打印

作者

来源

LANDSCAPE AND URBAN PLANNING,Vol.169,P.70-80

语言

英文

关键字

Private lands conservation; Invasive species; Conservation behavior; Boosted regression trees; Species distribution models; BOOSTED REGRESSION TREES; COLLECTIVE ACTION; URBAN ECOSYSTEMS; WEED MANAGEMENT; SPECIES CONTROL; UNITED-STATES; BIODIVERSITY; LANDS

作者单位

[Niemiec, R. M.] Stanford Univ, Emmett Interdisciplinary Program Environm & Resou, 473 Via Ortega Way,Suite 226, Stanford, CA 94305 USA. [Asner, G. P.; Brodrick, P. G.] Carnegie Inst Sci, Dept Global Ecol, Stanford, CA 94305 USA. [Gaertner, J. A.] Univ Hawaii, Dept Geog, Hilo, HI 96720 USA. [Ardoin, N. M.] Stanford Univ, Grad Sch Educ, 485 Lausen Mall, Stanford, CA 94305 USA. [Ardoin, N. M.] Stanford Univ, Woods Inst Environm, 485 Lausen Mall, Stanford, CA USA. Niemiec, RM (reprint author), Stanford Univ, Emmett Interdisciplinary Program Environm & Resou, 473 Via Ortega Way,Suite 226, Stanford, CA 94305 USA. E-Mail: rniemiec@stanford.edu; gpa@carnegiescience.edu; pbrodrick@camegiescience.edu; jgaertne@hawaii.edu; nmardoin@stanford.edu

摘要

To reduce the spread and impacts of invasive species in human-dominated landscapes, numerous and diverse residents often need to engage in individual as well as collective invasive species control efforts on their property and in their community. Combating invasion thus requires an understanding of what socioeconomic factors, in addition to ecological factors, may slow an invasive species or facilitate its spread. However, few studies have examined associations between invasive species and multi-scale socioeconomic and environmental factors, which may influence residents' decisions to control invaders. We combine spatially explicit social and environmental datasets, and we apply gradient boosting regression to examine the socioeconomic, land use, and environmental factors associated with the distribution of albizia (Falcataria moluccana), an invasive tree species, in Hawaii. We find that environmental factors are the dominant controls on albizia cover at the landscape scale, but socioeconomic variables lead to a modest improvement in the ability to predict albizia distribution at the housing subdivision scale. At the subdivision scale, albizia is more common on properties with absentee and/or less-wealthy landowners. Albizia is also more common on smaller properties and non-agricultural land. Our study provides policy recommendations for reducing the spread of invaders and outlines an approach using computation machine learning for examining multiple socioeconomic and environmental factors associated with biological invasion in complex social landscapes.