Next Article in Journal
Water Balance Characteristics of the Salix Shelterbelt in the Kubuqi Desert
Previous Article in Journal
Functional Analysis of the HbREF1 Promoter from Hevea brasiliensis and Its Response to Phytohormones
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 15
Issue 2
10.3390/f15020277
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Modeling the Spatial Distribution of Acacia decurrens Plantation Forests Using PlanetScope Images and Environmental Variables in the Northwestern Highlands of Ethiopia

by
Bireda Alemayehu
1,2,*,
Juan Suarez-Minguez
3 and
Jacqueline Rosette
4
1
Space Science and Geospatial Institute, Addis Ababa P.O. Box 33679, Ethiopia
2
Department of Geography and Environmental Studies, Debre Markos University, Debre Markos P.O. Box 269, Ethiopia
3
Forest Research Agency of the Forestry Commission, Northern Research Station, Midlothian, Edinburgh EH25 9SY, UK
4
Department of Geography, Swansea University, Swansea SA2 8PP, UK
*
Author to whom correspondence should be addressed.
Forests 2024, 15(2), 277; https://doi.org/10.3390/f15020277
Submission received: 19 December 2023 / Revised: 18 January 2024 / Accepted: 25 January 2024 / Published: 1 February 2024
(This article belongs to the Section Forest Inventory, Modeling and Remote Sensing)
Browse Figures
Versions Notes

Abstract

:
Small-scale Acacia decurrens plantation forests, established by farmers on degraded lands, have become increasingly prevalent in the Northwestern Highlands of Ethiopia. This trend has been particularly notable in Fagita Lekoma District over the past few decades. Such plantations play a significant role in addressing concerns related to sustainable agricultural land use, mitigating the adverse effects of deforestation, and meeting the livelihood and energy requirements of a growing population. However, the spatial distribution of Acacia decurrens and the essential remote sensing and environmental variables that determine its distribution are not well understood. This study aimed to model the spatial distribution of Acacia decurrens plantation forests using PlanetScope data and environmental variables combined with a species distribution model (SDM). Employing 557 presence/absence points, noncollinear variables were identified and utilized as input for six SDM algorithms, with a 70:30 split between training and test data, and 10-fold bootstrap replication. The model performance was evaluated using the receiver operation characteristic curve (AUC) and true skill statics (TSS). The ensemble model, which combined results from six individual algorithms, was implemented to predict the spatial distribution of Acacia decurrens. The highest accuracy with the values of 0.93 (AUC) and 0.82 (TSS) was observed using random forest (RF), followed by SVM with values of 0.89 (AUC) and 0.71 (TSS), and BRT with values of 0.89 (AUC) and 0.7 (TSS). According to the ensemble model result, Acacia decurrens plantation forests cover 22.44% of the district, with the spatial distribution decreasing towards lower elevation areas in the northeastern and western parts of the district. The major determinant variables for identifying the species were vegetation indices, specifically CVI, ARVI, and GI, with AUC metric values of 39.3%, 16%, and 7.1%, respectively. The findings of this study indicate that the combination of high-resolution remote sensing-derived vegetation indices and environmental variables using SDM could play a vital role in identifying Acacia decurrens plantations, offering valuable insights for land use planning and management strategies. Moreover, comprehending the spatial distribution’s extent is crucial baseline information for assessing its environmental implications at a local scale.

1. Introduction

Plantation forests are forests established by planting or deliberate seeding to achieve principally economic goals [1] such as timber, energy, fiber, and non-woody forest products [2]. They are also established for soil and water conservation and carbon sequestration in the process of afforestation and reforestation [3]. According to the FAO [2] report, plantation forests cover about 131 million ha worldwide and account for 3% of the global forest. Between 1990 and 2020, their area increased by 55.8 million ha, with the biggest jump (21.2 million ha) recorded between 2000 and 2010.
The impact of plantation forests on the environment will depend on what land use they replace [4]. If they are established on frequently cultivated land for a long time or degraded lands, they may provide substantial opportunities for biodiversity conservation [5] and deliver vigorous ecosystem service [6], but plantations converted from the natural forest have adverse impacts on biodiversity [7]. Plantation forests are essentially significant in fragmented landscapes, where they may account for a large amount of remaining forest habitat [8] and can serve as corridors between habitats [4].
Small-scale plantation forests initiated by farmers on degraded lands have become important in Ethiopia, particularly since the mid-1990s [9]. An estimated area of 754,900 ha of the country is covered by small-scale plantation forests, and of this, 84.7% (639,400 ha) is found in the Amhara Region [9]. Critical levels of land degradation and reduced productivity forced farmers to start planting trees, often with predominantly exotic and fast-growing species at the expense of crop production [10,11,12].
In the Northwestern Highlands of Ethiopia, specifically in the Fagita Lekoma District, the study site and one of the districts in the Amhara Region, growing Acacia decurrens plantations on small-size farmlands have been rapidly increasing and are widely planted because of the economic and environmental benefits. This species is preferable due to its advantages of a fast-growing rate and adaptability to degraded and acidic soil conditions [13]. Additional reasons that motivate farmers to plant this species are its use as fuel and construction wood, animal fodder, charcoal production and availability of market, and soil fertility maintenance [14]. It also creates job opportunities for landless community parts and supports local livelihoods and rural developments, especially when managed by smallholders [13,15]. Land-use change from cultivated land and grassland into Acacia decurrens plantation has been common in the district in the past three decades and resulted in an increase in the forest cover of the district by more than 250% between 1887 and 2015 [16], and around 400% between 2006 and 2017 [17]. A range of ecosystem services can be obtained from plantation forests established on degraded lands that require restoration [18]. New plantation forests generated from former agricultural land can improve ecosystem service [19].
The rapid expansion of plantation forests significantly affects climate, hydrology, biodiversity, and the terrestrial carbon cycle. The expansion of plantation forests can amend the understory climate condition and soil properties [20], and water quality [21]. These changes occur as a result of changes in temperature, rainfall, land use type, and storm frequency and magnitude [4]. In addition, plantation forests can contribute strongly to regulating the environment, biodiversity, and socioeconomic functions, especially carbon sequestration [22]. To monitor such dynamics, remote sensing is an essential and effective source of data [23].
Remote sensing technologies drive developments in forest resource assessments and monitoring at various scales. They enable the provision of airborne and spaceborne data with a higher spatial resolution, frequent coverage, and expanded spectral coverage [24]. The remote sensing-based assessment of forest study is repetitive, affordable, competent, and non-destructive for monitoring [25]. Recently, in complementarity with field data, it has shown great adaptability in environmental studies such as droughts [26], floods [27], the spread of invasive species [28], disturbance [29], and other human-induced pressures [30]. The contribution of satellite data is becoming impressive to monitor the spatial distribution and temporal dynamics of plantation forests [31]. Optical data are spectrally sensitive to different species and can distinguish phenological characteristics unique to a particular plantation [32]. Different plantation forests can have distinct implications for the local ecosystem service [33].
The geographic distribution of species is dynamic at accelerating rates because of anthropogenic pressures, the introduction of non-native species, and climate change [34]. To understand the distribution of introduced or expanding species, researchers often map the suitability status of the habitat or the potential occurrence probability of species using different techniques such as expert opinion [35], mathematical models [36], or machine learning algorithms [37]. These methods help researchers and decision-makers to identify priority areas for environmental conservation [38], examine landscape planning approaches on the management and restoration of protected areas [39], assess species distribution under changing anthropogenic or environmental conditions [40], investigate the impact of environmental changes on the biodiversity [41], and model the invasion status of invasive species [28,42].
The species distribution model (SDM) is a popular technique in ecology and conservation biology to assess the impact of land use and climate changes on biodiversity distribution [43], predict species diversity and composition patterns over space and time [44], and provide spatially explicit and compressive maps that are specifically important to understand the distribution level and extent of a given species. It combines observations of species occurrence or abundance with environmental variables [45]; its performance depends on the collected data during field surveys and exists as simple presence/absence records, which are crucial to train and validate the model [46]. Moreover, the accuracy of SDM varies among algorithms [47], and integrating multiple algorithms is more reliable to get robust estimations of species distribution [48].
The use of SDM algorithms in combination with remotely sensed datasets is effective for mapping plant species across different management levels at local and regional scales [49]. In Sub-Saharan Africa, agricultural landscapes are highly fragmented and this is one of the challenges complicating their mapping [50]. Fragmented parcels of land with verities of coverage highlighting the need of methods based on high resolution satellite imageries [51].
PlanetScope (PS) satellite constellation can achieve daily coverage with a spatial resolution of 3–5 m, visible to near-infrared and atmospherically corrected imagery [52], which has been successfully applied in many fields, for example, rubber plantation mapping [53], forest canopy height estimation [54], biomass estimation [55], leaf area index production [56], cropland mapping [51], and crop yield prediction [57]. It provides effective spatial data for the extraction of plantation forest and agricultural information in the tropical and subtropical regions [58], and offers a good opportunity to overcome challenges in mapping smallholder agricultural fields [51].
Understanding the interplay between the spatial distribution of a species and its environmental determinants is a fundamental concept in ecology and conservation [59]. Establishing plantation forests on agricultural or degraded land presents significant prospects for biodiversity conservation [60]. Consequently, the impact of plantation forests is contingent upon their spatial extent of landscape coverage [61] and the specific land use they replace [4]. Despite prior studies in the study area focusing primarily on land use/land cover changes across all classes [11,16,17,62,63], there is a noticeable gap in research specifically addressing the species-level identification of Acacia decurrens plantations through the utilization of high-resolution satellite imagery and environmental variables combined with machine learning algorithms. Examining such a spatial pattern is crucial in any study aiming to ensure the provision of goods and services [61]. Therefore, the objectives of this research were (i) to model the spatial distribution of Acacia decurrens plantation forests using high-resolution satellite imagery, (ii) to evaluate the performance of SDM algorithms for Acacia decurrens distribution modeling, and (iii) to identify the relative importance of predictor variables for Acacia decurrens distribution modeling. Modeling and understanding the spatial spread of such species are high priorities for resource managers to assess the environmental implications of the sustained use of plantation forests and to scale up for the other degraded areas of the country based on scientific findings and with great attention. This is because in Ethiopia, it is believed that small-scale tree plantations can contribute to addressing issues related to sustainable agricultural land use, mitigating the negative impacts of deforestation, and meeting the needs for the livelihood and energy of the growing population [64]. Having comprehensive information about its distribution enables effective control and management.

