Predicting Cork Oak Suitability in Ma&amp;#226;mora Forest Using Random Forest Algorithm

Lahssini, Saïd; Lahlaoi, Hicham; Alaoui, Hicham Mharzi; Hlal, El Aid; Bagaram, Martin B.; Ponette, Quentin

doi:10.4236/jgis.2015.72017

Cited by 16 publications

(13 citation statements)

References 17 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning methods are becoming increasingly popular in land suitability analysis thanks to their ability to deal with complex relationships between predictor variables, robustness in managing big and noisy data, and being economical in terms of time required (Lahssini et al 2015). RF, as proposed by Breiman (2001, p. 6), is "a classifier consisting of a collection of tree-structured classifiers {h(x,Θk ), k=1, ...} where the {Θk} are independent identically distributed random vectors and each tree casts a unit vote for the most popular class at input x".…”

Section: Random Forest Methodsmentioning

confidence: 99%

“…However, many studies have shown that RF most often attains the best predictive performance (Garzon et al 2006). Lahssini et al (2015) and Vincenzi et al (2011) used RF to detect cork oak suitability and Ruditapes philippinarum's potential spatial distribution assessment respectively. The probability of correct predictions in both studies was more than 90%.…”

Section: Random Forest Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Land Suitability Analysis of Alvar Grassland Vegetation in Estonia Using Random Forest

Ismayilova¹,

Uuemaa²,

Helm³

et al. 2020

giforum

View full text Add to dashboard Cite

Calcareous alvar grasslands are one of the most species-rich habitats in Estonia. Land-use change and cessation of traditional agricultural practices have led to a decrease of the area of these valuable grasslands during the past century. Therefore, their conservation and restoration are becoming increasingly important. Efforts to restore these habitats have already been made in recent years. Land suitability analysis for potential restoration sites, using the machine learning technique Random Forest (RF), was performed for the first time in this study, which aimed to assess the use of RF for a suitability analysis of alvar grassland. RF predicted 610.91 km 2 of areas suitable for restoring alvar grasslands or for creating alvarlike habitats in Estonia. These areas include all existing alvar areas as well an additional 140.91 km 2 suitable for establishing new habitat similar to calcareous alvar grasslands. We discuss suitability analysis to help with restoration planning and find it to be a reasonable and efficient tool that has potential to provide relevant information. The quality of the prediction could be improved by including additional data relevant for alvar grasslands, such as soil depth, but such data was unfortunately unavailable.

show abstract

Section: Random Forest Methodsmentioning

confidence: 99%

Section: Random Forest Methodsmentioning

confidence: 99%

Land Suitability Analysis of Alvar Grassland Vegetation in Estonia Using Random Forest

Ismayilova¹,

Uuemaa²,

Helm³

et al. 2020

giforum

View full text Add to dashboard Cite

show abstract

“…Even though many studies on land suitability are conducted in agriculture [11]- [13], there are few in forestry. The land suitability methods are diverse and range from random forest [14], GIS techniques [15] [16], fuzzy logic [17] and AHP [12] but few authors used the integrated approach combining GIS (multicriteria analysis), fuzzy logic and AHP in the same study even though according to Zang et al [11], the integrated approach has a great potential to increase the effectiveness and accuracy of land suitability assessment.…”

Section: Introductionmentioning

confidence: 99%

Site Suitability Analysis for Cork Oak Regeneration Using GIS Based Multicriteria Evaluation Techniques in Maamora Forest-Morocco

Bagaram¹,

Mounir²,

Lahssini³

et al. 2016

OALib

Self Cite

View full text Add to dashboard Cite

Following the assumption that some soils are less suitable for the development of cork oak, a substitution of cork oak by exotic species has been carried out on a large part of Maamora forest. However, soil is not the only factor to consider in assessing the suitability. Therefore, some areas might just suit to cork oak but to what extent? To answer this question, major factors known to influence the regeneration of cork oak have been identified and mapped using multicriteria evaluation. Therefore, factors were incorporated by weighted linear combination to determine suitability indices for the regeneration of cork oak. The weight of each factor was estimated by expert judgment using the AHP (Analytic Hierarchy Process) approach. The main factors influencing the cork oak regeneration suitability in Maamora forest are climate expressed by "continentality", terrainslope, thickness of the sand, slope of the clay floor layer, and plant communities and soil types. These factors were mapped by incorporating fuzzy logic in order to avoid crisp classes and ensure a smooth transition between the preference and indifference. This allowed then to integrate uncertainty in the knowledge of the exact class limits. Finally, four classes of regeneration suitability were obtained. These classes are good, medium, low and very low suitability and which represent 17.40%, 40.18%, 34.84% and 4.28% of the forest area respectively.

show abstract

“…In SDM, many statistical models could be used (Hegel et al, 2010;Franklin, 2009). In addition to classical regression methods, machine learning based modeling is widely used including: Artificial Neural Networks (Ripley 1996); Maximum Entropy MaxEnt (Phillips et al, 2004); Random Forest (Lahssini et al, 2015); Classification and Regression Trees CART (Breiman et al, 1984). MaxEnt algorithm is the most popular and widely used and qualified as most efficient in handling complex interactions between response and predictor variables (Elith et al, 2011;Elith et al, 2006) and less sensitive to small sample sizes (Wisz et al, 2008).…”

Section: Introductionmentioning

confidence: 99%

Modelling Spatial Distribution of the Carob Tree (Ceratonia siliqua L.) in Azilal Province, Morocco

Lahssini¹,

Hajib²,

Lahlaoi³

et al. 2015

JGG

Self Cite

View full text Add to dashboard Cite

Factors determining forest species distribution include, in addition to external factors such as human interference and environmental management strategies, also soil and hydrological characteristics and climate conditions in any given areas. Modelling distribution has practical application in forest conservation and management, and help decision makers to develop strategies aimed at forest restoration, development of mountainous areas and the continuous and sustainable provision of forest-related services. Species distribution modelling (SDM) can be used for predicting species distribution based on tree presence records and on a number of environmental predictors. In this study we used MaxEnt for niche modelling in predicting carob (Ceratonia siliqua L.) trees spatial distribution in the Province of Azilal in Morocco. The results obtained show that a large area of the mountain regions is suitable for the expansion of Ceratonia siliqua stands. These findings will help decision makers in forest planning to better identify suitable sites for carob tree plantations and assess the potential of the exiting populations.

show abstract

Predicting Cork Oak Suitability in Maâmora Forest Using Random Forest Algorithm

Cited by 16 publications

References 17 publications

Land Suitability Analysis of Alvar Grassland Vegetation in Estonia Using Random Forest

Land Suitability Analysis of Alvar Grassland Vegetation in Estonia Using Random Forest

Site Suitability Analysis for Cork Oak Regeneration Using GIS Based Multicriteria Evaluation Techniques in Maamora Forest-Morocco

Modelling Spatial Distribution of the Carob Tree (Ceratonia siliqua L.) in Azilal Province, Morocco

Contact Info

Product

Resources

About