Assessment of drought frequency, severity, and duration and its impacts on vegetation greenness and agriculture production in Mediterranean dryland: A case study in Tunisia

2020 
Drought is among the highest-impact natural hazards affecting drylands around the world in a warming climate. The Mediterranean region, including Tunisia, is projected to experience the most predominant drying trends worldwide. However, a detailed regional scale study of drought for Tunisia has been limited, hampering an assessment of drought impact on the ecosystem and society. This study explored drought characteristics and its effect on vegetation greenness and agricultural productivity in three vegetation zones of Tunisia during 1982–2011, taking advantage of both meteorological and soil moisture drought indices and identifying the most appropriate index for each zone. The results revealed that meteorological droughts were short and frequent, triggering soil moisture droughts that were long-lasting and intense. The standardized precipitation index was identified as the best indicator of vegetation and agricultural droughts in the Northern forest, while the Palmer Drought Severity Index was best in the Central steppe and Southern desert (no crop data in the Southern desert). The lag-correlation analysis revealed that the response of both vegetation greenness and wheat productivity to droughts was more pronounced and had a longer significant lag in the Central steppe than in the other regions. These results suggest that arable land species (Central steppe and Southern desert), characterized by shallow roots, have a rapid response to rainfall variability when compared with forest species (Northern forest), which have deep roots allowing them an adequate supply of moisture. The region-specific indices identified here will provide a useful measure for drought monitoring and mitigation in Tunisia.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    65
    References
    2
    Citations
    NaN
    KQI
    []