Next Article in Journal
Microgrid Protection Using Magneto-Resistive Sensors and Superimposed Reactive Energy
Previous Article in Journal
Numerical Analysis of an Aerofoil Fin Integrated Double Pass Solar Air Heater for Thermal Performance Enhancement
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 1
10.3390/su15010600
sustainability-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 Impact of Climate Change on Sustainable Production of Two Legumes Important Economically and for Food Security: Mungbeans and Cowpeas in Ethiopia

by
Birhanu Kagnew
1,2,*,
Awol Assefa
2 and
Asfaw Degu
2
1
Department of Biology, College of Natural and Computational Science, Hawassa University, Hawasssa P.O. Box 05, Ethiopia
2
Department of Plant Biology and Biodiversity Management, College of Natural and Computational Sciences, Addis Ababa University, Addis Ababa P.O. Box 1176, Ethiopia
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(1), 600; https://doi.org/10.3390/su15010600
Submission received: 1 December 2022 / Revised: 20 December 2022 / Accepted: 22 December 2022 / Published: 29 December 2022
Browse Figures
Review Reports Versions Notes

Abstract

:
Climate change is one of the most serious threats to global crops production at present and it will continue to be the largest threat in the future worldwide. Knowing how climate change affects crop productivity might help sustainability and crop improvement efforts. Under existing and projected climate change scenarios (2050s and 2070s in Ethiopia), the effect of global warming on the distribution of V. radiata and V. unguiculata was investigated. MaxEnt models were used to predict the current and future distribution pattern changes of these crops in Ethiopia using different climate change scenarios (i.e., lowest (RCP 2.6), moderate (RCP 4.5), and extreme (RCP 8.5)) for the years 2050s and 2070s. The study includes 81 and 68 occurrence points for V. radiata and V. unguiculata, respectively, along with 22 environmental variables. The suitability maps indicate that the Beneshangul Gumuz, Oromia, Amhara, SNNPR, and Tigray regions are the major Ethiopian regions with the potential to produce V. radiata, while Amhara, Gambella, Oromia, SNNPR, and Tigray are suitable for producing V. unguiculata. The model prediction for V. radiata habitat ranges distribution in Ethiopia indicated that 1.69%, 4.27%, 11.25% and 82.79% are estimated to be highly suitable, moderately suitable, less suitable, and unsuitable, respectively. On the other hand, the distribution of V. unguiculata is predicted to have 1.27%, 3.07%, 5.22%, and 90.44% habitat ranges that are highly suitable, moderately suitable, less suitable, and unsuitable, respectively, under the current climate change scenario by the year (2050s and 2070s) in Ethiopia. Among the environmental variables, precipitation of the wettest quarter (Bio16), solar radiation index (SRI), temperature seasonality (Bio4), and precipitation seasonality (Bio15) are discovered to be the most effective factors for defining habitat suitability for V. radiata, while precipitation of the wettest quarter (Bio16), temperature annual range (Bio7) and precipitation of the driest quarter (Bio17) found to be better habitat suitability indicator for V. unguiculata in Ethiopia. The result indicates that these variables were more relevant in predicting suitable habitat for these crops in Ethiopia. A future projection predicts that the suitable distribution region will become increasingly fragmented. In general, the study provides a scientific basis of suitable agro-ecological habitat for V. radiata and V. unguiculata for long-term crop management and production improvement in Ethiopia. Therefore, projections of current and future climate change impacts on such crops are vital to reduce the risk of crop failure and to identify the potential productive areas in the country.

1. Introduction

Food insecurity is a common occurrence in Ethiopia’s dry lands as a result of climatic unpredictability, climate change, population pressure, and subsistence agricultural techniques have all which increased the danger of food insecurity. Ethiopia has experienced climate-related disasters in the past and regularly mentioned several times as a country that is extremely vulnerable to climate change and fluctuation [1]. Agriculture, which uses a “low input and low output” subsistence production strategy, is the mainstay of the Ethiopian economy. Smallholder rain-fed farming accounts for 95% of the nation’s yearly crop production and cropping area [1]. Future climate change could have an impact on agricultural productivity and projected to diminish crop yield in developing countries worldwide [2,3]. Over the last few decades, Ethiopia has been frequently affected by droughts and climate extremes, resulting in severe food shortages in the country [4,5,6]. Especially, food insecure farmers are particularly vulnerable to intermittent rainfall and rising temperatures across the country [7]. Consequently, either negatively or positively, the climate change could affect the Ethiopian current or future agricultural production in which millions of the population depends on it [6].
Legumes are important crops for Ethiopian smallholder agriculture, providing an economic advantage and source of protein in Ethiopia [8]. Legume crops (mungbean and cowpea) productions are negligible compared to cereals crop production in Ethiopia. The main reason could be climate change, which affecting pulse crop production in a variety of ways. Therefore, there is an urgent need for modeling or mapping of these crops’ agricultural potential areas in Ethiopia. Crop models have been used in studies to examine the links between crop productivity and environmental conditions based on weather, crop and management methods [9]. Species distribution modeling (SDM) integrates bioclimatic environmental variables from known distribution areas to determine a species’ environmental requirements for survival, and then forecasts its distribution in future [10,11]. BIOCLIM, DOMAIN, GARP and MaxEnt are among algorithms which have been created for modeling species distribution across geographic areas [9,12,13,14]. Even though their performance varies significantly, majority of these algorithms are user-friendly [14]. MaxEnt algorithms predict the existing and projected agricultural potential spaces based on the statistical relationship between species distribution and environmental variables. MaxEnt model only consider environmental variables that control species distribution and ignore factors such as associations and extensive interference events [15].
Due to its high anticipated performance, the MaxEnt program, which uses maximum entropy modeling of species’ geographic distribution, has emerged as one of the most popular techniques [14]. MaxEnt algorithm only requires presence data and less concerned with the amount of occurrence records required to create accurate models. The selection of appropriate environmental variables and model parameters is an important step that must be considered [16,17]. Continuous mapping of economically important pulse crops such as mungbean and cowpea is critical for long-term sustainable agricultural production improvement in the face of climate change and species diversity problems or crises.
Mungbeans (V. radiata) is creeping annual crop belongs to the legume family growing up to 90 cm height. The crop is short-lived (75–90 days) grows in all types of soil. It is distinguished by its greenish yellow to pale yellow flowers, alternate, trifoliate leaves, and dangling, glabrous fruits (pods) that grow up to 12 cm and contain 10–15 ellipsoidal green, yellow to brown, or black mottled pods [18]. Mungbean is a native of Bangladesh, India, and Pakistan. Mungbean have been traditionally consumed all over the world for over 3500 years [19]. For instance, Indian farmers have grown mungbeans extensively for 3500 years, and these home-grown mungbeans quickly spread from India to China and various regions of Southeast Asia [20]. Currently, the China National Crop Gene Bank has identified and documented more than 5000 mungbean accessions [21].
In Ethiopia’s semi-arid regions, mungbeans become the most important crop for food, feed, and cash crops. It is well-known for its rich vitamin A and protein content, which can contribute to a well-balanced diet when mixed with cereal [22]. In addition to its economic benefit the crop can grow grown in different agro-ecological zones around the world [23]. Through the addition of organic matter and biological nitrogen fixation, mungbeans also help to increase the natural resource base [8]. Mungbean is an essential pulse that is grown in a variety of agricultural ecozones around the world [23]. Drought adaptability, quick maturation periods, adaptability to harsh (semiarid and arid) climatic conditions and thrive in rain-fed areas are all characteristics of mungbean due to their fast growing life cycle [24,25,26]. Mungbean is grown on an estimated 41,633.2 hectares in Ethiopia, with a total production of 514,227.41 tons and an average yield of 1.23 tons per hectare. Ethiopia’s current regional production of mungbeans is projected to be 122.14 hectares [27]. This indicates that the country had huge potential for production on mungbeans for sustainable development and food security. However, compared to cereals, which have been chosen for high grain yield under high input circumstances, the productivity of most pulses is quite low in Ethiopia [28,29]. This could be because of several reasons; (i) mungbean productivity is declining as a result of biotic and abiotic factors such as diseases, severe temperatures, insect pests, and pest incidence, and recurrent drought [25,29]; (ii) mungbeans are given less attention in Ethiopia and are generally farmed by smallholder farmers in marginal environmental conditions, with a lesser production potential than other pulse crops [28]. Like mungbeans, another economically important legume crop in Ethiopia is cowpea.
Cowpea (Vigna unguiculata L. Walp.) is multipurpose legume which provides food, feed, medicine, soil fertility improvement, nitrogen fixation and income generation for small farm holders [30,31,32,33,34]. Cowpea is vital to the lives of millions of people in poor countries worldwide [35]. The crop is grown mainly in the lowlands of Ethiopia. In comparison to other major producing countries, Ethiopia has very low production and consumption. Nevertheless, the crop contributes significantly to the livelihood of smallholder farmers [30]. Cowpea is an indigenous to African which is produced all over the tropical and subtropical regions of the world which is crucial for human nutrition. All parts of cowpea are nutritious including carbohydrates, proteins, minerals, and vitamins [32,34,36]. Minerals and vitamins are crucial in determining human growth and development. The proper growth, development, and economic status of humans are hampered by vitamin and mineral deficits [37]. Deficits in macro- and micronutrients are substantial obstacles to raising the socioeconomic level of food nutrition in the twenty-first century [37]. Additionally, cowpea provides a number of economic, environmental, and agronomic benefits that support environmental preservation and food security [35]. Legume production and productivity are poor, particularly in Africa, with best-performing countries reporting average yields of 5000 kg per hectares [38]. The main reason could be that cowpea cultivation is primarily done using conventional methods, and yields in farmers’ fields are low, particularly in the African region, average production is about 0.025–0.3 tons per hectares [31]. Several studies reported that the reason for its low productivity can be attributed to a variety of factors, including prolonged drought periods, poor soil fertility, inappropriate cultivar, inadequate planting systems, lack of improved seeds and pest attacks [30,31,32,33,36]. Cowpea contributes significant to the long-term viability of agricultural systems and the improvement of soil fertility on marginal lands [39]. It serves as a ground cover, as well as a nitrogen fixer and weed suppressor. The crop grows even in poor soils with more than 85% sand and less than 0.2% organic matter [40]. Although Ethiopia has high potential arable lands for legume crops, their production is insignificant and there is information gap related to mungbean and cowpea diversity, agricultural potential areas, climate impacts and distributions of these legume crops in Ethiopia. To this effect, the main objective of this study was to map the existing and projected potential agricultural lands for better production of mungbean and cowpea to enhance sustainable production and food security in Ethiopia.

