Comparing distance-based methods of measuring plant density in an arid sparse scrubland: testing field and simulated sampling.

Abstract

Plant density is an important indicator of rangeland monitoring and management. In this paper, we aimed at comparing distance-based methods of density measurement in field and in a simulated approach for two species of Astragalus verus Olivier and Astragalus albispinus Sirj. & Bornm with random and regular distribution pattern respectively. Then, we simulated a sampling scheme based on field measured statistics of the two species using Stochastic Geometry program and repeated the measurements using the same density methods. Moreover, the total individual plants of each species were counted in the sampling plots as the control method in Marjan rangelands of Iran during summer 2018. Results indicated that the best method for A. verus is nearest neighbor and point-centered quarter in field and simulated datasets in terms of both accuracy and precision respectively. For A. albispinus, the nearest neighbor and third closest individual were the best methods when compared with control in term of precision and accuracy in the field respectively and third closet individual in simulated dataset in terms of both accuracy and precision compared with control, respectively. The results reveal that there is no statistically significant difference between the nearest neighbor and point-centered quarter for A. verus and third closest individual and nearest neighbor for A. albispinus when field/simulated datasets are compared with each other and controls. Therefore, we recommend using these distance methods when this is a known species distribution pattern, basic number per unit, and dimensions of the unit as an alternative.