This time interval (corresponding to the Holocene) is a key period to investigate short and long-term climate variability and to improve prediction in a warming climate. In this study we present an integrated approach focusing on a mountainous lake (Aguelmam Azigza). The modern lake system study is based on site monitoring (2012-2016) and available regional hydro-climatic data. These data show that lake level changes during the instrumented period were mainly driven by precipitation following the high inter-annual variability. These data are then compared with accurately dated short sediment cores retrieved in the same lake. Micro-scale geochemical and sedimentological analyses of these sequences enable us to identify various sedimentary facies that can be linked with periods of high (low) lake levels over the past decades.