2. Method and Materials

Study Area

Ethiopia is a nation in the Horn of Africa that shares borders with South Sudan to the west, Kenya to the south, Somalia to the south and east, Djibouti to the east, Eritrea to the north, and Sudan to the northwest. Geographically speaking, Ethiopia is located between the equator and the tropic of cancer, between latitudes 3° and 15° N and longitudes 33° and 48° E. The nation’s 1.14 million km2 is made up of high, rocky plateaus as well as peripheral dry and semi-arid lowlands [8]. The study incorporates all areas of Ethiopia to project the existing and future distribution of both mungbeans and cowpea in Ethiopia.

3. Data Collection

Species Occurrence

The data contains occurrence records for the focal species as well as environmental data for its habitats. Using GPS data with high-quality locality points and confident taxonomic identification is required for selecting and then processing these points [41]. The occurrence data of mungbean and cowpea was collected from Addis Ababa University’s National Herbarium (38°45′55.04″ E and 9°1′58.77″ N), the Ethiopian Biodiversity Institute (38°46′56.53″ E and 9°2′0.80″ N), the Global Biodiversity Information Facility (http://www.gbif.org, accessed on 4 April 2022) online data source, and the published literature [25,32,36]. Updated and recent records (1970s up to present time) geographical occurrence points of species were considered because the quality/accuracy of location data frequently declines with specimen age [42,43]. For this study, by checking duplicate occurrences, 81 occurrences for mungbean and 68 for cowpea were used for training and testing the MaxEnt modeling algorithm. The MaxEnt model performs better with more presence observations (sample size) under constant prevalence, and worsens with more prevalence under constant sample size [44]. However, the lower bounds for producing nonrandom models range from 14 for narrow-ranged species to 25 for widespread species [44,45], confirming that the current sample is more than adequate. The present study used 1 km2 distances between two sample points or species occurrence records.

4. Bioclimatic Variables

Data from the environment are used to quantify the ecological tolerance of the target species. Environmental variables that are relevant to the species being modeled were included in the model’s analysis. Many environment variables are so closely correlated that some are redundant, making it difficult to interpret the impact of each variable [46]. Bioclimatic variables for both plant species’ distribution modeling were extracted from the WorldClim version 1.4 (http://www.worldclim.org, accessed on 27 May 2022) dataset with a spatial resolution of about 1 km2 for the years 1960–1990 [47]. Elevation, topographic position index (combined indices for slope, aspect, and altitude), and solar radiation index were added as predictor variables in addition to 19 bioclimatic variables. Future climatic factors for inter-comparison project Phase 5 (CMIP 5) (IPCC Fifth Assessment Report, AR5) were used to predict the potential habitat distribution of the species. Predictions were based on three future climate change scenarios (RCP 2.6, RCP 4.5, and RCP 8.5) of the 2050s and 2070s, named after the potential radiative forcing value in 2100), relative to the pre-industrial values (+2.6, +4.5, +6.0, and +8.5 W/m2, respectively) [48]. In accordance with [49], the RCP 2.6 scenario has the lowest expected GHG emissions (a likely average global temperature increase of 0.3–1.7 °C at the end of the century), followed by the intermediate RCP 4.5 and RCP 6.0 scenarios, and the most extreme RCP 8.5 scenario (a likely average global temperature increase of 2.6–4.8 °C at the end of the century) with the highest expected GHG emissions. The potential suitable production and distribution areas for both cowpea and mungbeans in Ethiopia were predicted using six climate change scenario combinations (RCP 2.6, RCP 4.5, RCP 8.5 for the 2050s and RCP 2.6, RCP 4.5, and RCP 8.5 for the 2070s) which represent the average GHG emissions for the years 2041–2060 and 2061–2080. Finally, the environmental variables were geo-referenced and converted to ASCII files with the same spatial resolution of 30-s latitude/longitude and ground resolution of around 1 km2.

5. Environmental Variable Selection for Modeling

Environmental variables for this study were screened in three ways. First, the 22 environmental variables were connected to the occurrence points (longitude and latitude) of each plant species. Then, using the jackknife test and percentage contribution to the model, an initial MaxEnt model with all environmental variables was performed, and the least contributing variables were eliminated [50,51]. Secondly, the bioclimatic variables were categorized into three groups (Bio1- bio11-temperature groups, bio12-bio19- precipitation), and other groups (topographic positioning index, solar radiation, and elevation). Thirdly, R program (version 4.1.3) was run, and statistically significant factors in projecting V. radiata and V. unguiculata distribution, physiologically significant for the establishment and survival of these species but uncorrelated with environmental factors, were selected. Finally, to account for the pairwise correlation between the predictors, the final set of variables was chosen using a Pearson’s correlation coefficient r < 0.80 [52] (Supplementary Tables S1 and S2 for both species).
The R program was run again for a variance inflation factors (VIF) test (Supplementary Figures S1 and S2) to find collinearity among all the bioclimatic variables, and all variables with VIF values greater than 10 were eliminated to reduce the influence of multicollinearity and model over fitting [53,54]. In general, a VIF value greater than 10 is often thought to be highly correlated to other variables, resulting in multi-collinearity difficulties [55]. For this study, eleven environmental variables for V. radiata and ten environmental variables for V. unguiculata were retained from 22 variables in the form of ASCII raster grids format which was used by the MaxEnt algorithm.
In order to analyze changes in appropriate habitat ranges in the 2050s (average for 2041–2060) and 2070s (2061–2080) climate conditions, greenhouse scenarios such as RCP 2.6 (lowest GHG emissions), RCP 4.5 (intermediate GHG emissions), and RCP 8.5 (highest GHG emissions) were used. Environmental variables such as monthly average minimum and maximum temperatures and precipitation are in multiple of ten downloaded from the WorldClim 1.4 average value climate data for 1970–2000 as baseline climate [56]. Using ArcGIS 10.7, the existing and prospective areas of suitability changes were studied under two categories to identify the locations that are appropriate and those that are not suitable. The permutation importance and contribution percentage are important parameters in determining the importance of an environmental variable. Permutation importance rather than the path was used in the individual run-up to the model’s final results, which makes it simple to assess the significance of a particular variable.