2. Data and Methods

2.1. Study Area

The study was conducted in Fagita Lekoma District which is part of the Northwestern Highlands of Ethiopia. The total area of the district is 65,579 ha [16], where the elevation extends from 1879 m to 2922 m above sea level (Figure 1). The mean daily temperature in the district ranges from 15 °C to 24 °C, and it receives an average annual rainfall of 2454 mm, with peak precipitation occurring between June and September [65]. The climatic conditions in the study area comprise 84% humid subtropical (Weynadega) and 16% moist subtropical (Dega) agroecological zones [16].
The major economic activity in the area is a mixed crop–livestock system [66] and recently, charcoal production from Acacia decurrens plantation has become another significant source of income for residents [67], replacing crop and grazing lands [16]. Consequently, the vegetation status of the district has increased over the last two decades, resulting in an improvement in the ecosystem condition [68].

2.2. Data

2.2.1. Remote Sensing Data

The PS satellite constellation comprises more than 180 satellites in orbit, allowing it to provide high-resolution data with daily global coverage [69]. For this study, cloud-free data from the PS satellite for February 2022 were collected. The images were a level 3B product that had undergone pre-processing, including radiometric and geometric correction [58]. They had a spatial resolution of 3m and consisted of four bands in the visible (blue, green, and red) and near-infrared regions of the spectrum. An image mosaic was created for six scenes of the PS image using QGIS 3.28 to fully cover the study area. Then, it was projected to Adindan Universal Transvers Mercator (UTM) Zone 37N.
Vegetation indices (Appendix A), known for their capabilities in detecting and mapping the distribution of plant species, were selected from the Index Database (https://www.indexdatabase.de/ accessed on 17 March 2023). Indices that have proven effective in characterizing various vegetation types were selected, particularly those sensitive to reflectance in the visible and NIR regions. These regions have been recognized as effective in discriminating commercial forest species [70]. The four spectral bands of PS and the thirty-one vegetation indices derived were tested to identify Acacia decurrens plantation forest species.

2.2.2. Environmental Variable Data

In addition to the remote sensing data, environmental variables such as elevation, slope, aspect, road proximity, temperature, and rainfall were used in this study. These variables can be categorized as climatic and topographic variables.
A significant correlation exists between climatic factors, such as temperature and rainfall, and the spatial distribution of forest cover [71]. Rainfall data from Climate Hazard Group InfraRed Precipitation with Station data (CHIRPS) were acquired from the Famine Early Warning Systems Network (FEWS NET) dataset, accessible at a spatial resolution of 5 km (https://earlywarning.usgs.gov/fews accessed on 24 March 2023). Additionally, land surface temperature (LST) data from Moderate Resolution Imaging Spectroradiometer (MODIS), with a spatial resolution of 250 m, were utilized for temperature data. It is noteworthy that forests can influence local temperatures, including LST [72].
Vegetation dynamics were determined using topographic factors such as elevation, slope, and aspect [73,74], as well as road accessibility [75]. Aspects, elevations, and slopes intricately govern the spatial and temporal distribution of critical elements, including radiation, temperature, and precipitation, thereby significantly influencing species composition [76]. A digital elevation model (DEM), extracted from ALOS PALSAR with a spatial resolution of 12.5 m (https://search.asf.alaska.edu/#/ accessed on 24 March 2023), was used to generate topographic variables such as the elevation, aspect, slope, and streams of the study area. The heterogeneity of soil types gives rise to niches characterized by specific conditions, consequently influencing the distribution patterns of plants [77]. All variables underwent various pre-processing stages, which involved resampling to match the spatial resolution of PS imagery and masking to align with the study area’s extent using QGIS 3.28.

2.2.3. Presence/Absence Data

Presence/absence data were collected in the field using Garmin eTrex GPS with an error below two meters, concurrently with the capture of PS imagery in February 2022. A stratified random sampling, based on land use/land cover types, was employed. A total of 557 points were collected, with the distribution proportional to the extent of each land use/land cover type in the district. Of these, 195 points (35%) represented presence data in Acacia decurrens planted areas, while the remaining 362 points (65%) were collected in croplands, grasslands, natural forests, and settlement areas. The proportions were determined based on previous studies conducted in the district by Teshome and Wondimu [62] and Worku et al. [63]. The minimum distance between two consecutive points was two hundred meters, and the Spatially Rarefy Tool in the SDM ToolBox v2.10 under ArcGIS 10.8 was used to reduce the spatial autocorrelation between points.

2.3. Variable Selection

Multicollinearity among predictor variables (i.e., remote sensing and environmental variables) was assessed using the Variance Inflation Factor (VIF) step (vifstep) function of usdm package. Variables above the threshold (greater than 10) were considered as collinear and excluded from further processing [78]. This test aids in the selection of predictor variables by assessing their relative importance [79].

2.4. Modelling Algorithms

Numerous SDM algorithms have been developed to predict species distribution based on environmental factors [80]. In this study, SDM was employed using six commonly used algorithms, which can be categorized into two regression methods, namely, generalized linear models (GLMs) [81] and multivariate adaptive regression spline (MARS) [82]; three machine learning methods, namely, boosted regression trees (BRT) [83], random forest (RF) [84], and support vector machine (SVM) [85]; and one classification and regression method, namely, classification and regression trees (CART) [86]. Furthermore, the prediction results of individual algorithms were combined based on their weighted mean of TSS to create an ensemble model [87], which is widely recognized as a powerful, well-referenced, and stable method for tree species prediction [88]. The weights assigned to the ensemble model were determined proportionally to the computed TSS derived from the cross-validation runs of the best performing models. This methodology ensures a nuanced and optimized weighting scheme for the ensemble model within the context of each model’s performance across multiple cross-validation iterations [89], and significantly enhances the accuracy of SDM [47].

2.5. Model Evaluation

Models were calibrated with 70% of the presence/absence data, while the remaining 30% were utilized to evaluate the predictive performance of each model [90]. This evaluation involved a 10-fold bootstrap replication, employing both threshold-independent metrics, such as area under a receiver operating characteristic (ROC) curve (AUC) [91], and threshold-dependent metrics, such as the true skill statistic (TSS) [92].
The AUC metric evaluates a model’s ability to distinguish between sites where a species is present and those where it is absent. It serves as an indicator of how effectively the models prioritize areas based on their suitability as habitat for a particular species [93]. A model is deemed excellent when the AUC is greater than 0.9, good for values between 0.8 and 0.9, acceptable for values between 0.7 and 0.8, poor for values between 0.6 and 0.7, and invalid for values between 0.5 and 0.6 [90].
The TSS is calculated as sensitivity (the proportion of observed presence to predicted presence, or true positive rate) plus specificity (the proportion of observed absence to predicted absence, or true negative rate) minus one [91]. TSS is not sensitive to prevalence, while keeping all the advantages of Kappa, such as considering omission and commission errors [92]. The TSS value ranges from −1 to 1, where +1 indicates perfect agreement between observations and predictions, and values of 0 or less implies results not better than random grouping [94]. The overall workflow adopted in this study is outlined in Figure 2 below.

3. Results

3.1. Multicollinearity Test

Strongly correlated variables were eliminated using the multicollinearity test. Out of forty-two variables used on the test, eleven predictor variables with a VIF below the specified threshold were selected (Table 1). Atmospheric Resistant Vegetation Index (ARVI), Aspect, Chlorophyll Vegetation Index (CVI), Green Index (GI), Modified Soil Adjusted Vegetation Index (mSAVI), Elevation, Rainfall, Road, Soil type, Slope, and Temperature were the variables that satisfied the VIF threshold value. After excluding the collinear variables, the linear correlation coefficients ranged between 0.004907765 (Soil type~GI) and 0.8013945 (CVI~ARVI).

