Monotemporal assessment of the population structure of Acacia tortilis (Forssk.) Hayne ssp. raddiana (Savi) Brenan in Bou Hedma National Park, Tunisia: A terrestrial and remote sensing approach

• Field-based measurements were confronted with remote sensing techniques to model key tree attributes of A. tortilis.
• Remote sensing based models were reasonable accurate and considered valuable.
• Main strengths of remote sensing methods are their spatially explicit character and their automatic and repeatable usability.

Soil erosion and desertification are the main problems faced by the Bou-Hedma National Park in South-Tunisia. Restoration of the original woodland cover, particularly by afforestation and reforestation with Acacia tortilis ssp. raddiana, has been recognized as efficient to combat further degradation of the environment in this protected area. In order to study effects of woodland restoration and future trends in Bou-Hedma, it is essential to accurately assess its current situation. This paper addresses a monotemporal assessment of the population structure of A. tortilis. An extensive field inventory was performed to provide deeper insight into the dendrometric characteristics of this keystone species. Next, a spatially explicit and repeatable method is developed to model key tree attributes like crown diameter, volume and tree height from which the structural composition of the A. tortilis population in the Bou Hedma National Park is derived. This method involves analysis of a very high resolution (VHR) GeoEye-1 image within an OBIA (Object-based Image Analysis) framework. The remote sensing (RS) results show that the population of A. tortilis is typified by an irregular population structure and confirm the findings of Noumi and Chaieb (2012) suggesting regressive population dynamics. The RS approach demonstrates potential for monitoring purposes in this particular setting of an arid environment.