6. Data Analysis

Before running the model, the multicollinearity and variable inflation factors (VIF) of the 22 environmental factors were assessed using Pearson’s pairwise correlation matrix analysis (Supplementary Tables S1 and S2). R program version 4.1.3 was used for this purpose, and only weakly correlated variables (r < 0.80) were included in the model. Then MaxEnt version 3.4.4 was calibrated to predict potential agricultural areas for mungbean and cowpea across Ethiopia [57]. To study the association between species habitat suitability and environmental variable distribution, species response curves were created [15]. The predicted species distribution chart has values ranging from 0 to 1. These figures are classified into four categories: high potential (>0.6), good potential (0.4–0.6), moderate potential (0.2–0.4), and low potential (0.2–0.2) [58]. The logistic output, which generates a continuous map with an estimated probability of occurrence between 0 and 1, was used for model computations. For each plant species, fifteen cross validation runs were performed and the results were averaged to form a single model. The ability of each model to predict the one locality that was not included in the training data set was then utilized to evaluate predictive performance. The remaining possibilities were left as default. Developed models were transferred through time to estimate changes in suitable zones for species under various climatic scenarios.

7. Evaluation of Model Performance and Simulation

MaxEnt is user-friendly, less sensitive to sample size, and accurately simulates species distribution [14] and hence was used for this data analysis. For the model simulation, the random test percentage was set to 25, allowing MaxEnt to withhold 25% of the presence data to be used randomly to evaluate model performance while keeping the remaining 75% for training the model [14]. MaxEnt algorithms were programmed to run the fifteen model replicates before averaging the results from all of the models, 5000 iterations, 0.00001 threshold convergence, 10,000 background points, subsample replicate run type, auto-features, and logistic output format [59]. The jackknife test was used to determine which predictor factors were most important to the species’ distribution [58,60]. Then, the MaxEnt output was classified in ArcGIS using the maximum training sensitivity plus specificity logistic threshold values determined from model results as highly suitable (0.85–1), suitable (0.69–0.85), moderately suitable (0.54–0.69), and less suitable regions (0.54–0.69) [61]. The area under the curve (AUC) value determined from receiver-operating characteristic (ROC) analysis was used to assess the model’s ability to forecast species distribution [14]. AUC, a threshold-independent indicator in the modeling process, evaluates the MaxEnt model’s ability to separate background into a high likelihood of presence and absence [62]. Perfect discrimination can be seen in the ranges between 0 and 1, where a value of 1 denotes perfect discrimination. Model performance is given a failing grade (0.5–0.6), a bad grade (0.6–0.7), a decent grade (0.7–0.8), a better grade (0.8–0.9), or a superb grade (0.9–1) [63].

8. Result

8.1. Environmental Variable Contributions for Mungbeans and Cowpea Distribution in Ethiopia

The environmental data give landscape-level information to calculate the ecological tolerances of the focal species. Tightly uncorrelated environmental variables should be used to prevent some redundancy, and this includes variables that are probably directly relevant to the species being modeled. The geographic MaxEnt analysis show that the precipitation of the wettest quarter (Bio16), solar radiation index (SRI), elevation (Elv), temperature seasonality (Bio4) and precipitation seasonality (Bio15) make significantly higher contributions to the distribution modeling of mungbeans (V. radiata) in Ethiopia under various global warming scenarios (Table 1). The outcome of the jackknife test of variable significance confirmed that precipitation of the wettest quarter (Bio16), temperature seasonality (Bio4), annual range temperature (Bio7), and precipitation of the driest quarter (Bio17) made a greater contribution to the geographical distribution of cowpea (V. unguiculata), while the contribution of the remaining environmental variables ranged from medium to least for the distribution of both crops in Ethiopia (Figure 1A–D; Table 1).
With an AUC value of 0.723 and 0.880 under current climate scenario, the MaxEnt algorithm projected the possible appropriate habitats for V. radiata and V. unguiculata in Ethiopia, respectively with acceptable statistical accuracy (Figure 2A,B). The internal Jackknife test of the MaxEnt model revealed that the variable Bio16 is a more relevant variable in determining habitat appropriateness for both V. radiata and V. unguiculata in Ethiopia (Figure 1 and Table 1). Bio16, the environmental variable with the largest gain when used in isolation, appears to have the most helpful information by itself. Therefore, Bio16 is the environmental factor whose absence significantly reduces the gain, indicating that it contains the most information not found in the other factors taken into account in this study (Figure 1A–D; Table 1).
The receiver operating characteristic (ROC) curve for the same data is shown in Figure 2. This curve was also averaged over the replicated 15 runs. The analysis revealed that the average test AUC for the replicate runs of ROC was found to be 0.723 ± 0.052, and 0.880 ± 0.046 for V. radiata and V. unguiculata, respectively (Figure 2A,B). The AUC values that were considerably different from the standard set for random prediction (AUC = 0.5 and close to 1) show that the MaxEnt model could accurately estimate the locations of potentially appropriate habitats for both species [61]. The average AUC value for predicting V. radiata and V. unguiculata distributions under future climate change scenarios was also found to be between 0.71 and 0.91. Since the AUC values for all future years/scenarios were greater than 0.70, the model proved robust enough to accurately classify the background into presence and absence categories for these crops (Figure 2).

8.2. Current Distribution of Mungbeans and Cowpea in Ethiopia

The point-wise mean and standard deviation of the 15 run output grids are depicted (Figure 3). The distributions of both V. radiata and V. unguiculata in several categories of habitat suitability (i.e., highly suitable, moderately suitable, unsuitable and less suitable) were concentrated in drought-prone areas of Ethiopia. The result indicates that the highly and moderately suitable distribution areas for V. radiata mainly include areas such as northern, northwestern and western Gonder, western, northern, eastern, central and southern Tigray, northern and southern Wello, northern and western Shewa, West Hararge, West Bale, Arsi, Kelem Wollega, Metekel, Assosa area, western Burji, Mano special woreda and Filtu areas, to mention a few under the current scenario (Figure 3A). Highly suitable and moderately suitable habitats for V. unguiculata include central, eastern, northwest, southeast, and northern Tigray regions, Wag Hamra, North and East Shewa, West Hararghe, Harari, Dire Dawa, Arsi, Gofa, Gamo, Derashe, South Omo, Basketo special woreda, Konso, Agnewak, Kelem Wellaga, and Metekel areas, while the remaining areas of Ethiopia become less suitable and unsuitable areas under the existing and projected global warming situations (Figure 4A–D). From the total land areas of Ethiopia, only highly suitable (1.69%), moderately suitable (4.27), less suitable (11.25%) and unsuitable (82.79%) are predicted habitat ranges of V. radiata under current climatic scenarios (Figure 3A and Table 2).
For V. unguiculata distribution in Ethiopia under the current climate scenario, predicted habitats are: highly suitable (1.27%), moderately suitable (3.07%), less suitable (5.22%), and unsuitable area (90.44%)(Figure 4A and Table 2). The distributions of V. radiata and V. unguiculata are mainly determined by the major environmental variables which include Bio16, SRI, Elv and Bio16, Bio4, Bio7, and Bio17 (Table 1). The current and future map of the result, under different climate scenarios (2.6, 4.5, 8.5/2050s and 2070s) indicate a slight difference on the distribution and suitable potential habitats for both of V. radiata and V. unguiculata crops in Ethiopia (Figure 3A–D). This implies that the climate changes are threatening these crops’ distribution and production in Ethiopia and will continue in the future in the same manner. Therefore, the result indicates that the major future distribution areas of both crops are restricted to the current distribution areas in Ethiopia.