3.2. Performance of Modelling Algorithms

All modelling algorithms for Acacia decurrens identification were generally effective in terms of both AUC and TSS values (Figure 3 and Table 2). AUC values ranged between 0.84 (GLM and CART) and 0.93 (RF). TSS ranged from 0.64 to 0.81 (Table 2). RF received the highest TSS, and GLM received the lowest TSS among all SDM algorithms. GLM, CART, and MARS had the lowest performance out of all six algorithms based on both AUC and TSS values, and RF achieved the highest performance. ROC plots are graphical representations of sensitivity (the true positive rate) plotted against 1-specificity (the false positive rate). In these plots, an algorithm that exhibits a curve closer to the top-left corner demonstrates superior performance compared to an algorithm with the curve closer to the 45-degree line within the ROC space. For instance, the ROC plots generated using the GLM and CART algorithms (Figure 3) exhibit a section of the plot that leans closer to the 45-degree line, indicating relatively lower performance. These results are different from the RF ROC plots, which illustrate relatively higher performance. Sensitivity and specificity scores were high across all algorithms, signifying the effective identification of both the presence and absence areas of Acacia decurrens (Table 2). This indicates that the proportion of correctly classified samples was maximum.
The presence values of Acacia decurrens plantation forest for the GLM, MARS, BRT, RF, SVM, and CART algorithms were 19.07%, 23.56%, 19.09%, 23.48%, 24.15%, and 24.72%, respectively. The best performing RF algorithm shows that Acacia decurrens covered an area of approximately 15,818.53 ha within the study area.
Ensemble modeling was employed to combine all the selected algorithms, reducing bias, and providing a relative evaluation of the significance of each predictor variable across all chosen modeling algorithms, with the intention of enhancing prediction performance [45]. According to the ensemble model result, 14,203.97 ha (22.44%) of the district was covered by Acacia decurrens, with a higher prevalence in the Southcentral and Central regions (Figure 4).

3.3. Variable Importance

CVI is the most contributing variable, followed by ARVI and GI, based on the average result of all used SDM algorithms. Most of the environmental predictors have relatively the same contribution for Acacia decurrens prediction (Figure 5). Moreover, the relative importance of each variable for Acacia decurrens identification determined and variables with below 1% contribution (Aspect and Soil type) were excluded from the final modelling [95].
The optimal environmental conditions that best represent the occurrence probability of Acacia decurrens concerning both the vegetation indices and environmental variables are presented in Figure 6 and Figure 7, respectively. They provide a quantitative representation of the relationship between predictors and the logistic probability of an Acacia decurrens distribution. These curves assist in comprehending the ecological niche of the Acacia decurrens within the predictor’s range. Peaks in each curve indicate the values that influenced optimal model performance. The shape, the location of the peak, and the range of values around the peak provide valuable information that influences the distribution of Acacia decurrens. Among vegetation indices, the response curves with the highest peak value were observed for CVI, with a high probability in areas that have a value between ~30 and 40, followed by GI (~0.8 and 0.9) and ARVI (~0.4 and 0.7). The response curve for temperature values exhibited the highest peak among the environmental variables, with a high probability in areas that have a temperature ~28 °C, followed by elevation (~2600 m) and rainfall (~470 mm), respectively (Figure 7).

4. Discussion

This study modeled the spatial distribution of small-scale Acacia decurrens plantation forests, utilizing high-resolution satellite data and environmental variables by employing six different algorithms available in the SDM package.
The assessment of multicollinearity was a crucial phase aimed at identifying and addressing strong correlations among multiple predictor variables used in the identification process. The implementation of the VIF method, recognized for discerning collinearity among predictor variables [78], resulted in the exclusion of almost three-fourths of the total variables due to collinearity issues, potentially decreasing the efficiency of prediction and increasing the uncertainty of the SDM [96]. When VIF exceeds 10, it serves as an indicator of collinearity issues within the model [97]. Furthermore, collinearity represents a significant concern that can potentially result in the incorrect identification of relevant predictor variables [98]. The high correlation among many of the input predictor variables can be attributed to the fact that observations were made within a relatively local scale and due to the similarity in spectral vegetation indices. It is worth noting that all the reduced variables were vegetation indices and raw bands, implying a relatively higher similarity between the generated vegetation indices. This result aligns with the findings of [99], possibly due to the limited spectral resolution of the PS image with only four bands.
The use of a combination of SDM algorithms, such as GLM, MARS, BRT, RF, SVM, and CART, in a complementary manner, along with the incorporation of accuracy estimators based on presence/absence data, enables a more effective representation of the spatial distribution of species at a local scale. The overall accuracy of the algorithms is relatively good, with the values exceeding 0.8 for AUC and 0.6 for TSS (Table 2). Predictive accuracy pertains to the ability of the algorithms to gauge the disparity between observed and predicted values [100]. RF exhibits an AUC above 0.9 and a TSS above 0.8, signifying near-perfect agreement. BRT and SVM demonstrate AUC values above 0.8 and TSS values above 0.7, indicating substantial agreement. The result aligns with the findings of Maxwell et al. [101], who found that machine learning outperformed regression algorithms for species identification. RF attains higher values than other algorithms for both evaluation metrics. This aligns with previous studies that have shown RF’s superior performance in species identification [28,102,103,104,105], remote sensing image classification [106,107], and data mining [108] compared to other algorithms. This is because the RF algorithm generates predictions by creating thousands of trees and aggregating their results through averaging [86]. This approach allows the algorithm to prevent overfitting, thereby enhancing predictive performance and reducing variance [109]. Thus, RF proves to be a robust technique for modeling species distribution prediction, as supported by previous studies [110,111,112,113]. In other studies, it has been noted that generative methods such as RF tend to yield improved results with small sample sizes, possibly due to their faster convergence toward their higher asymptotic error when compared to discriminative methods [114].
The performance of all the applied algorithms was effective, as indicated by the mentioned measures, enabling their inclusion in the ensemble modeling. The spatial extent of Acacia decurrens plantation in the study area was 22.44% during 2022. Worku et al. [63] reported that 17.97% of the district was covered by the plantations during 2017. The proportion of plantations was described as 33.9% in one of the watersheds in the district during 2017 [17]. Another study by Wondie and Mekuria [13] stated that 25.6% of the district was covered by forestlands. The difference in the extent between this study and previous studies might be attributed to differences in the satellite data properties, methodology used, and study time.
The ensemble model unveiled the variable importance and response curves in the prediction of Acacia decurrens distribution. The results indicate that CVI, ARVI, and GI are ranked as the first, second, and third variables, respectively, with values of 39.3%, 16%, and 7.1% based on the AUC metric, signifying the relatively strong influence of vegetation indices. These indices played a predominant role in determining the distribution and proliferation of Acacia decurrens. Vegetation indices played a crucial role in identifying tree species, demonstrating the highest relative importance, and can effectively serve as a classification variable to differentiate evergreen trees [115]. The study by Anna [116] also reported similar variable importance, with vegetation indices being the best predictors of tropical evergreen species. This result is consistent with the description of vegetation index variables exerting a more substantial influence than the bioclimatic variables, significantly contributing to defining the distribution range and landscape patterns in the Chelodina longicollis model, with a total contribution of 50.75 compared to 36.94 for the 11 bioclimatic variables [117]. Moreover, Engler et al. [99] also indicated that variables derived from remote sensing are significantly crucial for mapping the spatial distribution of both broadleaved and coniferous tree species at high-resolution data. Elevation, distance to roads, and mSAVI occupy the fourth, fifth, and sixth ranks, respectively, making significant contributions to the species’ distribution. Elevation significantly influences the distribution of plant species [118,119], and in particular, the decision of where and whether to establish plantations depends on environmental factors such as elevation [120]. Further, the findings of Altamirano and Lara [121] also indicate that plantation forests tend to be located in areas characterized by moderate elevation levels and a short distance from roads. Rainfall holds the seventh rank in terms of its relative importance in determining the distribution of this species. The lower ranked predictor variables, specifically, temperature and slope, also play a role in the distribution of Acacia decurrens. However, in relative terms, they are not as influential as other predictor variables in the model, as indicated by the AUC metrics. This may be attributed to the size of the study area in relation to the resolution of the utilized data.
The response curves in Figure 6 show that the probability of Acacia decurrens occurrence generally increases with higher values of vegetation indices. Among the environmental variables, the probability of occurrence increases with higher elevation, keeping other variables constant at their mean value. The probability of occurrence is rare at low-elevation areas of the district because these areas are more suitable for agricultural practices, especially small-scale irrigation activities. This is clearly seen in Figure 7, where the distribution of Acacia decurrens is rare in the Northeastern and Western parts, where small-scale irrigation activities are being carried out along the Guder and Tinbil rivers, respectively. Conversely, the probability of occurrence decreases with an increase in distance from roads and slope (Figure 7). The community cultivates Acacia decurrens to generate income by selling the standing trees or producing charcoal, and road access plays a vital role in facilitating this. Access to roads reduces input costs and in certain situations, leads to higher prices for plantation products [122].
In summary, the findings of this study proved that PS-derived variables and environmental variables integrated with SDM are effective in identifying the distribution of small-scale Acacia decurrens plantation forests. This can be attributed to the high spatial resolution of the PS image, suggesting potential applications of PS imagery in small-scale forestry.

