Habitat protection and planning for three Ephedra using the MaxEnt and Marxan models

As renowned species for use in sand fixation and traditional Chinese medicines, Ephedra sinica Stapf, Ephedra intermedia Schrenk ex Mey and Ephedra equisetina Bunge have important ecological and medicinal value. However, although conservation measures have been implemented, the suitable habitat for these species has been shrinking at a remarkable rate in recent years due to climate change and anthropogenic activities. The identification of suitable habitats for these three Ephedra is essential for their protection and cultivation. Thus, this study proposed a new method for planning the best species conservation habitat through fitting errors caused by a single model. The method modeled the current suitable habitats of the three Ephedra species using the MaxEnt model and evaluated the key environmental variables defining the Ephedra distribution based on recorded occurrence data for Xinjiang, Inner Mongolia, Gansu, and Qinghai. The results showed that human activities were the most important environmental parameter affecting the distribution of the three species, and land cover, ground frost frequency and precipitation accounted for an important proportion of the model evaluation. Second, the results obtained with the MaxEnt model were coupled with the finding from the Marxan model to identify priority conservation areas for the three Ephedra species. The results revealed that the most suitable habitats for E. sinica were mainly concentrated in eastern and central Inner Mongolia, E. intermedia was relatively distributed in southern and central Gansu, and the areas of E. equisetina were mainly distributed in northern Xinjiang. The priority conservation areas for E. sinica, E. intermedia, and E. equisetina were 38068.12 km2, 26446.74 km2 and 36360.6 km2, respectively. The comprehensive analysis indicated that be use of MaxEnt coupled with the Marxan model to model the suitable habitat of these Ephedra species eliminates the distribution errors introduced by each model alone. The protection of wild Ephedra should be a priority, and the scale of the artificial planting of Ephedra should be enlarged. Given these results, the three Ephedra species included in our study urgently need to be subjected to targeted conservation management to ensure the renewal of Ephedra communities and to maintain the supply of medicinal materials. This study provides a useful reference for the protection and cultivation of these three Ephedra.