8.3. Future Projected Distributions of Mungbeans and Cowpea Crops in Ethiopia

The MaxEnt model predicted that central, northern, and south eastern Tigray, East and northern Shewa, Wag Hamra, North and South Wello, West Hararge, Harari, Yem special woreda, Agnewak, Kellem Wollega, South Omo, Gamo, Gofa, Alle, Konso, and Basketo special woreda areas are the best appropriate habitats for V. unguiculata in Ethiopia (Figure 4B–D). The model predicted future potential appropriate areas suitable for V. radiata which includes North and western Gonder, South and North Wello, Western Hararge, Western Wollega, Metekel, Harar, western and eastern Tigray, North and southern Shewa, Jimma, Kelem Wollega, Assosa areas, Gamo, Gofa, Dawuro, Wolayita and Yem special woredas (Figure 3B–D). The model’s output revealed that an area of 1.75 million hectares (Mha) (1.54%) (2.6/2050s), 2.06 Mha (1.82%) (4.5/2050s), 1.92 Mha (1.7%) (8.5/2050s), 1.61 Mha (1.42%) (2.6/2070s), 1.89 Mha (1.68%) (4.5/2070s), 1.97 Mha (1.74%) (8.5/2070s) under the three future climate scenarios had high potential for V. radiata distribution and production (Figure 3B–D and Table 2), while 1.41 Mha (1.24%) (2.6/2050s), 1.54 Mha (1.36%) (4.5/2050s), 1.41 Mha (1.24%) for 8.5/2050s, 1.52 Mha (1.34%) (2.6/2070s), 1.53 Mha (1.36%) (4.5/2070s), 1.36 Mha (1.21%) (8.5/2070s) areas under the three future climate scenarios indicated highly suitable habitat distribution for V. unguiculata distribution in Ethiopia (Figure 4B–D and Table 2).
Similarly, an area of approximately 92.32 Mha (81.53%) (2.6/2050s), 91.56 Mha (80.87%) (4.5/2050s), 92.15 Mha (81.38%) (8.5/2050s), 94.09 Mha (83.1%) (2.6/2070s), 93.54 Mha (82.61%) (4.5/2070s), 91.02 Mha (80.39%) (8.5/2070s) and 10.29 Mha (90.9%) (2.6/2050s), 10.32 Mha (91.11%) (4.5/2050s), 10.29 Mha (90.96%) (8.5/2050s), 10.21 Mha (90.16%) (2.6/2070s), 10.26 Mha (91.63%) (4.5/2070s), 10.28 Mha (90.79%) (8.5/2070s) areas under the three future climate scenarios were predicted as unsuitable distribution areas for V. radiata and V. unguiculata, respectively, in Ethiopia (Figure 3B–D; Figure 4B–D and Table 2). Furthermore, the reaming areas of the model maps, indicated moderately to less suitable habitats of both V. radiata and V. unguiculata crops in Ethiopia (Figure 3 and Figure 4B–D and Table 2). The map shows that drought prone areas had high potential habitat suitability for both V. radiata and V. unguiculata from low to mid-altitude in Ethiopia (Figure 3 and Figure 4). Distribution and production unsuitability dominates Ethiopia’s total land area, followed by less suitable, moderately suitable, and highly suitable areas for both crop distributions in Ethiopia (Table 2).
The future map MaxEnt models projection identified almost similar current and future distribution habitat suitability for both crops in Ethiopia (Figure 3 and Figure 4). This implies that climatic change has an effect on the spread and production of V. radiata and V. unguiculata crops in Ethiopia.

8.4. Geographical Distribution and Species Response to the Environmental Variables

Figure 5 below depicts how each environmental variable influences the MaxEnt predictions. The curves depict the mean response of the 15 replicate runs (red) and the mean +/− one standard deviation (blue, two shades for categorical variables). These plots depict the relationship between anticipated appropriateness and the chosen variable. The distribution of V. radiata was restricted to precipitation of wettest quarter (Bio16) value range of 600 mm, and outside this range the species’ distribution dramatically reduced as Bio16 values increased, with very low rates of presence over 800 mm (Figure 5A). The probability of presence also increased as precipitation seasonality (Bio15) rose, with the maximum probability of presence occurring in the 120–140 mm range (Figure 5D).
Similarly, as temperature seasonality (Bio4) climbed from 0 °C to 150 °C, the probability of presence increased, then quickly declined beyond (Figure 5C). In contrast, a negative association was discovered between the solar radiation index (Srad) and the likelihood of V. radiata’s existence (Figure 5B).
The probability of presence increased dramatically with the increase in precipitation of the wettest quarter (Bio16), even beyond 400mm value, which indicates the distribution and presence of V. unguiculata highly depends on precipitation of wettest quarter (Figure 6A). The chance of existence of V. unguiculata depends on a range of 20–220mm of precipitation of the driest quarter (Figure 6D), while temperature seasonality (Figure 6B) and temperature annual range (Figure 6C) are major limiting environmental variables for the distribution of V. unguiculata in Ethiopia (Table 1).

9. Discussions

Achieving the AUC values of 0.723 ± 0.052 and 0.880 ± 0.046 under the current climate scenario, the model is considered as well fit and accurate for appropriate habitat prediction under the present climate change scenarios in Ethiopia (Figure 2A, B). When projected using two future climates (2050s and 2070s) with three various global warming scenarios for each year (RCP 2.6, RCP 4.5 and RCP 8.5), V. radiata and V. unguiculata were found to be extremely vulnerable to future climate change as it is confined to specific suitable areas in Ethiopia. From the total land area of Ethiopia, based on the four categories of the model for both species’ distribution, unsuitable areas are dominant followed by less suitable, moderately suitable and highly suitable (Figure 3; Figure 4 and Table 2). The prediction map and interception calculation for both present and projected (in the 2050s and again in the 2070s) climate change scenarios confirmed that all the RCPs are shown to be almost similar in suitable habitat which implies the continuous threats of climate change on crops distribution and production in Ethiopia (Figure 3 and Figure 4A–D). Plants with a confined ecological niche and a limited geographic range are more vulnerable to the effects of climate change [64]. The finding of the present study confirmed that the distribution and production of V. radiata and V. unguiculata are limited to specific suitable areas and show fragmented distribution (Figure 3A–D and Figure 4A–D). Owing to the fact that the range is very susceptible to anthropogenic pressure and general climate change, scenarios for species that are naturally constrained to small habitats show that the vulnerability of a plant species to extinction is closely related to its habitat range size and range shift performance [65].
The internal Jackknife approach and variable importance test suggest that both environmental variables and non-climatic variables (solar radiation) have a higher relative value, indicating that V. radiata has a high particular preference for these environmental variables for its growth and distribution (Figure 1A,B and Table 1). Among the environmental variables studied for habitat suitability determination, Bio16 was found to be the most important driver for both crop species distribution in Ethiopia. The precipitation of the wettest quarter (Bio16) values ranging from 200–800mm have a higher probability (p > 0.5) of having good habitat appropriateness for the distribution of both crop plant species in Ethiopia (Figure 5 and Figure 6A). These crops require between 750 and 5000 mm of rainfall annually in their natural habitat [66], and they can tolerate up to six dry months per year with just 40 mm of mean monthly rainfall [67]. This indicates that V. radiata and V. unguiculata crops grow in drought-prone, marginal and lowland areas of Ethiopia.
Precipitation of the wettest quarter (Bio16), solar radiation index (SRI), temperature seasonality (Bio4), and precipitation seasonality (Bio15) were limiting factors for V. radiata adaptability while and precipitation of wettest quarter (Bio16), temperature seasonality (Bio4), temperature annual range (Bio7) and precipitation of the driest quarter (Bio17) were the limiting factors for V. unguiculata adaptability in Ethiopia. Changes in precipitation and temperature are anticipated to have a negative impact on the distribution of these species [68]. The current study also revealed that V. radiata distribution and production prefers a temperature range of up to 42 °C to 100 °C (Figure 5C). V. radiata is a creeping annual crop that can reach a height of 90 cm in temperatures as high as 35 °C. It has more flexibility adaptability and a short growing season (75–90 days). In spite of the most challenging arid and semi-arid environments, V. radiata thrives widely on all types of soil [18]. Warm season crop V. radiata requires 90 to 120 days and prefers temperatures ranges between 27–30 °C [69,70]. This indicates that the crop is often grown in the summer [69]. To obtain a good grain yield, water is needed from flowering to late pod setting stages [71]. Temperature and light intensity influence photosynthetic rate during critical growth stages of the plants [71]. The crop growth and development are mainly affected by solar radiation. The solar radiation index is proportionate to the amount of direct solar radiation that reaches the earth’s surface as a function of its aspect, slope, and latitude [72], and significantly affects the distribution of V. radiata and V. unguiculata in Ethiopia (Figure 5). Solar energy provides the light needed for seed germination, leaf expansion, root, stem, and shoot growth, flowering, and fruiting. Solar radiation also effects nutrition assimilation and dry matter distribution (www.agritech.tnau.ac.in/agriculture/agri_agrometeorology_solar.html, accessed on 4 April 2022).
The finding maps indicate that the largest potential V. radiata production areas in Ethiopia are moisture-stressed areas, which include Assossa, Kamash, West Hararge, West Wollega, Kellem Wollega, Bale; Delo Menna and Mera woredas, Dire Dawa, Gurage-Abeshege woreda, Oromia special zone, East, West and North Shewa, western and northern Gonder, East Gojam, Central, North and South Wello, North, Western and Southern Tigray, Koysha, Wolayita, Halaba, Amaro, Burji, Gamo, Konso, Basketo and Majang areas, to mention a few (Figure 3A–D). The study agrees with a report that, small-scale farmers of North Debere Sina, Shewa, Qallu, South Wollo and some districts of Benishangul Gumuz Region produces V. radiata in Ethiopia [27]. Another study reported that V. radiata crops are potentially produced in drought prone areas of the Ethiopia such as South Omo zone, Konso, Gofa, and Konta special district [28].
Similarly, the study maps reveal that the main production areas of V. unguiculata in Ethiopia are East Gojam, North Wello, South Wello, central Tigray, northern Tigray, eastern Tigray, Gambella, Arsi, Gamo, Iluu Ababor, Basketo special woreda, eastern Shewa, Konso, Dire Dawa, Harari and eastern Hararghe (Figure 4A–D). The major V. unguiculata production area includes the Amhara and Tigray regions [32], while Gambella, Oromia, Dire Dawa and SNNPR were reported by [34]. According to [32], V. unguiculata is currently being produced in the country’s northwest, east, central rift valley, and southern regions. It is regarded as the most significant lowland pulse utilized for food, feed, restoring soil fertility, and for revenue generation [73,74]. Cowpea is grown mostly as a single crop (48.75%), followed by intercropping (35%) [73,74]. The result reveals that V. unguiculata growth and production was higher at optimal precipitation of wettest quarter (Bio16) ranging from 200 mm to 450 mm (Figure 6A). Another study found that V. unguiculata outperforms most tropical legumes, reaching adequate yield performance with minimal rainfall of up to 450 mm per year and heat stress [75]. Furthermore, higher CO2 level enhances N2 fixation in V. unguiculata, resulting to increased photosynthesis and yield [75,76]. Climatic factors such as temperature, CO2 concentration, and light (UV radiation) which affects biological nitrogen fixation [77], high temperature [78] and elevated ultraviolet radiation [79] decreased the performance of V. unguiculata crops because these variables impede symbiosis [75]. Even though both crops are distributed at wider lowland and midland agro-ecological areas of Ethiopia (Figure 3 and Figure 4A–D), the productivity of these crops, important both economically and for food security, remained low in Ethiopia as a result of the low attention paid to the role these crops can play to ensure food security in Ethiopia. The study report in 2014 indicated that the first ranked cowpea producer in the world was Nigeria with an annual production of 2,137,900 tons from 3,701,500 ha of land and 578 kg/ha [80]. According to a survey conducted in the main cowpea-producing districts of Ethiopia (Amhara, Gambella, Oromia, SNNPR, and Tigray), yearly cowpea output is projected to be 55,600 tons grown on 69,500 acres of land [39]. The average grain productivity of cowpeas on farmers’ fields was found to be low (0.8 tons ha−1) when compared to the yield on a farm demonstration research station, which produced 1.7 to 2.1 tons ha−1 for improved varieties, whereas the average yield of cowpea on the research plot was 2.2 to 3.2 tons ha−1 [39]. The major production constraints and challenges facing V. radiata and V. unguiculata production in Ethiopia includes limited attention to these crop, pests, infertile soil, water stress, market failure, poor management practices, and lack of technology dissemination [28,30,31,32,33,36]. This study only anticipated potential habitats for V. radiata and V. unguiculata in Ethiopia. However, there could be huge regions of potential habitats outside of refugia of these important crops in Ethiopia. Owing to artificial and natural constraints, both crops may have been unable to disperse from their geographic distributions [58]. This potential habitat distribution map (Figure 3A–D and Figure 4A–D) for V. radiata and V. unguiculata can serve as a reference for agricultural resource managers, policymakers, stakeholders and future researchers to pay attention to increasing V. radiata and V. unguiculata production and to enhance the livelihood of smallholder farmers and to improve food security in Ethiopia.