5. Conclusions

The Acacia decurrens plantation holds significant importance in the study area. Over the past few decades, the rapid expansion of these plantations has brought about both environmental and economic implications. Accurately predicting the spatial distribution of Acacia decurrens plantations is essential for concerned authorities to develop effective management policies.
This study successfully employed high-resolution satellite data and advanced techniques to assess the spatial distribution of small-scale Acacia decurrens plantation forests. The thorough evaluation of multicollinearity using the VIF method was pivotal, leading to the exclusion of a substantial portion of variables to address collinearity issues, and enhanced the robustness of the SDM. Among the six algorithms utilized, RF emerged as the standout performer, exhibiting near-perfect agreement with AUC above 0.9 and TSS above 0.8. The ensemble model further identified that significant portions of the study area were covered with Acacia decurrens plantations and emphasized the critical role of vegetation indices (CVI, ARVI, and GI) in determining its distribution and proliferation, underscoring their high relative importance. This study also revealed that the decision to establish plantations is influenced by environmental considerations such as elevation, with areas of moderate elevation and proximity to roads being favorable for plantation activities. The geographical rarity of Acacia decurrens in certain areas was attributed to the prevalence of agricultural practices such as small-scale irrigation.
Ultimately, this research demonstrates the effectiveness of integrating PS imagery-derived variables with environmental factors within SDM, offering valuable insights for identifying and managing small-scale Acacia decurrens plantation forests and providing comprehensive information for efficient decision making in land use planning and forestry management at a local scale.

Author Contributions

Conceptualization, B.A., J.S.-M. and J.R.; methodology, B.A.; software, B.A.; validation, B.A.; formal analysis, B.A.; investigation, B.A.; resources, B.A., J.S.-M. and J.R.; data curation, B.A.; writing—original draft preparation, B.A.; writing—review and editing, B.A., J.S.-M. and J.R.; visualization, B.A.; supervision, J.S.-M. and J.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The data from this are available from corresponding author upon reasonable request.

Acknowledgments

The authors are sincerely thankful to Swansea University for providing PlanetScope images.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

No.NameAbbreviationFormula
1Atmospherically Resistant Vegetation Index ARVI I R − R − y ( R − B ) I R + R − y ( R − B )
2Blue Green Pigment IndexBGI B G
3Blue Normalized Difference Vegetation IndexBNDVI   IR − B   IR   +   B  
4Chlorophyll Vegetation IndexCVI IR     R   G   2
5Difference Vegetation IndexDVI I R − R
6Differenced Vegetation Index MSSDVIMSS 2.4 I R − R
7Enhanced Vegetation IndexEVI 2.5 I R − R ( I R + 6 R − 7.5 B ) + 1
8Enhanced Vegetation Index 2EVI2 2.4 I R − R I R + R + 1
9Green Atmospherically Resistant Vegetation IndexGARI I R − ( G − ( B − R ) ) I R − ( G + ( B − R ) )
10Green-Blue NDVIGBNDVI I R − ( G + B ) I R + ( G + B )
11Greenness IndexGI G R
12Green Leaf IndexGLI 2 G − ( R − B ) 2 G + ( R + B )
13Green NDVIGNDVI   IR − G   IR   +   G  
14Green Optimized SAVIGOSAVI   IR − G   IR   +   G + 0.16
15Green-Red NDVIGRNDVI I R − ( G + R ) I R + ( G + R )
16Green Ratio Vegetation IndexGRVI I R G
17Infrared Percentage Vegetation IndexIPVI I R I R + R 2 ( N D V I + 1 )
18Leaf Area IndexLAI 3.618 E V I − 0.118
19Modified NDVImNDVI I R − R I R + ( R − 2 B )
20Modified Simple RatiomSR I R − B R − B
21Modified SAVImSAVI 2 I R + 1 − ( 2 I R + 1 ) 2 − 8 ( I R − R ) 2
22Normalized Difference Plant Pigment RatioPPR G − B G + B
23Normalized Difference Photosynthetic Vigor RatioPVR G − R G + R
24Normalized Difference 682/553ND682/553 R − G R + G
25Normalized Difference Vegetation IndexNDVI I R − R I R + R
26Red-Blue NDVIRBNDVI I R − ( R + B ) I R + ( R + B )
27Renormalized Difference Vegetation IndexRDVI   I R − R   IR + R  
28Soil Adjusted Vegetation IndexSAVI IR − R     IR   + R + L ( 1 + L )
29Simple RatioSR IR R
30Transformed NDVITNDVI I R − R I R + R + 0.5
31Weighted Difference Vegetation IndexWDVI I R − α R
32Wide Dynamic Range Vegetation IndexWDRVI 0.1 I R − R 0.1 I R + R
where B is blue band, G is green band, R is red band, IR is infrared band, α is 0.2, and L is 0.5.

