Partitioning Variation in Vegetation and Landscape Functionality in Arid Land Piosphere
Abstract
A piosphere is an interaction between vegetation community, watering point, and grazing livestock/wildlife. In this dissertation, I begin by reviewing the concept of the piosphere, and the progress, knowledge gaps and common statistical approaches and errors in piosphere research. Landscape functionality, plant community distribution and their influencing factors on a piosphere in Iran are then investigated. A total 862 piosphere publications of multiple types from 68 countries and 10 different languages during the period of 1915 – 2018 were reviewed. Australia was the most productive country followed by South Africa, the USA, Botswana and Argentina. Ivan Thrash is the most contributed author with 11 publications. The most frequently cited article is "Responses of Mediterranean grassland plants to grazing and protection" by Noy-Meir et al. (1989) in The Journal of Ecology. Rapid growth was found in piosphere research since 2000. Vegetation, soil, wildlife, and livestock were the most frequent research topics of piosphere investigations. Advances in statistics, remote sensing, GIS and programming enlivened piosphere research, especially since the 1990s. One-way ANOVA, multiple linear regression, Pearson correlation, permutational multivariate analysis of variance, canonical correspondence analysis, mean and Tukey's honest significance test were the most frequent statistical methods used in piosphere investigations. Seventy-one common statistical errors in piosphere publications were found. The most common were not choosing proper statistical techniques, not checking the assumptions and diagnostics of statistical techniques, partial and wrong interpretation of results and not using informative figures and tables. Two field research studies of Lajaneh piosphere, Iran, were completed. In the first one, soil surface condition indices of infiltration, stability, and nutrient cycling were measured along a grazing gradient from an artificial watering point in Lajaneh piosphere, and landscape functionality of the piosphere was investigated. One-way ANOVA and Tukey’s post hoc tests were used to discover significant differences in infiltration, stability and nutrient cycling indicators among three distances 10, 100 and 1000 m from the watering point. The three indices of nutrient cycling, infiltration and stability increased with increasing distance from the watering point. In the second field study, Linear Discriminant Analysis (LDA), Redundancy Analysis (RDA) and partial Redundancy Analysis (pRDA) were used to establish and evaluate vegetation distribution as related to soil phosphorus, soil potassium, distance from water, slope, aspect, and elevation at Lajaneh piosphere. The cover of palatable plant species increased with distance from the artificial watering point. Although the redundancy analysis resulted in much-unexplained variation, I found the role of soil fertility factors and distance from the water most important to plant species composition; soil nitrogen (N) 7.26 %, soil phosphorus (P) 5.13 %, soil potassium (K) 4.47 %, and distance from water (1.77 %) explained variations in vegetation community more than elevation (0.91%), aspect (0.42%) and slope (0.05%). In general, grazing pressure decreased with distance from water and correlated with improved soil chemical characteristics. Collectively, findings from this dissertation improve the understanding of the size and effect of piospheres and environmental stresses in arid and semi-arid regions and provide suggestions for managing natural and artificial watering points.
Subject Area
Environmental science|Ecology|Land Use Planning
Recommended Citation
Shahriary, Eahsan, "Partitioning Variation in Vegetation and Landscape Functionality in Arid Land Piosphere" (2020). ETD Collection for University of Texas, El Paso. AAI27998113.
https://scholarworks.utep.edu/dissertations/AAI27998113