10. Conclusions

A study has examined the potential distribution suitability of the current and future agricultural areas for the sustainable production of V. radiata and V. unguiculata in Ethiopia. The maximum entropy model was used to analyze the potential suitable habitat for both crops in Ethiopia based on present and future sets of environmental factors. The adaptability of V. radiata and V. unguiculata habitat for current and future climate change scenarios was accurately modeled in this study. Final suitability work, from the total land areas of Ethiopia, highly suitable (1.69%), moderately suitable (4.27), less suitable (11.25%), and unsuitable (82.79%) habitat ranges for V. radiata are predicted, whereas V. unguiculata distribution in Ethiopia, highly suitable (1.27%), moderately suitable (3.07%), less suitable (5.22%), and 90.44% the study area was not suitable for V. unguiculata under the existing climate scenarios. The results revealed that the distribution, habitat suitability, and productivity of V. radiata were limited by the precipitation of the wettest quarter (Bio16), solar radiation index (SRI), and temperature seasonality (Bio4), whereas precipitation seasonality (Bio15), precipitation of the wettest quarter (Bio16), temperature annual range (Bio7), and precipitation of the driest quarter (Bio17) were found to be the most important environmental variables to define V. unguiculata habitat suitability in Ethiopia. Furthermore, the results showed that both crops adapted to wide ranges of agro-ecological areas of Ethiopia. Further distribution and production areas will be restricted as a result of climate change’s detrimental effects. Production and utilization of V. radiata and V. unguiculata crops in Ethiopia are very low, remaining below the potential of the crops and the world production average yield as compared to other major producing countries. Nevertheless, such crops make a significant contribution to the livelihood of the smallholder farming communities in Ethiopia, which have rapidly growing populations and need to meet their food needs. Knowing the current and future distribution of these economically important crops is critical to decide an appropriate production strategy that improves these crops’ production. This study could be useful to optimize these crops’ distribution and production in areas of high and moderate suitability where the crops are presently lacking. The habitat suitability map might be more realistic if more environmental covariates, such as anthropogenic threats were added, which will be strongly recommended for future researchers.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/su15010600/s1, Figure S1: Correlation plots indicating the variable inflation factor analysis output using R program version 10.7 for V. radiata; Figure S2: Correlation plots indicating the variable inflation factor analysis output using R program version 10.7 for V. unguiculata; Table S1: Pearson correlation matrix among 22 predictor variables for V. radiata; Table S2: Pearson correlation matrix among 22 predictor variables for V. unguiculata.

Author Contributions

Conceptualization, B.K. and A.A.; Methodology, B.K; Formal analysis, B.K.; Investigation, B.K.; Data curation, B.K.; Writing—original draft, B.K.; Writing—review & editing, A.A. and A.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data can be obtained from the corresponding author based on request via email.

