干旱区生态环境知识资源中心(Arid): Predicting potential reforestation areas by Quercus ilex (L.) species using machine learning algorithms: case of upper Ziz, southeastern Morocco

Arid

DOI	10.1007/s10661-023-11680-1
	Predicting potential reforestation areas by Quercus ilex (L.) species using machine learning algorithms: case of upper Ziz, southeastern Morocco
	Manaouch, Mohamed; Sadiki, Mohamed; Pham, Quoc Bao; Zouagui, Anis; Batchi, Mohcine; Al Karkouri, Jamal
通讯作者	Manaouch, M
来源期刊	ENVIRONMENTAL MONITORING AND ASSESSMENT
ISSN	0167-6369
EISSN	1573-2959
出版年	2023
卷号	195 期号:9
英文摘要	The selection of appropriate areas for reforestation remains a complex task because of influence by several factors, which requires the use of new techniques. Based on the accurate outcomes obtained through machine learning in prior investigations, the current study evaluates the capacities of six machine learning techniques (MLT) for delineating optimal areas for reforestation purposes specifically targeting Quercus ilex, an important local species to protect soil and water in upper Ziz, southeast Morocco. In the initial phase, the remaining stands of Q. ilex were identified, and at each site, measurements were taken for a set of 12 geo-environmental parameters including slope, aspect, elevation, geology, distance to stream, rainfall, slope length, plan curvature, profile curvature, erodibility, soil erosion, and land use/land cover. Subsequently, six machine learning algorithms were applied to model optimal areas for reforestation. In terms of models' performance, the results were compared, and the best were obtained by Bagging (area under the curve (AUC) = 0.98) and Naive Bayes (AUC = 0.97). Extremely favorable areas represent 8% and 17% of the study area according to Bagging and NB respectively, located to the west where geological unit of Bathonian-Bajocian with low erodibility index (K) and where rainfall varies between 250 and 300 mm/year. This work provides a roadmap for decision-makers to increase the chances of successful reforestation at lower cost and in less time.
英文关键词	Forest restoration Holm oak Site suitability Machine learning Semi-arid areas
类型	Article
语种	英语
收录类别	SCI-E
WOS记录号	WOS:001093863000003
WOS关键词	GIS ; CLASSIFICATION
WOS类目	Environmental Sciences
WOS研究方向	Environmental Sciences & Ecology
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/396182
推荐引用方式 GB/T 7714	Manaouch, Mohamed,Sadiki, Mohamed,Pham, Quoc Bao,et al. Predicting potential reforestation areas by Quercus ilex (L.) species using machine learning algorithms: case of upper Ziz, southeastern Morocco[J],2023,195(9).
APA	Manaouch, Mohamed,Sadiki, Mohamed,Pham, Quoc Bao,Zouagui, Anis,Batchi, Mohcine,&Al Karkouri, Jamal.(2023).Predicting potential reforestation areas by Quercus ilex (L.) species using machine learning algorithms: case of upper Ziz, southeastern Morocco.ENVIRONMENTAL MONITORING AND ASSESSMENT,195(9).
MLA	Manaouch, Mohamed,et al."Predicting potential reforestation areas by Quercus ilex (L.) species using machine learning algorithms: case of upper Ziz, southeastern Morocco".ENVIRONMENTAL MONITORING AND ASSESSMENT 195.9(2023).