References

  1. Stephens, S.S.; Wagner, M.R. Forest Plantations and Biodiversity: A Fresh Perspective. J. For. 2007, 105, 307–313. [Google Scholar]
  2. FAO. FAO Global Forest Resources Assessment 2020 Main Report; FAO: Rome, Germany, 2020; ISBN 9789251329740. [Google Scholar]
  3. Carnus, J.M.; Parrotta, J.; Brockerhoff, E.; Arbez, M.; Jactel, H.; Kremer, A.; Lamb, D.; O’Hara, K.; Walters, B. Planted Forests and Biodiversity. J. For. 2006, 104, 65–77. [Google Scholar]
  4. Pawson, S.M.; Brin, A.; Brockerhoff, E.G.; Lamb, D.; Payn, T.W.; Paquette, A.; Parrotta, J.A. Plantation Forests, Climate Change and Biodiversity. Biodivers. Conserv. 2013, 22, 1203–1227. [Google Scholar] [CrossRef]
  5. Paquette, A.; Hawryshyn, J.; Senikas, A.V.; Potvin, C. Enrichment Planting in Secondary Forests: A Promising Clean Development Mechanism to Increase Terrestrial Carbon Sinks. Ecol. Soc. 2009, 14, 31. [Google Scholar] [CrossRef]
  6. Van Der Meer, P.J.; Kanninen, M. Ecosystem Goods and Services from Plantation Forests. In Ecosystem Goods and Services from Plantation Forests; Routledge: Oxfordshire, UK, 2016. [Google Scholar]
  7. More, S.; Karpatne, A.; Wynne, R.H.; Thomas, V.A. Deep Learning for Forest Plantation Mapping in Godavari Districts of Andhra Pradesh, India. Earth Day KDD 2019, 1–5. Available online: https://vtechworks.lib.vt.edu/server/api/core/bitstreams/5da5a7e7-01e3-4546-a062-35f011047556/content (accessed on 8 June 2023).
  8. Brockerhoff, E.G.; Berndt, L.A.; Jactel, H. Role of Exotic Pine Forests in the Conservation of the Critically Endangered New Zealand Ground Beetle Holcaspis Brevicula (Coleoptera: Carabidae). N. Z. J. Ecol. 2005, 29, 37–43. [Google Scholar]
  9. Lemenih, M.; Kassa, H. Re-Greening Ethiopia: History, Challenges and Lessons. Forests 2014, 5, 1896–1909. [Google Scholar] [CrossRef]
  10. Bayle, G.K. Ecological and Social Impacts of Eucalyptus Tree Plantation on the Environment. J. Biodivers. Conserv. Bioresour. Manag. 2019, 5, 93–104. [Google Scholar] [CrossRef]
  11. Mekuria, W.; Wondie, M.; Amare, T.; Wubet, A.; Feyisa, T.; Yitaferu, B. Restoration of Degraded Landscapes for Ecosystem Services in North-Western Ethiopia. Heliyon 2018, 4, e00764. [Google Scholar] [CrossRef]
  12. Abiyu, A.; Teketay, D.; Gratzer, G.; Shete, M. Tree Planting by Smallholder Farmers in the Upper Catchment of Lake Tana Watershed, Northwest Ethiopia. Small Scale For. 2016, 15, 199–212. [Google Scholar] [CrossRef]
  13. Wondie, M.; Mekuria, W. Planting of Acacia Decurrens and Dynamics of Land Cover Change in Fagita Lekoma District in the Northwestern Highlands of Ethiopia. Mt. Res. Dev. 2018, 38, 230–239. [Google Scholar] [CrossRef]
  14. Tadesse, W.; Gezahgne, A.; Tesema, T.; Shibabaw, B. Plantation Forests in Amhara Region: Challenges and Best Measures for Future Improvements. World J. Agric. Res. 2019, 7, 149–157. [Google Scholar]
  15. Nambiar, E.S.; Harwood, C.E.; Kien, N.D. Acacia Plantations in Vietnam: Research and Knowledge Application to Secure a Sustainable Future. South. For. 2015, 77, 1–10. [Google Scholar] [CrossRef]
  16. Alemayehu, B. GIS and Remote Sensing Based Land Use/Land Cover Change Detection and Prediction in Fagita Lekoma Woreda, Awi Zone, Northwestern Ethiopia. Master’s Thesis, Addis Ababa University, Addis Ababa, Ethiopia, 2015; pp. 1–85. [Google Scholar]
  17. Berihun, M.L.; Tsunekawa, A.; Haregeweyn, N.; Meshesha, D.T.; Adgo, E.; Tsubo, M.; Masunaga, T.; Fenta, A.A.; Sultan, D.; Yibeltal, M. Exploring Land Use/Land Cover Changes, Drivers and Their Implications in Contrasting Agro-Ecological Environments of Ethiopia. Land Use Policy 2019, 87, 104052. [Google Scholar] [CrossRef]
  18. Baral, H.; Keenan, R.J.; Fox, J.C.; Stork, N.E.; Kasel, S. Spatial Assessment of Ecosystem Goods and Services in Complex Production Landscapes: A Case Study from South-Eastern Australia. Ecol. Complex. 2013, 13, 35–45. [Google Scholar] [CrossRef]
  19. Chazdon, R.L. Beyond Deforestation: Restoring Forests and Ecosystem Services on Degraded Lands. Science 2008, 320, 1458–1460. [Google Scholar] [CrossRef] [PubMed]
  20. Parrotta, J.A. The Role of Plantation Forests in Rehabilitating Degraded Tropical Ecosystems. Agric. Ecosyst. Environ. 1992, 41, 115–133. [Google Scholar] [CrossRef]
  21. Baillie, B.R.; Neary, D.G. Water Quality in New Zealand’s Planted Forests: A Review. N. Z. J. For. Sci. 2015, 45, 7. [Google Scholar] [CrossRef]
  22. Freer-smith, P. Plantation Forests: Potential and Impacts in Europe New EFI Study. 2019. Available online: https://efi.int/sites/default/files/files/thinkforest/2019/ThinkForest%20-%20Dec%202019_Freer-Smith.pdf (accessed on 12 June 2023).
  23. Akingbogun, A.; Kosoko, O.; Aborisade, D.K. Remote Sensing and GIS Application for Forest Reserve Degredation Prediction and Monitoring. In Proceedings of the FIG Young Surveyors Conference, Rome, Italy, 4–5 May 2012; pp. 1–27. [Google Scholar]
  24. Calders, K.; Jonckheere, I.; Nightingale, J.; Vastaranta, M. Remote Sensing Technology Applications in Forestry and REDD+. Forests 2020, 11, 188. [Google Scholar] [CrossRef]
  25. Lausch, A.; Erasmi, S.; King, D.J.; Magdon, P.; Heurich, M. Understanding Forest Health with Remote Sensing-Part I-A Review of Spectral Traits, Processes and Remote-Sensing Characteristics. Remote Sens. 2016, 8, 1029. [Google Scholar] [CrossRef]
  26. Liu, Q.; Zhang, S.; Zhang, H.; Bai, Y.; Zhang, J. Monitoring Drought Using Composite Drought Indices Based on Remote Sensing. Sci. Total Environ. 2020, 711, 134585. [Google Scholar] [CrossRef]
  27. Shahabi, H.; Shirzadi, A.; Ghaderi, K.; Omidvar, E.; Al-Ansari, N.; Clague, J.J.; Geertsema, M.; Khosravi, K.; Amini, A.; Bahrami, S.; et al. Flood Detection and Susceptibility Mapping Using Sentinel-1 Remote Sensing Data and a Machine Learning Approach: Hybrid Intelligence of Bagging Ensemble Based on K-Nearest Neighbor Classifier. Remote Sens. 2020, 12, 266. [Google Scholar] [CrossRef]
  28. Ahmed, N.; Atzberger, C.; Zewdie, W. Species Distribution Modelling Performance and Its Implication for Sentinel-2-Based Prediction of Invasive Prosopis Juliflora in Lower Awash River Basin, Ethiopia. Ecol. Process 2021, 10, 18. [Google Scholar] [CrossRef]
  29. de Simone, W.; Di Musciano, M.; Di Cecco, V.; Ferella, G.; Frattaroli, A.R. The Potentiality of Sentinel-2 to Assess the Effect of Fire Events on Mediterranean Mountain Vegetation. Plant Sociol. 2020, 57, 11–22. [Google Scholar] [CrossRef]
  30. Meng, Y.; Wei, C.; Guo, Y.; Tang, Z. A Planted Forest Mapping Method Based on Long-Term Change Trend Features Derived from Dense Landsat Time Series in an Ecological Restoration Region. Remote Sens. 2022, 14, 961. [Google Scholar] [CrossRef]
  31. Torbick, N.; Ledoux, L.; Salas, W.; Zhao, M. Regional Mapping of Plantation Extent Using Multisensor Imagery. Remote Sens. 2016, 8, 236. [Google Scholar] [CrossRef]
  32. Chi, H.; Sun, G.; Huang, J.; Guo, Z.; Ni, W.; Fu, A. National Forest Aboveground Biomass Mapping from ICESat/GLAS Data and MODIS Imagery in China. Remote Sens. 2015, 7, 5534–5564. [Google Scholar] [CrossRef]
  33. Fagan, M.E.; DeFries, R.S.; Sesnie, S.E.; Arroyo-Mora, J.P.; Soto, C.; Singh, A.; Townsend, P.A.; Chazdon, R.L. Mapping Species Composition of Forests and Tree Plantations in Northeastern Costa Rica with an Integration of Hyperspectral and Multitemporal Landsat Imagery. Remote Sens. 2015, 7, 5660–5696. [Google Scholar] [CrossRef]
  34. Frans, V.F.; Augé, A.A.; Fyfe, J.; Zhang, Y.; McNally, N.; Edelhoff, H.; Balkenhol, N.; Engler, J.O. Integrated SDM Database: Enhancing the Relevance and Utility of Species Distribution Models in Conservation Management. Methods Ecol. Evol. 2022, 13, 243–261. [Google Scholar] [CrossRef]
  35. Adem Esmail, B.; Geneletti, D. Multi-Criteria Decision Analysis for Nature Conservation: A Review of 20 Years of Applications. Methods Ecol. Evol. 2018, 9, 42–53. [Google Scholar] [CrossRef]
  36. Smeraldo, S.; Di Febbraro, M.; Ćirović, D.; Bosso, L.; Trbojević, I.; Russo, D. Species Distribution Models as a Tool to Predict Range Expansion after Reintroduction: A Case Study on Eurasian Beavers (Castor Fiber). J. Nat. Conserv. 2017, 37, 12–20. [Google Scholar] [CrossRef]