Acknowledgments

The authors are distinctly thankful to Addis Ababa University, National Herbarium of Addis Ababa University, Ethiopia, the Ethiopian Biodiversity Institute (EBI), Global Biodiversity Information Facility (GBIF) database and the published literature for providing the necessary information on the two crops. The three anonymous reviewers are duly acknowledged for their critical comments and suggestions which improved the quality of this paper.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Tesfaye, K.; Mamo, G.; Debela, S.; Tadesse, M.; Mekuriaw, A.; Debele, B.; Semane, B.; Gebrekidan, B.; Lemenih, M.; Zenebe, M. First Assessment Report on Agriculture and Food Security. Working Group II—Climate Change Impact, Vulnerability, Adaptation, and Mitigation; Ethiopian Academy of Sciences (EAS): Addis Ababa, Ethiopia, 2015; p. 230. [Google Scholar]
  2. IPCC (Intergovernmental Panel on Climate Change). Climate Change 2007–The Physical Science Basis; Cambridge University Press: Cambridge, UK, 2007. [Google Scholar]
  3. Jones, P.G.; Thornton, P.K. The potential impacts of climate change on maize production in Africa and Latin America in 2005. Glob. Environ. Chang. 2009, 13, 51–59. [Google Scholar] [CrossRef]
  4. Conway, D.; Schipper, E.L.F. Adaptation to climate change in Africa: Challenges and opportunities identified from Ethiopia. Glob. Environ. 2014, 21, 227–237. [Google Scholar] [CrossRef]
  5. Demeke, A.B.; Keil, A.; Zeller, M. Using panel data to estimate the effect of rainfall shocks on smallholders’ food security and vulnerability in rural Ethiopia. Clim. Chang. 2011, 108, 185–206. [Google Scholar] [CrossRef]
  6. Araya, A.; Stroosnijder, L.; Habtu, S.; Keesstra, S.D.; Berhe, M.; Hadgu, K.M. Risk assessment by sowing date for barley (Hordeum vulgare) in northern Ethiopia. Agric. For. Meteorol. 2012, 154, 30–37. [Google Scholar] [CrossRef]
  7. Deresa, T.T. Measuring the Economic Impact of Climate Change on Ethiopian Agriculture. Centre for Environmental Economics and Policy in Africa, University of Pretoria; Ethiopian Revenue and Custom Authority (ERCA): Addis Ababa, Ethiopia, 2006.
  8. Central Statistical Agency (CSA). Agricultural Sample Survey Report on Area and Production of Crops (Private Peasant Holdings ‘Meher’ Season); Central Statistical Agency (CSA): Addis Ababa, Ethiopia, 2016.
  9. 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] [Green Version]
  10. 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]
  11. Guisan, A.; Zimmermann, N.E. Predictive habitat distribution models in ecology. Ecol. Model. 2000, 135, 147–186. [Google Scholar] [CrossRef]
  12. Carpenter, G.; Gillison, A.N.; Winter, J. DOMAIN: A flexible modelling procedure for mapping potential distributions of plants and animals. Biodivers. Conserv. 1993, 2, 667–680. [Google Scholar] [CrossRef]
  13. Stockman, A.K.; Beamer, D.A.; Bond, J.E. An Evaluation of a GARP Model as an Approach to Predicting the Spatial Distribution of Non-Vagile Invertebrate Species. Divers. Distrib. 2006, 12, 81–89. [Google Scholar] [CrossRef]
  14. Phillipsa, S.J.; Andersonb, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  15. Hussein, A.; Workeneh, S. Modeling the Impacts of Climate Changes on the Distribution of Aloe Vera Species in Ethiopia. J. Earth Sci. Clim. Chang. 2021, 12, 567. [Google Scholar]
  16. 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. Meth. Ecol. Evol. 2015, 6, 1126–1136. [Google Scholar] [CrossRef] [Green Version]
  17. Porfirio, L.L.; Harris, R.M.; Lefroy, E.C.; Hugh, S.; Gould, S.F.; Lee, G.; Bindoff, N.L.; Mackey, B. Improving the use of species distribution models in conservation planning and management under climate change. PLoS ONE 2014, 9, e113749. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  18. Lim, T.K. Edible Medicinal and Non-Medicinal Plants Fruits; Springer: Dordrecht, Germany; Berlin/Heidelberg, Germany; London, UK; New York, NY, USA, 2012; Volume 2, pp. 966–967. [Google Scholar]
  19. Lambrides, C.J.; Godwin, I. Mungbean. In Genome Mapping and Molecular Breeding in Plants Pulses, Sugar and Tuber Crops; Kole, C., Ed.; Springer: Berlin/Heidelberg, Germany, 2007; Volume 3, pp. 69–90. [Google Scholar]
  20. Shi, Z.; Yao, Y.; Zhu, Y.; Ren, G. Nutritional composition and antioxidantactivity of twenty mung bean cultivars in China. Crop J. 2016, 4, 398–406. [Google Scholar] [CrossRef] [Green Version]
  21. Liu, C.Y.; Chen, X.Z.; Wang, S.H.; Wang, L.X.; Sun, L.; Mei, L.; Xu, N. The genetic diversity of mungbean germplasm in China. J. Plant Genet. Resour. 2006, 7, 459–463. [Google Scholar]
  22. Rahman, M.; Bhuiyan, M.; Sutradhar, G.; Rahman, M.; Paul, A. Effect of phosphorus, molybdenum and rhizobium inoculation on yield and yield attributes of mungbean. Int. J. Sust. Crop Prod. 2008, 3, 26–33. [Google Scholar]
  23. Khan, M.A.; Naveed, K.; Ali, K.; Bashir, A.; Samin, J. Impact of mungbean-maize intercropping on growth and yield of mungbean. Pak. J. Weed Sci. Res. 2012, 18, 191–200. [Google Scholar]
  24. Anjum, M.S.; Ahmed, Z.I.; Rauf, C.A. Effect of rhizobium inoculation and nitrogen fertilizer on yield and yield components of mungbean. Int. J. Agric. Biol. 2006, 8, 238–240. [Google Scholar]
  25. Ayalew, T. Mungbean (Vigna radiata (L.) Wilczek) (Fabaceae) Landrace Diversity in Ethiopia. Master’s Thesis, Addis Ababa University, Addis Ababa, Ethiopia, 2015. [Google Scholar]
  26. Yagoob, H.; Yagoob, M. The effects of water deficit stress on protein yield of mungbean genotypes. Peak J. Agri. Sci. 2014, 2, 30–35. [Google Scholar]
  27. CSA (Central Statistical Authority). Report on the Preliminary Results of Area, Production and Yield of Temporary Crops in Ethiopia; CSA: Addis Ababa, Ethiopia, 2018.
  28. Asfaw, A.; Gurum, F.; Alemayehu, F.; Rezene, Y. Analysis of multi environment grain yield trials in mungbean (Vigna radiata L.) Wilczek based on GGE bipot in Southern Ethiopia. J. Agric. Sci. Technol. 2012, 14, 389–398. [Google Scholar]
  29. Das, S.; Shekhar, U.D.; Ghosh, P. Assessment of molecular genetic diversity in some green gram cultivars as revealed by ISSR analysis. Adv. Appl. Sci. Res. 2014, 5, 93–97. [Google Scholar]
  30. Erana, K.; Zelalem, B. Expounding the production and importance of cowpea (Vigna unguiculata (L.) Walp.) in Ethiopia. Cogent Food Agric. 2020, 6, 1769805. [Google Scholar] [CrossRef]
  31. Kamara, A.Y.; Omoigui, L.O.; Kamai, N.; Ewansiha, S.U.; Ajeigbe, H.A. Improving Cultivation of Cowpea in West Africa. In Achieving Sustainable Cultivation of Grain Legumes; Sivasankar, S., Bergvinson, D., Gaur, P., Kumar, S., Beebe, S., Tamò, M., Eds.; Burleigh Dodds Science Publishing Limited: Cambridge, UK, 2018; Volume 2, pp. 235–252. [Google Scholar]
  32. Mulugeta, A.; Zemede, A.; Zerihun, W.; Berhanu, A.F.; Beth, M. Cowpea (Vigna unguiculata (L.) Walp.) (Fabaceae) landrace diversity in northern Ethiopia. Int. J. Biodi. Conserv. 2016, 8, 297–309. [Google Scholar] [CrossRef] [Green Version]
  33. Ojiewo, C.O.; Omoigui, L.O.; Pasupuleti, J.; Lenné, J.M. Grain legume seed systems for smallholder farmers: Perspectives on successful innovations. Outlook Agric. 2020, 49, 286–292. [Google Scholar] [CrossRef]
  34. Sisay, A.; Mulugeta, A.; Zemede, A.; Woldu, Z.; Berhanu, A. Cowpea (Vigna unguiculata (L.)Walp., Fabaceae) landrace (local farmers’ varieties) diversity and ethnobotany in Southwestern and Eastern parts of Ethiopia. Afr. J. Agric. Res. 2019, 14, 1029–1041. [Google Scholar] [CrossRef] [Green Version]
  35. Carneiro da Silva, A.; da Costa Santos, D.; Junior, L.T.; Silva, P.B.; Santos, R.C.; Siviero, A. Cowpea: A strategic legume species for food security and health. In Legume Seed Nutraceutical Research; Jimenez-Lopez, J.C., Clemente, A., Eds.; IntechOpen: London, UK, 2018; pp. 47–65. [Google Scholar] [CrossRef] [Green Version]
  36. Sisay, A. Cowpea (Vigna unguiculata (L.) Walp.) (Fabaceae) Landrace Diversity in southern Ethiopia. Master’s Thesis, Addis Ababa University, Addis Ababa, Ethiopia, 2015. [Google Scholar]
