A comparison of distance sampling methods in saxaul (Halloxylon ammodendron C.A. Mey Bunge) shrub-lands

The primarily goal of plot-less sampling methods is to reduce costs and rapid survey of plant communities. First full inventory was conducted in two 30-ha sites of Saxaul populations geo-morphologically different. In first site (site I), population had random pattern while in second site (site II) clumped pattern was observed. Crown diameters and spatial situation of all trees were recorded using distance and azimuth. Data were transferred to computer and stem map was generated with ArcGIS Software. Distance sampling methods include point-centred quarter method (PCQ), joint-point method (JP), Random pairs method (RP), T-Square method (T-Sq) and Quartered neighbour methods beside fixed area plot (FAP), n-tree and variable area transect (VAT) methods were conducted on generated stem maps. A time study was done aiding indices determined in field works. In site I, point centred quarter estimator with measurements to the second closest individual in each quadrant had the lowest relative bias (RBIAS) in estimating density followed by 3-tree and closest individual methods. In clumped pattern (site II), variable area transect method with measurements to the 4th and 5th closest individuals in each transect brought the best results. The most time consuming methods after fixed area plot, were point centred quarter estimators while methods considering measurement to the closest individual were rapid. Considering RBIAS and Time together, VAT method was the best sampling method in clumped pattern followed by point centred quarter estimator with measurements to the second closest individual in each quadrant and closest individual estimators. In random pattern, point centred quarter estimator with measurements to the second closest individual in each quadrant was the best method followed by 3-tree and closest individual estimators. But for estimating cover per unit area N-tree methods performed well. As in this site, VAT method located in 4th grade, and due to simplicity of field works related to this method, in the case that the investigator would not be able to clearly define spatial pattern of the population, this method can be recommended as well.