  37. Lorena, A.C.; Jacintho, L.F.O.; Siqueira, M.F.; Giovanni, R.D.; Lohmann, L.G.; De Carvalho, A.C.P.L.F.; Yamamoto, M. Comparing Machine Learning Classifiers in Potential Distribution Modelling. Expert. Syst. Appl. 2011, 38, 5268–5275. [Google Scholar] [CrossRef]
  38. Ortega-Huerta, M.A.; Peterson, A.T. Modelling Spatial Patterns of Biodiversity for Conservation Prioritization in North-Eastern Mexico. Divers. Distrib. 2004, 10, 39–54. [Google Scholar] [CrossRef]
  39. Angelieri, C.C.S.; Adams-Hosking, C.; Paschoaletto, K.M.; De Barros Ferraz, M.; De Souza, M.P.; McAlpine, C.A. Using Species Distribution Models to Predict Potential Landscape Restoration Effects on Puma Conservation. PLoS ONE 2016, 11, e0145232. [Google Scholar] [CrossRef]
  40. Schwartz, M.W. Using Niche Models with Climate Projections to Inform Conservation Management Decisions. Biol. Conserv. 2012, 155, 149–156. [Google Scholar] [CrossRef]
  41. Peterson, A.T.; Ortega-Huerta, M.A.; Bartley, J.; Sánchez-Cordero, V.; Soberón, J.; Buddemeler, R.H.; Stockwell, D.R.B. Future Projections for Mexican Faunas under Global Climate Change Scenarios. Nature 2002, 416, 626–629. [Google Scholar] [CrossRef]
  42. Rodríguez-Merino, A.; Fernández-Zamudio, R.; García-Murillo, P. An Invasion Risk Map for Non-Native Aquatic Macrophytes of the Iberian Peninsula. An. Del Jard. Bot. De Madr. 2017, 74, e055. [Google Scholar] [CrossRef]
  43. Cheung, W.W.L.; Rondinini, C.; Avtar, R.; van den Belt, M.; Hickler, T.; Metzger, J.P.; Scharlemann, J.P.W.; Velez-Liendo, X.; Yue, T.X. The Methodological Assessment Report on Scenarios and Models of Biodiversity and Ecosystem Services: Summary for Policymakers; IPBES: Bonn, Germany, 2016; ISBN 9789280735697. [Google Scholar]
  44. Elith, J.; Leathwick, J.R. Species Distribution Models: Ecological Explanation and Prediction across Space and Time. Annu. Rev. Ecol. Evol. Syst. 2009, 40, 677–697. [Google Scholar] [CrossRef]
  45. Guisan, A.; Thuiller, W.; Zimmermann, N.E. Habitat Suitability and Distribution Models; Cambridge University Press: Cambridge, UK, 2017. [Google Scholar]
  46. Václavík, T.; Meentemeyer, R.K. Invasive Species Distribution Modeling (ISDM): Are Absence Data and Dispersal Constraints Needed to Predict Actual Distributions? Ecol. Modell. 2009, 220, 3248–3258. [Google Scholar] [CrossRef]
  47. Marmion, M.; Parviainen, M.; Luoto, M.; Heikkinen, R.K.; Thuiller, W. Evaluation of Consensus Methods in Predictive Species Distribution Modelling. Divers. Distrib. 2009, 15, 59–69. [Google Scholar] [CrossRef]
  48. Amiri, M.; Tarkesh, M.; Shafiezadeh, M. Modelling the Biological Invasion of Prosopis Juliflora Using Geostatistical-Based Bioclimatic Variables under Climate Change in Arid Zones of Southwestern Iran. J. Arid. Land. 2022, 14, 203–224. [Google Scholar] [CrossRef]
  49. Arogoundade, A.M.; Odindi, J.; Mutanga, O. Modelling Parthenium Hysterophorus Invasion in KwaZulu-Natal Province Using Remotely Sensed Data and Environmental Variables. Geocarto Int. 2020, 35, 1450–1465. [Google Scholar] [CrossRef]
  50. Lesiv, M.; Laso Bayas, J.C.; See, L.; Duerauer, M.; Dahlia, D.; Durando, N.; Hazarika, R.; Kumar Sahariah, P.; Vakolyuk, M.; Blyshchyk, V.; et al. Estimating the Global Distribution of Field Size Using Crowdsourcing. Glob. Chang. Biol. 2019, 25, 174–186. [Google Scholar] [CrossRef]
  51. Rufin, P.; Bey, A.; Picoli, M.; Meyfroidt, P. Large-Area Mapping of Active Cropland and Short-Term Fallows in Smallholder Landscapes Using PlanetScope Data. Int. J. Appl. Earth Obs. Geoinf. 2022, 112, 102937. [Google Scholar] [CrossRef]
  52. Roy, D.P.; Huang, H.; Houborg, R.; Martins, V.S. A Global Analysis of the Temporal Availability of PlanetScope High Spatial Resolution Multi-Spectral Imagery. Remote Sens. Environ. 2021, 264, 112586. [Google Scholar] [CrossRef]
  53. Cui, B.; Huang, W.; Ye, H.; Chen, Q. The Suitability of PlanetScope Imagery for Mapping Rubber Plantations. Remote Sens. 2022, 14, 1061. [Google Scholar] [CrossRef]
  54. Csillik, O.; Kumar, P.; Asner, G.P. Challenges in Estimating Tropical Forest Canopy Height from Planet Dove Imagery. Remote Sens. 2020, 12, 1160. [Google Scholar] [CrossRef]
  55. Gargiulo, J.; Clark, C.; Lyons, N.; de Veyrac, G.; Beale, P.; Garcia, S. Spatial and Temporal Pasture Biomass Estimation Integrating Electronic Plate Meter, Planet Cubesats and Sentinel-2 Satellite Data. Remote Sens. 2020, 12, 3222. [Google Scholar] [CrossRef]
  56. Kimm, H.; Guan, K.; Jiang, C.; Peng, B.; Gentry, L.F.; Wilkin, S.C.; Wang, S.; Cai, Y.; Bernacchi, C.J.; Peng, J.; et al. Deriving High-Spatiotemporal-Resolution Leaf Area Index for Agroecosystems in the U.S. Corn Belt Using Planet Labs CubeSat and STAIR Fusion Data. Remote Sens. Environ. 2020, 239, 111615. [Google Scholar] [CrossRef]
  57. Skakun, S.; Kalecinski, N.I.; Brown, M.G.L.; Johnson, D.M.; Vermote, E.F.; Roger, J.C.; Franch, B. Assessing Within-Field Corn and Soybean Yield Variability from Worldview-3, Planet, Sentinel-2, and Landsat 8 Satellite Imagery. Remote Sens. 2021, 13, 872. [Google Scholar] [CrossRef]
  58. Jin, Y.; Guo, J.; Ye, H.; Zhao, J.; Huang, W.; Cui, B. Extraction of Arecanut Planting Distribution Based on the Feature Space Optimization of Planetscope Imagery. Agriculture 2021, 11, 371. [Google Scholar] [CrossRef]
  59. Kaky, E.; Nolan, V.; Alatawi, A.; Gilbert, F. A Comparison between Ensemble and MaxEnt Species Distribution Modelling Approaches for Conservation: A Case Study with Egyptian Medicinal Plants. Ecol. Inform. 2020, 60, 101150. [Google Scholar] [CrossRef]
  60. Parrotta, J.A.; Turnbull, J.W.; Jones, N. Introduction Catalyzing Native Forest Regeneration on Degraded Tropical Lands. For. Ecol. Manag. 1997, 99, 1–7. [Google Scholar] [CrossRef]
  61. Barrio-Anta, M.; Castedo-Dorado, F.; Cámara-Obregón, A.; López-Sánchez, C.A. Integrating Species Distribution Models at Forest Planning Level to Develop Indicators for Fast-Growing Plantations. A Case Study of Eucalyptus Globulus Labill. in Galicia (NW Spain). For. Ecol. Manage 2021, 491, 119200. [Google Scholar] [CrossRef]
  62. Teshome, T.; Wondimu, A. Best Practices on Development and Utilatation of Acacia Decurrens in Fagta Lekoma District, Awi Zone, Amhara Region. 2019, 37. Available online: https://www.epa.gov.et/images/PDF/ForestManuals/BEST%20PRACTICES%20ON%20DEVELOPMENT%20AND%20UTILIZATION%20OF%20ACACIA%20DECURRENS.pdf (accessed on 7 January 2022).
  63. Worku, T.; Mekonnen, M.; Yitaferu, B.; Cerdà, A. Conversion of Crop Land Use to Plantation Land Use, Northwest Ethiopia. Trees For. People 2021, 3, 100044. [Google Scholar] [CrossRef]
  64. Keneni, Y.G.; Senbeta, A.F.; Sime, G. Role of small-scale trees plantation and farmers’ attitude and skill toward propagation of indigenous and exotic trees: The case of Sidama, Ethiopia. Afr. J. Food Agric. Nutr. Dev. 2021, 21, 18804–18823. [Google Scholar] [CrossRef]
  65. Yibeltal, M.; Tsunekawa, A.; Haregeweyn, N.; Adgo, E.; Meshesha, D.T.; Aklog, D.; Masunaga, T.; Tsubo, M.; Billi, P.; Vanmaercke, M.; et al. Analysis of Long-Term Gully Dynamics in Different Agro-Ecology Settings. Catena 2019, 179, 160–174. [Google Scholar] [CrossRef]
  66. Bazie, Z.; Feyssa, S.; Amare, T. Effects of Acacia Decurrens Willd. Tree-Based Farming System on Soil Quality in Guder Watershed, North Western Highlands of Ethiopia. Cogent Food Agric. 2020, 6, 1743622. [Google Scholar] [CrossRef]
  67. Chanie, Y.; Abewa, A. Expansion of Acacia Decurrens Plantation on the Acidic Highlands of Awi Zone, Ethiopia, and Its Socio-Economic Benefits. Cogent Food Agric. 2021, 7, 1917150. [Google Scholar] [CrossRef]
  68. Alemayehu, B.; Suarez-Minguez, J.; Rosette, J.; Khan, S.A. Vegetation Trend Detection Using Time Series Satellite Data as Ecosystem Condition Indicators for Analysis in the Northwestern Highlands of Ethiopia. Remote Sens. 2023, 15, 5032. [Google Scholar] [CrossRef]