  37. Jaiswal, D.K.; Krishna, R.; Chouhan, G.K.; Pereira, A.P.D.A.; Ade, A.B.; Prakash, S.; Verma, S.K.; Prasad, R.; Yadav, J.; Verma, J.P. Bio-fortification of minerals in crops: Current scenario and future prospects for sustainable agriculture and human health. Plant Growth Regul. 2022, 98, 5–22. [Google Scholar] [CrossRef]
  38. Food and Agriculture Organization Corporate Statistical Database (FAOSTAT)). Food and Agriculture Organization of the United Nations Statistics Division. 2016. Available online: http://faostat3.fao.org/download/Q/QC/E (accessed on 31 December 2016).
  39. Beshir, B.; Amsalu, B.; Dagmawit, T.; Selamawit, K.; Teamir, M.; Bezawit, Y. Cowpea Production, Marketing and Utilization in Ethiopia (Research Report 121). 2019. Available online: https://www.researchgate.net/publication/341042797_Cowpea_Production_Marketing_and_Utilization_in_Ethiopia (accessed on 12 April 2022). [CrossRef]
  40. Bilatu, A.; Binyam, K.; Solomon, Z.; Eskinder, A.; Ferede, A. Animal feed potential and adaptability of some cowpea (Vigna unguiculata) varieties in North West lowlands of Ethiopia. J. Agri. Res. 2012, 1, 478–483. [Google Scholar]
  41. Chan, L.M.; Brown, J.L.; Yoder, A.D. Integrating statistical genetic and geospatial methods brings new power to phylogeography. Mol. Phylogenet. Evol. 2011, 59, 523–537. [Google Scholar] [CrossRef]
  42. Bloom, T.D.; Flower, A.; DeChaine, E.G. Why georeferencing matters: Introducing a practical protocol to prepare species occurrence records for spatial analysis. Ecol. Evol. 2018, 8, 765–777. [Google Scholar] [CrossRef] [Green Version]
  43. Murphey, P.C.; Guralnick, R.P.; Glaubitz, R.; Neufeld, D.; Ryan, J.A. Georeferencing of museum collections: A review of problems and automated tools, and the methodology developed by the mountain and plains spatiotemporal database-informatics initiative (Mapstedi). PhyloInformatics 2004, 3, 1–29. [Google Scholar]
  44. Van Proosdij, A.S.; Sosef, M.S.; Wieringa, J.J.; Raes, N. Minimum required number of specimen records to develop accurate species distribution models. Ecography 2016, 39, 542–552. [Google Scholar] [CrossRef]
  45. Støa, B.; Halvorsen, R.; Stokland, J.N.; Gusarov, V.I. How much is enough? Influence of number of presence observations on the performance of species distribution models. Sommerfeltia 2019, 39, 1–28. [Google Scholar]
  46. Brown, J.L. SDMtoolbox: A python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods Ecol. Evol. 2014, 5, 694–700. [Google Scholar] [CrossRef]
  47. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 2005, 25, 1965–1978. [Google Scholar] [CrossRef]
  48. Weyant, J.; Azar, C.; Kainuma, M.; Kejun, J.; Nakicenovic, N.; Shukla, P.R.; Rovere, E.L.; Yohe, G. Report of 2.6 Versus 2.9 Watts/m2 RCPP Evaluation Panel. IPCC Secretariat: Geneva, Switzerland, 2009. [Google Scholar]
  49. Intergovernmental Panel on Climate Change (IPCC). Climate Change and Water; Technical Paper of the Intergovernmental Panel on Climate Change; IPCC Secretariat: Geneva, Switzerland, 2008; p. 210. [Google Scholar]
  50. Baldwin, R.A. Use of maximum entropy modeling in wildlife research. Entropy 2009, 11, 854–866. [Google Scholar] [CrossRef]
  51. Van, H.A.; Conti, F.; Ciaschetti, G.; Westinga, E. Fine resolution distribution modeling of endemics in Majella National Park, Central Italy. Plant Biosyst. Int. J. Deal. All Asps. Plant Biol. 2012, 146, 276–287. [Google Scholar]
  52. Zhao, X.; Ren, B.; Li, D.; Garber, P.A.; Zhu, P.; Xiang, Z. Climate change, grazing, and collecting accelerate habitat contraction in an endangered primate. Biol. Conserv. 2019, 231, 88–97. [Google Scholar] [CrossRef]
  53. Guisan, A.; Thuiller, W.; Zimmermann, N.E. Habitat Suitability and Distribution Models: With Applications in R; Cambridge University Press: Cambridge, UK, 2017. [Google Scholar]
  54. Yoon, S.; Lee, W.H. Methodological analysis of bioclimatic variable selection in species distribution modeling with application to agricultural pests (Metcalfa pruinosa and Spodoptera litura). Comput. Electron. Agric. 2021, 190, 106430. [Google Scholar] [CrossRef]
  55. Montgomery, P. An Introduction to Statistical Modelling. Technometrics 1992, 42, 435–436. [Google Scholar]
  56. Fick, S.E.; Hijmans, R.J. WorldClim 2: New 1 km Spatial Resolution climate Surfaces for Global GBIF.org. 2017. Available online: https://www.gbif.org/occurrence/download/0223934-210914110416597 (accessed on 15 April 2022).
  57. Steven, J.; Phillips, M.; Robert, E.S. Maxent Software for Modeling Species Niches and Distributions (Version 3.4.4). Available online: https://biodiversityinformatics.amnh.org/open_source/maxent/ (accessed on 12 December 2022).
  58. Yang, X.Q.; Kushwaha, S.P.S.; Saran, S.; Xu, J.; Roy, P.S. Maxent modeling for predicting the potential distribution of medicinal plant, Justicia adhatoda L in Lesser Himalayan foothills. Ecol. Eng. 2013, 51, 83–87. [Google Scholar] [CrossRef]
  59. Antúnez, P.; Suárez-Mota, M.E.; Valenzuela-Encinas, C.; Ruiz-Aquino, F. The Potential Distribution of Tree Species in Three Periods of Time under a Climate Change Scenario. Forests 2018, 9, 628. [Google Scholar] [CrossRef]
  60. Phillips, S.J.; Dud’ık, M.; Schapire, R.E. A maximum entropy approach to species distribution modeling. In Proceedings of the 21st International Conference on Machine Learning, Banff, AB, Canada, 4–8 July 2004; pp. 655–662. [Google Scholar]
  61. Enkossa, T.; Nemomissa, S.; Lemessa, D. Predicting the potential suitable habitats of forest spices Piper capense and Aframomum corrorima under climate change in Ethiopia. J. Trop. Ecol. 2022, 38, 219–232. [Google Scholar] [CrossRef]
  62. Liu, C.; Berry, P.M.; Dawson, T.P.; Pearson, R.G. Selecting thresholds of occurrence in the prediction of species distributions. Ecography 2005, 28, 385–393. [Google Scholar] [CrossRef]
  63. Swets, J.A. Measuring the accuracy of diagnostic systems. Science 1988, 240, 1285–1293. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  64. Trew, B.T.; Maclean, I.M.D. Vulnerability of global biodiversity hotspots to climate change. Glob. Ecol. Biogeogr. 2021, 30, 768–783. [Google Scholar] [CrossRef]
  65. Purvis, A.; Gittleman, J.L.; Cowlishaw, G.; Mace, G.M. Predicting extinction risk in declining species. Proc. R. Soc. London Ser. B Biol. Sci. 2002, 67, 1947–1952. [Google Scholar] [CrossRef] [Green Version]
  66. Louppe, D.; Oteng-Amoaka, A.A.; Brink, M. Plant resources of tropical Africa. Prota 2008, 7, 562–563. [Google Scholar]
  67. Tchoundjeu, Z.; Atangana, A.R. Irvingia gabonensis (Aubry-Lecomte ex ’Rorke) Baill. In Record from Protabase; Van der Vossen, H.A.M., Mkamilo, G.S., Eds.; Prota: Wageningen, The Netherlands, 2011. [Google Scholar]
  68. Khanum, R.; Mumtaz, A.S.; Kumar, S. Predicting impacts of climate change on medicinal asclepiads of Pakistan using Maxent modeling. Acta Oecol. 2013, 49, 23–31. [Google Scholar] [CrossRef]
  69. Umata, H. Evaluation of Adaptability of Mung Bean Varieties in Moisture Stress of Eastern Harerghe Zone. Agric. Res. Technol. 2018, 13, 555880. [Google Scholar] [CrossRef]
  70. Zare, M.; Dehghani, B.; Alizadeh, O.; Azarpanah, A. The evaluation of various agronomic traits of mungbean (Vigna radiate L.) genotypes under drought stress and non-stress conditions. Int. J. Farming Allied Sci. 2013, 2, 764–770. [Google Scholar]
  71. Karim, A.; Fukamachi, H.; Hidaka, T. Photosynthetic performance of Vigna radiata L. leaves developed at different temperature and irradiance levels. Plant Sci. 2003, 164, 451–458. [Google Scholar]
  72. Keating, K.A.; Gogan, P.J.; Vore, J.M.; Irby, L.R. A simple solar radiation index for wildlife habitat studies. J. Wildl. Manag. 2007, 71, 1344–1348. [Google Scholar] [CrossRef]
  73. Ashinie, S.K.; Tesfaye, B.; Wakeyo, G.K.; Fenta, B.A. Genetic diversity for immature pod traits in Ethiopian cowpea [Vigna unguiculata (L.) Walp.] Landrace collections. Afr. J. Biotechnol. 2020, 19, 171–182. [Google Scholar]
