Predicting Mangrove Distributions in the Beibu Gulf, Guangxi, China, Using the MaxEnt Model: Determining Tree Species Selection

Abstract

Mangrove restoration is challenging within protected coastal habitats. Predicting the dominant species distributions in mangrove communities is essential for appropriate species selection and spatial planning for restoration. We explored the spatial distributions of six mangrove species, including their related environmental factors, thereby identifying potentially suitable habitats for mangrove protection and restoration. Based on six dominant mangrove species present in the Beibu Gulf, Guangxi, China, we used a linear correlation analysis to screen environmental factors. In addition, we used the maximum entropy model to analyze the spatial distributions of potential mangrove afforestation areas. Based on the spatial superposition analysis, we identified mangrove conservation and restoration hot spots. The findings indicate that topographic and bioclimatic factors affect the distribution of suitable mangrove habitats in the Beibu Gulf, followed by land use type, salinity, and substrate type. We identified 13,816 hm2 of prime mangrove habitat in the Beibu Gulf that is primarily distributed in protected areas. The protection rate for existing mangroves was 42.62%. According to the predicted spatial distributions of the mangrove plants, the findings suggest that mangrove restoration should be based on suitable species and site selection.

Climate change may drive the distribution of tribe Zyginelline pests in China and the Indo‐China Peninsula to shift towards higher latitude river‐mountain systems

“…The RF algorithm employs multiple decision trees and makes predictions based on the class with the highest number of votes, 47 and has been successfully used to predict species distribution ranges 31,41,48–52 . Additionally, the MaxEnt estimates the geographical range of a species by finding the distribution with maximum entropy subject to constraints derived from environmental conditions at recorded occurrence locations, 37 with over 1000 published ecological applications having demonstrated its usefulness and reliable predictive performance 19,53–56 . In addition to the aforementioned (2) settings, for RF, the number of trees was set to 500, mtry was set to one‐third of variables, and the nodesize was set to 5 57–60 .…”

“…31,41,[48][49][50][51][52] Additionally, the MaxEnt estimates the geographical range of a species by finding the distribution with maximum entropy subject to constraints derived from environmental conditions at recorded occurrence locations, 37 with over 1000 published ecological applications having demonstrated its usefulness and reliable predictive performance. 19,[53][54][55][56] In addition to the aforementioned (2) settings, for RF, the number of trees was set to 500, mtry was set to one-third of variables, and the nodesize was set to 5. [57][58][59][60] For MaxEnt, the type of features was set to 'auto', and the regularization multiplier was set to 1.…”

Phenomenon and Mechanisms of Sonneratia apetala Introduction and Spread Promoting Excessive Growth of Derris trifoliata

“…The MaxEnt model 3.4.1 (Columbia University in the United States, New York, NY, USA) with the Kuenm package optimization parameters [16] was used to predict the spatial distributions of D. trifoliata and S. apetala. The sources of environmental data (Table 1) and the selection of environmental factors (Table 2) in this study refer to the research of Li et al [13]. This study obtained the parameters for the MaxEnt model using the Kuenm package.…”

“…Larger pixel values indicated higher potential distributions and higher habitat suitability of D. trifoliata and S. apetala. In this study, we used the natural breakpoint method to grade the fitness results as follows: 0-0.2, no fitness; 0.2-0.5, low fitness; 0.5-0.7, medium fitness; and >0.7, high fitness [13,17].…”

“…Derris trifoliata Lour., the fish vine, is another species of concern that has experienced significant spread and excessive growth in mangrove areas [12][13][14]. The harmful effects of D. trifoliata were predominantly noted in communities dominated by Aegiceras corniculatum (L.) Blanco.…”