  69. Harrison, T.N. Introduction to Planet’s New 8-Band Data and Access via NASA’s Commercial SmallSat Data Acquisition (CSDA) Program. 2022. Available online: https://www.earthdata.nasa.gov/s3fs-public/2022-06/PlanetCSDA8-Band_Data_0.pdf (accessed on 23 August 2023).
  70. Lottering, R.; Mutanga, O.; Peerbhay, K. Detecting and Mapping Levels of Gonipterus Scutellatus-Induced Vegetation Defoliation and Leaf Area Index Using Spatially Optimized Vegetation Indices. Geocarto Int. 2018, 33, 277–292. [Google Scholar] [CrossRef]
  71. Dong, J.; Zhou, C.; Liang, W.; Lu, X. Determination Factors for the Spatial Distribution of Forest Cover: A Case Study of China’s Fujian Province. Forests 2022, 13, 2070. [Google Scholar] [CrossRef]
  72. Shunlin Liang, J.W. Forest Cover Changes. In Advanced Remote Sensing; Elsevier: Amsterdam, The Netherlands, 2020; pp. 915–952. ISBN 9780128158265. [Google Scholar]
  73. Singh, S. Understanding the Role of Slope Aspect in Shaping the Vegetation Attributes and Soil Properties in Montane Ecosystems. Trop. Ecol. 2018, 59, 417–430. [Google Scholar]
  74. Xie, C.; Li, M.; Jim, C.Y.; Liu, D. Environmental Factors Driving the Spatial Distribution Pattern of Venerable Trees in Sichuan Province, China. Plants 2022, 11, 3581. [Google Scholar] [CrossRef]
  75. Agujetas Perez, M.; Tegebu, F.N.; Steenbergen, F. van Roadside Planting in Ethiopia: Turning a Problem into an Opportunity. Sustain. Environ. 2016, 1, 98. [Google Scholar] [CrossRef]
  76. Stage, A.R.; Salas, C. Interactions of Elevation, Aspect, and Slope in Models of Forest Species Composition and Productivity. For. Sci. 2007, 53, 486–492. [Google Scholar]
  77. Zhang, X.N.; Yang, X.D.; Li, Y.; He, X.M.; Lv, G.H.; Yang, J.J. Influence of Edaphic Factors on Plant Distribution and Diversity in the Arid Area of Xinjiang, Northwest China. Arid Land Res. Manag. 2018, 32, 38–56. [Google Scholar] [CrossRef]
  78. Naimi, B.; Hamm, N.A.S.; Groen, T.A.; Skidmore, A.K.; Toxopeus, A.G. Where Is Positional Uncertainty a Problem for Species Distribution Modelling? Ecography 2014, 37, 191–203. [Google Scholar] [CrossRef]
  79. Adeyemo, S.M.; Granger, J.J. Habitat Suitability Model and Range Shift Analysis for American Chestnut (Castanea Dentata) in the United States. Trees For. People 2023, 11, 100360. [Google Scholar] [CrossRef]
  80. Naimi, B.; Araújo, M.B. Sdm: A Reproducible and Extensible R Platform for Species Distribution Modelling. Ecography 2016, 39, 368–375. [Google Scholar] [CrossRef]
  81. Cheek, P.J.; McCullagh, P.; Nelder, J.A. Generalized Linear Models, 2nd ed.; Applied Statistics; Routledge: Oxfordshire, UK, 1990; Volume 39, p. 385. [Google Scholar]
  82. Friedman, J.H. Multivariate Adaptive Regression Splines. Ann. Stat. 2007, 19, 1–67. [Google Scholar] [CrossRef]
  83. Friedman, J.H. Greedy Function Approximation: A Gradient Boosting Machine. Ann. Stat. 2001, 29, 1189–1232. [Google Scholar] [CrossRef]
  84. Breiman, L. Random Forests; University of California: Oakland, CA, USA, 2001; Volume 7, pp. 1–33. [Google Scholar] [CrossRef]
  85. Vapnik, V.N. The Nature of Statistical Learning Theory; Springer: New York, NY, USA, 2000; ISBN 978-1-4419-3160-3. [Google Scholar]
  86. Breiman, L.; Friedman, J.H.; Olshen, R.A.; Stone, C.J. Classification and Regression Trees; 1984; ISBN 9780412048418. Available online: https://rafalab.dfci.harvard.edu/pages/649/section-11.pdf (accessed on 19 May 2023).
  87. Araújo, M.B.; New, M. Ensemble Forecasting of Species Distributions. Trends Ecol. Evol. 2007, 22, 42–47. [Google Scholar] [CrossRef]
  88. Pecchi, M.; Marchi, M.; Burton, V.; Giannetti, F.; Moriondo, M.; Bernetti, I.; Bindi, M.; Chirici, G. Species Distribution Modelling to Support Forest Management. A Literature Review. Ecol. Modell. 2019, 411, 108817. [Google Scholar] [CrossRef]
  89. Thuiller, W.; Lafourcade, B.; Engler, R.; Araújo, M.B. BIOMOD—A Platform for Ensemble Forecasting of Species Distributions. Ecography 2009, 32, 369–373. [Google Scholar] [CrossRef]
  90. Araújo, M.B.; Pearson, R.G.; Thuiller, W.; Erhard, M. Validation of Species-Climate Impact Models under Climate Change. Glob. Chang. Biol. 2005, 11, 1504–1513. [Google Scholar] [CrossRef]
  91. Fielding, A.H.; Bell, J.F. A Review of Methods for the Assessment of Prediction Errors in Conservation Presence/Absence Models. Environ. Conserv. 1997, 24, 38–49. [Google Scholar] [CrossRef]
  92. Allouche, O.; Tsoar, A.; Kadmon, R. Assessing the Accuracy of Species Distribution Models: Prevalence, Kappa and the True Skill Statistic (TSS). J. Appl. Ecol. 2006, 43, 1223–1232. [Google Scholar] [CrossRef]
  93. Jiménez-Valverde, A. Insights into the Area under the Receiver Operating Characteristic Curve (AUC) as a Discrimination Measure in Species Distribution Modelling. Glob. Ecol. Biogeogr. 2012, 21, 498–507. [Google Scholar] [CrossRef]
  94. Zhang, L.; Liu, S.; Sun, P.; Wang, T.; Wang, G.; Zhang, X.; Wang, L. Consensus Forecasting of Species Distributions: The Effects of Niche Model Performance and Niche Properties. PLoS ONE 2015, 10, e0120056. [Google Scholar] [CrossRef]
  95. Duque-Lazo, J.; van Gils, H.; Groen, T.A.; Navarro-Cerrillo, R.M. Transferability of Species Distribution Models: The Case of Phytophthora Cinnamomi in Southwest Spain and Southwest Australia. Ecol. Modell. 2016, 320, 62–70. [Google Scholar] [CrossRef]
  96. De Marco, P.; Nóbrega, C.C. Evaluating Collinearity Effects on Species Distribution Models: An Approach Based on Virtual Species Simulation. PLoS ONE 2018, 13, e0202403. [Google Scholar] [CrossRef]
  97. Chatterjee, S.; Hadi, A.S. Regression-Analysis-by-Example; Wiley: Hoboken, NJ, USA, 2012; pp. 1–421. [Google Scholar]
  98. Dormann, C.F.; Elith, J.; Bacher, S.; Buchmann, C.; Carl, G.; Carré, G.; Marquéz, J.R.G.; Gruber, B.; Lafourcade, B.; Leitão, P.J.; et al. Collinearity: A Review of Methods to Deal with It and a Simulation Study Evaluating Their Performance. Ecography 2013, 36, 27–46. [Google Scholar] [CrossRef]
  99. Engler, R.; Waser, L.T.; Zimmermann, N.E.; Schaub, M.; Berdos, S.; Ginzler, C.; Psomas, A. Combining Ensemble Modeling and Remote Sensing for Mapping Individual Tree Species at High Spatial Resolution. For. Ecol. Manage 2013, 310, 64–73. [Google Scholar] [CrossRef]
  100. Li, J. Assessing the Accuracy of Predictive Models for Numerical Data: Not r nor R2, Why Not? Then What? PLoS ONE 2017, 12, e0183250. [Google Scholar] [CrossRef]
  101. Maxwell, A.E.; Sharma, M.; Donaldson, K.A. Explainable Boosting Machines for Slope Failure Spatial Predictive Modeling. Remote Sens. 2021, 13, 4991. [Google Scholar] [CrossRef]
  102. González-Irusta, J.M.; González-Porto, M.; Sarralde, R.; Arrese, B.; Almón, B.; Martín-Sosa, P. Comparing Species Distribution Models: A Case Study of Four Deep Sea Urchin Species. Hydrobiologia 2015, 745, 43–57. [Google Scholar] [CrossRef]
  103. Qiao, H.; Soberón, J.; Peterson, A.T. No Silver Bullets in Correlative Ecological Niche Modelling: Insights from Testing among Many Potential Algorithms for Niche Estimation. Methods Ecol. Evol. 2015, 6, 1126–1136. [Google Scholar] [CrossRef]
  104. Mugo, R.; Saitoh, S.-I. Ensemble Modelling of Skipjack Tuna (Katsuwonus pelamis) Habitats in the Western North Pacific Using Satellite Remotely Sensed Data; a Comparative Analysis Using Machine-Learning Models. Remote Sens. 2020, 12, 2591. [Google Scholar] [CrossRef]
  105. Montoya-Jiménez, J.C.; Valdez-Lazalde, J.R.; Ángeles-Perez, G.; de los SantosPosadas, H.M.; Cruz-Cárdenas, G. Predictive Capacity of Nine Algorithms and an Ensemble Model to Determine the Geographic Distribution of Tree Species. IForest 2022, 15, 363–371. [Google Scholar] [CrossRef]
  106. Talukdar, S.; Singha, P.; Mahato, S.; Shahfahad; Pal, S.; Liou, Y.A.; Rahman, A. Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations-A Review. Remote Sens. 2020, 12, 1135. [Google Scholar] [CrossRef]
  107. Rodriguez-Galiano, V.F.; Ghimire, B.; Rogan, J.; Chica-Olmo, M.; Rigol-Sanchez, J.P. An Assessment of the Effectiveness of a Random Forest Classifier for Land-Cover Classification. ISPRS J. Photogramm. Remote Sens. 2012, 67, 93–104. [Google Scholar] [CrossRef]