  74. Etana, A.; Tadesse, E.; Mengistu, A.; Hassen, A. Advanced evaluation of cowpea (Vigna unguiculata) accessions for fodder production in the central rift valley of Ethiopia. J. Agric. Extens. Rural Dev. 2013, 5, 55–61. [Google Scholar]
  75. Ayalew, T.; Yoseph, T. Cowpea (Vigna unguiculata L. Walp.): A choice crop for sustainability during the climate change periods. J. Appl. Biol. Biotechnol. 2022, 10, 154–162. [Google Scholar] [CrossRef]
  76. Rogers, A.; Gibon, Y.; Stitt, M.; Morgan, P.B.; Bernacchi, C.J.; Ort, D.R. Increased C availability at elevated carbon dioxide concentration improves N assimilation in a legume. Plant Cell Environ. 2006, 29, 1651–1658. [Google Scholar] [CrossRef]
  77. Magadza, C. Climate change impacts and human settlements in Africa: Prospects for adaptation. Environ. Monit Assess 2000, 61, 193–205. [Google Scholar] [CrossRef]
  78. Hungria, M.; Franco, A.; Sprent, J. New sources of high-temperature tolerant rhizobia for Phaseolus vulgaris L. Plant Soil 1993, 149, 103–109. [Google Scholar] [CrossRef]
  79. McNamara, A.; Hill, W. UV-B irradiance gradient affects photosynthesis and pigments but not food quality of perihyton. Fresh Water Biol. 2000, 43, 649–662. [Google Scholar] [CrossRef]
  80. Boukar, O.; Belko, N.; Chamarthi, S.; Togola, A.; Batieno, J.; Fatokun, C. Cowpea (Vigna unguiculata): Genetics, genomics and breeding. Plant Breed. 2018, 138, 415–424. [Google Scholar] [CrossRef]
Sustainability 15 00600 g001 550
Figure 1. The results of the jackknife test of variable relevance on test data for both crops under the current climate change scenario in Ethiopia. (A,B) refers variable importance for V. radiata and (C,D) for V. unguiculata, respectively.
Figure 1. The results of the jackknife test of variable relevance on test data for both crops under the current climate change scenario in Ethiopia. (A,B) refers variable importance for V. radiata and (C,D) for V. unguiculata, respectively.
Sustainability 15 00600 g001
Sustainability 15 00600 g002 550
Figure 2. The receiver operating characteristic (ROC) curve for the identical data, averaged over the duplicate runs for V. radiata (A) and V. unguiculata (B), in the context of the current Ethiopian situation.
Figure 2. The receiver operating characteristic (ROC) curve for the identical data, averaged over the duplicate runs for V. radiata (A) and V. unguiculata (B), in the context of the current Ethiopian situation.
Sustainability 15 00600 g002
Sustainability 15 00600 g003a 550Sustainability 15 00600 g003b 550
Figure 3. Maps indicating the current (A) and future (B–D), suitable habitats of V. radiata in Ethiopia.
Figure 3. Maps indicating the current (A) and future (B–D), suitable habitats of V. radiata in Ethiopia.
Sustainability 15 00600 g003aSustainability 15 00600 g003b
Sustainability 15 00600 g004a 550Sustainability 15 00600 g004b 550
Figure 4. Maps indicating the current (A) and predicted (B–D) potential suitable habitats of V. unguiculata in Ethiopia.
Figure 4. Maps indicating the current (A) and predicted (B–D) potential suitable habitats of V. unguiculata in Ethiopia.
Sustainability 15 00600 g004aSustainability 15 00600 g004b
Sustainability 15 00600 g005 550
Figure 5. Relationships between major environmental indicators and the likelihood of V. radiata occurrence in Ethiopia: Bio 16 is precipitation of the wettest quarter in mm (A), Srad is solar radiation index (B), Bio4 is temperature seasonality (C), and Bio15 is precipitation seasonality (D).
Figure 5. Relationships between major environmental indicators and the likelihood of V. radiata occurrence in Ethiopia: Bio 16 is precipitation of the wettest quarter in mm (A), Srad is solar radiation index (B), Bio4 is temperature seasonality (C), and Bio15 is precipitation seasonality (D).
Sustainability 15 00600 g005
Sustainability 15 00600 g006 550
Figure 6. Relationships between key environmental variables and the likelihood of V. unguiculata being present in Ethiopia: Bio 16 is precipitation of the wettest quarter in mm (A), Bio4 is temperature seasonality (B), Bio7 is temperature annual range (C), and Bio17 is precipitation of the driest quarter in mm (D).
Figure 6. Relationships between key environmental variables and the likelihood of V. unguiculata being present in Ethiopia: Bio 16 is precipitation of the wettest quarter in mm (A), Bio4 is temperature seasonality (B), Bio7 is temperature annual range (C), and Bio17 is precipitation of the driest quarter in mm (D).
Sustainability 15 00600 g006
Table
Table 1. Contribution of environmental factors to the MaxEnt modeling for cowpea and mungbean in Ethiopia.
Table 1. Contribution of environmental factors to the MaxEnt modeling for cowpea and mungbean in Ethiopia.
Contribution Percentage of Environmental Variables
SpeciesPeriodsRCPsBio 2Bio 3Bio 4Bio 8Bio 15Bio16Bio17Bio18Bio19SRIElv
Vigna radiataCurrent-2.20.27.56.58.132.93.03.37.319.59.5
20502.62.80.26.86.75.435.02.75.15.919.110.3
4.52.00.110.15.15.932.33.77.37.416.99.2
8.52.20.29.06.25.836.33.46.65.318.56.4
20702.64.50.18.57.06.331.02.66.34.421.08.2
4.51.30.88.87.46.930.43.07.07.017.510.0
8.51.40.710.46.56.337.34.07.64.315.16.4
PeriodsRCPsBio3Bio4Bio7Bio8Bio16Bio17Bio18Bio19SRIElv
Vigna unguiculataCurrent-0.419.515.41.629.414.10.812.60.95.3
20502.60.319.414.52.031.313.10.612.50.65.9
4.50.319.613.82.036.112.00.89.11.05.3
8.50.418.915.50.533.213.90.810.50.85.5
20702.60.917.915.11.335.912.80.88.50.46.4
4.51.118.514.91.433.312.10.512.80.74.8
8.50.720.513.01.835.213.41.07.51.26.0
Table
Table 2. Predicted potential suitable areas (ha) of V. radiata and V. unguiculata in Ethiopia.
Table 2. Predicted potential suitable areas (ha) of V. radiata and V. unguiculata in Ethiopia.
SpeciesPeriodsRCPsHighly Suitable *Moderately Suitable *Less Suitable *Unsuitable *
Vigna radiataCurrent-1,909,414.094,837,553.3312,735,453.5993,743,817.41
20502.61,749,186.284,777,447.9614,382,120.0592,316,907.18
4.52,057,643.945,350,872.3714,254,604.1791,562,882.39
8.51,927,019.964,700,236.6314,451,621.5192,147,319.25
20702.61,610,365.244,426,141.1713,099,459.9194,089,793.06
4.51,899,146.984,590,284.0313,195,068.5293,541,422.19
8.51,972,956.945,274,628.7914,957,770.7591,020,754.08
Vigna unguiculataCurrent-1,439,051.113,473,013.605,907,325.59102,408,727.68
20502.61,408,126.913,016,977.975,868,713.78102,934,348.21
4.51,543,539.112,964,740.655,561,656.80103,158,174.98
8.51,408,407.032,909,946.925,919,024.48102,990,437.85
20702.61,518,309.153,137,969.176,483,449.87102,088,452.69
4.51,534,859.093,097,101.095,978,729.21102,616,632.11
8.51,369,446.982,953,955.836,101,788.30102,803,076.12
* Refers to the hectares of estimated area to the relevant habitat type area.
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

Kagnew, B.; Assefa, A.; Degu, A. Modeling the Impact of Climate Change on Sustainable Production of Two Legumes Important Economically and for Food Security: Mungbeans and Cowpeas in Ethiopia. Sustainability 2023, 15, 600. https://doi.org/10.3390/su15010600

AMA Style

Kagnew B, Assefa A, Degu A. Modeling the Impact of Climate Change on Sustainable Production of Two Legumes Important Economically and for Food Security: Mungbeans and Cowpeas in Ethiopia. Sustainability. 2023; 15(1):600. https://doi.org/10.3390/su15010600

Chicago/Turabian Style

Kagnew, Birhanu, Awol Assefa, and Asfaw Degu. 2023. "Modeling the Impact of Climate Change on Sustainable Production of Two Legumes Important Economically and for Food Security: Mungbeans and Cowpeas in Ethiopia" Sustainability 15, no. 1: 600. https://doi.org/10.3390/su15010600

APA Style

Kagnew, B., Assefa, A., & Degu, A. (2023). Modeling the Impact of Climate Change on Sustainable Production of Two Legumes Important Economically and for Food Security: Mungbeans and Cowpeas in Ethiopia. Sustainability, 15(1), 600. https://doi.org/10.3390/su15010600

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