  108. Hassan, C.A.; Khan, M.S.; Shah, M.A. Comparison of Machine Learning Algorithms in Data Classification. In Proceedings of the ICAC 2018—2018 24th IEEE International Conference on Automation and Computing: Improving Productivity through Automation and Computing, Newcastle Upon Tyne, UK, 6–7 September 2018; Institute of Electrical and Electronics Engineers Inc.: Piscataway, NJ, USA, 2018. [Google Scholar]
  109. Elith, J.; Leathwick, J.R.; Hastie, T. A Working Guide to Boosted Regression Trees. J. Anim. Ecol. 2008, 77, 802–813. [Google Scholar] [CrossRef]
  110. Cheng, L.; Lek, S.; Lek-Ang, S.; Li, Z. Predicting Fish Assemblages and Diversity in Shallow Lakes in the Yangtze River Basin. Limnologica 2012, 42, 127–136. [Google Scholar] [CrossRef]
  111. He, Y.; Wang, J.; Lek-Ang, S.; Lek, S. Predicting Assemblages and Species Richness of Endemic Fish in the Upper Yangtze River. Sci. Total Environ. 2010, 408, 4211–4220. [Google Scholar] [CrossRef]
  112. Grenouillet, G.; Buisson, L.; Casajus, N.; Lek, S. Ensemble Modelling of Species Distribution: The Effects of Geographical and Environmental Ranges. Ecography 2011, 34, 9–17. [Google Scholar] [CrossRef]
  113. Makhkamov, T.K.; Brundu, G.; Jabborov, A.M.; Gaziev, A.D. Predicting the Potential Distribution of Ranunculus Sardous (Ranunculaceae), a New Alien Species in the Flora of Uzbekistan and Central Asia. Bioinvasions Rec. 2023, 12, 63–77. [Google Scholar] [CrossRef]
  114. Aguirre-Gutiérrez, J.; Carvalheiro, L.G.; Polce, C.; van Loon, E.E.; Raes, N.; Reemer, M.; Biesmeijer, J.C. Fit-for-Purpose: Species Distribution Model Performance Depends on Evaluation Criteria—Dutch Hoverflies as a Case Study. PLoS ONE 2013, 8, e63708. [Google Scholar] [CrossRef]
  115. Yuan, J.; Wu, Z.; Li, S.; Kang, P.; Zhu, S. Multi-Feature-Based Identification of Subtropical Evergreen Tree Species Using Gaofen-2 Imagery and Algorithm Comparison. Forests 2023, 14, 292. [Google Scholar] [CrossRef]
  116. Anna, C. Potential of Multi-Temporal Remote Sensing Data for Modeling Tree Species Distributions and Species Richness in Mexico; Institute for Geography and Geology, Julius Maximilian University of Würzburg: Wuerzburg, Germany, 2012. [Google Scholar]
  117. Wen, L.; Saintilan, N.; Yang, X.; Hunter, S.; Mawer, D. MODIS NDVI Based Metrics Improve Habitat Suitability Modelling in Fragmented Patchy Floodplains. Remote Sens. Appl. 2015, 1, 85–97. [Google Scholar] [CrossRef]
  118. Nogués-Bravo, D.; Araújo, M.B.; Romdal, T.; Rahbek, C. Scale Effects and Human Impact on the Elevational Species Richness Gradients. Nature 2008, 453, 216–219. [Google Scholar] [CrossRef]
  119. Muche, M.; Molla, E.; Rewald, B.; Tsegay, B.A. Diversity and Composition of Farm Plantation Tree/Shrub Species along Altitudinal Gradients in North-Eastern Ethiopia: Implication for Conservation. Heliyon 2022, 8, e09048. [Google Scholar] [CrossRef] [PubMed]
  120. Thaiutsa, B.C. Commercial Plantation in Thailand: A Case Study of the Forest Industry Organization. In Proceedings of the FORTROP II: Tropical Forestry Change in a Changing World, Bangkok, Thailand, 17–20 November 2008. [Google Scholar]
  121. Altamirano, A.; Lara, A. Deforestación En Ecosistemas Templados de La Precordillera Andina Del Centro-Sur de Chile. Bosque 2010, 31, 53–64. [Google Scholar] [CrossRef]
  122. Kaczan, D.J. Can Roads Contribute to Forest Transitions? Duke University: Durham, NC, USA, 2017. [Google Scholar]
Forests 15 00277 g001
Figure 1. Map of Fagita Lekoma District.
Figure 1. Map of Fagita Lekoma District.
Forests 15 00277 g001
Forests 15 00277 g002
Figure 2. An overview of the methodological framework applied in the predictive modeling of Acacia decurrens distribution.
Figure 2. An overview of the methodological framework applied in the predictive modeling of Acacia decurrens distribution.
Forests 15 00277 g002
Forests 15 00277 g003
Figure 3. ROC curves generated using ten-fold bootstrap replication for the applied algorithms to identify the distribution of Acacia decurrens. The vertical axis represents sensitivity (true positive rate) and the horizontal axis represents 1-specificity (false positive rate), depicting the proportions of correctly and incorrectly classified samples. The red and blue smoothed curves stand for the mean AUC when using training and testing data, respectively.
Figure 3. ROC curves generated using ten-fold bootstrap replication for the applied algorithms to identify the distribution of Acacia decurrens. The vertical axis represents sensitivity (true positive rate) and the horizontal axis represents 1-specificity (false positive rate), depicting the proportions of correctly and incorrectly classified samples. The red and blue smoothed curves stand for the mean AUC when using training and testing data, respectively.
Forests 15 00277 g003
Forests 15 00277 g004
Figure 4. The spatial distribution of Acacia decurrens from the ensemble model.
Figure 4. The spatial distribution of Acacia decurrens from the ensemble model.
Forests 15 00277 g004
Forests 15 00277 g005
Figure 5. Relative importance of variables for Acacia decurrens prediction.
Figure 5. Relative importance of variables for Acacia decurrens prediction.
Forests 15 00277 g005
Forests 15 00277 g006
Figure 6. Vegetation index response curve for Acacia decurrens presence: blue-black curves show the mean response and gray shades are ±SD calculated through 10 replicate runs in the ensemble model.
Figure 6. Vegetation index response curve for Acacia decurrens presence: blue-black curves show the mean response and gray shades are ±SD calculated through 10 replicate runs in the ensemble model.
Forests 15 00277 g006
Forests 15 00277 g007
Figure 7. Environmental variable response curve for Acacia decurrens presence: blue-black curves show the mean response and gray shades are ±SD calculated through 10 replicate runs in the ensemble model.
Figure 7. Environmental variable response curve for Acacia decurrens presence: blue-black curves show the mean response and gray shades are ±SD calculated through 10 replicate runs in the ensemble model.
Forests 15 00277 g007
Table 1. VIFs of the predictive variables after multicollinearity test with a cut-off threshold of 10.
Table 1. VIFs of the predictive variables after multicollinearity test with a cut-off threshold of 10.
No.VariablesVIF
1ARVI7.414995
2Aspect1.190828
3CVI6.169403
4Elevation3.323751
5GI3.797256
6mSAVI3.771155
7Rainfall3.080133
8Road1.192230
9Slope1.220056
10Soil type1.724052
11Temperature5.638469
Table 2. Performance of SDM algorithms for Acacia decurrens prediction.
Table 2. Performance of SDM algorithms for Acacia decurrens prediction.
AlgorithmAUCSensitivitySpecificityTSS
GLM0.840.810.830.64
MARS0.850.820.830.65
BRT0.890.830.870.7
RF0.930.90.920.82
SVM0.890.820.890.71
CART0.840.80.850.65
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Alemayehu, B.; Suarez-Minguez, J.; Rosette, J. Modeling the Spatial Distribution of Acacia decurrens Plantation Forests Using PlanetScope Images and Environmental Variables in the Northwestern Highlands of Ethiopia. Forests 2024, 15, 277. https://doi.org/10.3390/f15020277

AMA Style

Alemayehu B, Suarez-Minguez J, Rosette J. Modeling the Spatial Distribution of Acacia decurrens Plantation Forests Using PlanetScope Images and Environmental Variables in the Northwestern Highlands of Ethiopia. Forests. 2024; 15(2):277. https://doi.org/10.3390/f15020277

Chicago/Turabian Style

Alemayehu, Bireda, Juan Suarez-Minguez, and Jacqueline Rosette. 2024. "Modeling the Spatial Distribution of Acacia decurrens Plantation Forests Using PlanetScope Images and Environmental Variables in the Northwestern Highlands of Ethiopia" Forests 15, no. 2: 277. https://doi.org/10.3390/f15020277

APA Style

Alemayehu, B., Suarez-Minguez, J., & Rosette, J. (2024). Modeling the Spatial Distribution of Acacia decurrens Plantation Forests Using PlanetScope Images and Environmental Variables in the Northwestern Highlands of Ethiopia. Forests, 15(2), 277. https://doi.org/10.3390/f15020277

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop