
The impacts of climate change are projected to become more intense and frequent. One of the indirect impacts of climate change is food insecurity. Agriculture in Pakistan, measured fourth best in the world, is already experiencing visible adverse impacts of climate change. Among many other food sources, potato crop remains one of the food security crops for developing nations. Potatoes are widely cultivated in Pakistan. To assess the impact of climate change on potato crop in Pakistan, it is imperative to analyze its distribution under future climate change scenarios using Species Distribution Models (SDMs). Maximum Entropy Model is used in this study to predict the spatial distribution of Potato in 2070 using two CMIP5 models for two climate change scenarios (RCP 4.5 and RCP 8.5). 19 Bioclimatic variables are incorporated along with other contributing variables like soil type, elevation and irrigation. The results indicate slight decrease in the suitable area for potato growth in RCP 4.5 and drastic decrease in suitable area in RCP 8.5 for both models. The performance evaluation of the model is based on AUC. AUC value of 0.85 suggests the fitness of the model and thus, it is applicable to predict the suitable climate for potato production in Pakistan. Sustainable potato cultivation is needed to increase productivity in developing countries while promoting better resource management and optimization.
Citation: Tayyaba Khalil, Saeed A. Asad, Nusaiba Khubaib, Ayesha Baig, Salman Atif, Muhammad Umar, Jürgen P. Kropp, Prajal Pradhan, Sofia Baig. Climate change and potential distribution of potato (Solanum tuberosum) crop cultivation in Pakistan using Maxent[J]. AIMS Agriculture and Food, 2021, 6(2): 663-676. doi: 10.3934/agrfood.2021039
[1] | José Héctor Gálvez, Helen H. Tai, Noelle A. Barkley, Kyle Gardner, David Ellis, Martina V. Strömvik . Understanding potato with the help of genomics. AIMS Agriculture and Food, 2017, 2(1): 16-39. doi: 10.3934/agrfood.2017.1.16 |
[2] | Johan Sukweenadhi, Eloqui Viectorica Wiranata, Ida Bagus Made Artadana, Kang-Se Chang . Isolation and in vitro screening of plant growth promoting bacteria from rhizosphere and root tissues of potato tuber (Solanum tuberosum L.). AIMS Agriculture and Food, 2023, 8(4): 1028-1037. doi: 10.3934/agrfood.2023055 |
[3] | A. Burhan, G. Rasul, T. Qadir, S. Hussain, M. Saqib, S.A.A. Bukhari . Environmental policies to protect pollinators: attributes and actions needed to avert climate borne crisis of oil seed agriculture in Pakistan. AIMS Agriculture and Food, 2017, 2(3): 233-250. doi: 10.3934/agrfood.2017.3.233 |
[4] | Abbas Ali Chandio, Yuansheng Jiang, Abdul Rehman, Rahman Dunya . The linkage between fertilizer consumption and rice production: Empirical evidence from Pakistan. AIMS Agriculture and Food, 2018, 3(3): 295-305. doi: 10.3934/agrfood.2018.3.295 |
[5] | J.M.A. Duncan, J. Dash, E.L. Tompkins . Observing adaptive capacity in Indian rice production systems. AIMS Agriculture and Food, 2017, 2(2): 165-182. doi: 10.3934/agrfood.2017.2.165 |
[6] | Heru Irianto, Mujiyo Mujiyo, Aulia Qonita, Ato Sulistyo, Erlyna Wida Riptanti . The development of jarak towo cassava as a high economical raw material in sustainability-based food processing industry. AIMS Agriculture and Food, 2021, 6(1): 125-141. doi: 10.3934/agrfood.2021008 |
[7] | Supriyono Loekito, Afandi, Auliana Afandi, Naomasa Nishimura, Hiroyuki Koyama, Masateru Senge . Gypsum supplies calcium to Ultisol soil and its effect on Phytophthora nicotianae, pineapple (Ananas comosus) growth, yield and fruit quality in lower single row bed under climate change issue. AIMS Agriculture and Food, 2022, 7(3): 721-736. doi: 10.3934/agrfood.2022044 |
[8] | John Samelis, Athanasia Kakouri . Major technological differences between an industrial-type and five artisan-type Greek PDO Galotyri market cheeses as revealed by great variations in their lactic acid microbiota. AIMS Agriculture and Food, 2019, 4(3): 685-710. doi: 10.3934/agrfood.2019.3.685 |
[9] | Fernando Arias, Maytee Zambrano, Kathia Broce, Carlos Medina, Hazel Pacheco, Yerenis Nunez . Hyperspectral imaging for rice cultivation: Applications, methods and challenges. AIMS Agriculture and Food, 2021, 6(1): 273-307. doi: 10.3934/agrfood.2021018 |
[10] | Nelson Mpumi, Revocatus L. Machunda, Kelvin M. Mtei, Patrick A. Ndakidemi . Insecticidal Efficacy of Syzygium aromaticum, Tephrosia vogelii and Croton dichogamus Extracts against Plutella xylostella and Trichoplusia ni on Brassica oleracea crop in Northern Tanzania. AIMS Agriculture and Food, 2021, 6(1): 185-202. doi: 10.3934/agrfood.2021012 |
The impacts of climate change are projected to become more intense and frequent. One of the indirect impacts of climate change is food insecurity. Agriculture in Pakistan, measured fourth best in the world, is already experiencing visible adverse impacts of climate change. Among many other food sources, potato crop remains one of the food security crops for developing nations. Potatoes are widely cultivated in Pakistan. To assess the impact of climate change on potato crop in Pakistan, it is imperative to analyze its distribution under future climate change scenarios using Species Distribution Models (SDMs). Maximum Entropy Model is used in this study to predict the spatial distribution of Potato in 2070 using two CMIP5 models for two climate change scenarios (RCP 4.5 and RCP 8.5). 19 Bioclimatic variables are incorporated along with other contributing variables like soil type, elevation and irrigation. The results indicate slight decrease in the suitable area for potato growth in RCP 4.5 and drastic decrease in suitable area in RCP 8.5 for both models. The performance evaluation of the model is based on AUC. AUC value of 0.85 suggests the fitness of the model and thus, it is applicable to predict the suitable climate for potato production in Pakistan. Sustainable potato cultivation is needed to increase productivity in developing countries while promoting better resource management and optimization.
Recently, the climatic change has become the cause of concern all over the globe due to unprecedented occurring of extreme weather events. Average temperatures have undergone an increase of 0.85 ℃ in the last century and it is further predicted to increase to a maximum of 8 ℃ by 2100 [1]. Less developed nations and low-income communities based on agricultural economy are more susceptible; thus, they are more likely to get impacted with climate change [2].
In coming years, the most suffering communities from climate change will be the ones in Global South, particularly Africa and South Asia [1]. Out of all the other sectors, agriculture is directly affected by global warming and climatic changes due to altered weather conditions [3,4]. The overall global climate change, mainly an increase in temperature has negatively influenced the natural ecosystems and agriculture [5]. There are complex means through which climate change influences agriculture and food production. Alteration in precipitation, extreme heat and drought conditions due to changes in climate are directly responsible for the change in food production patterns [6]. This also results in changes in growth and distribution of incomes with increased demand for agriculture. Future climatic conditions are predicted to be more variable than current climatic conditions as further severe extreme events of heat, droughts, cyclones and floods are expected. These events will be responsible for more fluctuating agricultural yields and thus would lead to food insecurity [7].
Pakistan is an agriculture-based country with an area of about 79.6 million hectares (Mha). The climate is arid and semi-arid. Of all the total area, about 22.1 Mha is suitable for crop cultivation. Of the cultivated land 72% is irrigated through surface and subsurface irrigation system and nearly 28% through rain-fed agriculture. According to World Bank, total agricultural land of Pakistan in 2017 was 202,000 km2 among top four best agriculture areas globally, but unfortunately this region is experiencing reduction in crop area due to several reasons, for example people migrating to cities, large housing societies, construction of roads and highways, and shortage of water for irrigation. Furthermore, the country has suffered through several natural disasters such as floods, earthquakes, landslides and droughts.
Despite its vibrant agriculture-based economy, Pakistan is more likely to be impacted with food shortage and agricultural down fall [8,9]. Thus, lack in agricultural output will be detrimental and a major cause of food insecurity risk that might impact the health and economy in future [10]. Pakistan is ranked poorly on Global Hunger Index that emphasizes the seriousness of the issue [11]. According to estimates, around 127 million people in Pakistan suffer from food insecurity [12]. Mostly children are malnourished and live below poverty line. With such depleting figures, extreme weather events affecting major food crops of Pakistan including wheat, maize, rice and potato, is extremely alarming.
Potato (Solanum tuberosum) can grow in warmer as well as cooler climates, however it cannot thrive in harsh climatic conditions such drought, high temperature and high humidity [13]. In many countries Potato is grown as a major crop under different climatological zones, such as temperate regions, the sub-tropics and tropics, under very different agro-ecological conditions, lowlands and highlands. Presently, the potato crop is cultivated in about 130 countries and there are nearly 5000 potato varieties worldwide [14]. With regard to the global food consumption, potato ranks third as an important non-grain crop. FAO has declared potato as the food security crop because it provides nutritious food to poor and hungry where the world is facing issues with food accessibility [15]. After wheat, rice, and maize, potato is now the fourth most important global food crop due to better yield and high nutritive value. Potatoes are a good source of carbohydrates, vitamins, and minerals. Vitamins includes niacin, thiamin, riboflavin and vitamin C [16]. The economically significant part of the potato is tuber, site for storage of carbohydrates and consists of about 77% water, 20% indigestible carbohydrate [14].
China is the biggest producer of potatoes worldwide, with about one third of the world's potatoes produced in China and India. According to FAO estimates, in 2019, over 370 million metric tons of potatoes were produced worldwide, a substantial increase from a production volume of 333.6 million tons in 2010. However, this region is ecologically fragile, and agricultural production is vulnerable to climate change [17]. Hinjmans [18] studied the effect of climate change on global potato production using LINTUL simulation model. Potential yields were calculated for current (1961–1990) and projected (2010–2039 and 2040–2069) conditions. The study revealed that at high latitudes, global warming will likely lead to a shift of the location of potato production. Model predicted changes in potato yield to be relatively small, and sometimes positive. In the tropics, where there is little temperature change during the year, there is not much scope for adaptation of potato production. In India and Bangladesh where potato is a winter crop already grown in the coldest season, climate change might reduce potato yield. It is predicted that an overall fall of 9.56% at national level if needed steps are not taken to mitigate the effects of climate change [19].
Potatoes are grown over a large area of Pakistan with the production of roughly 4.1 million tons. They are largely grown in the central and northern plains of Punjab and KPK. Some parts or districts of Baluchistan in west supporting the production of potato, include Pishin, Killa Saifulla, Kalat and Gilgit district in Gilgit-Baltistan. Potato in Pakistan is not yet being produced to its maximum capacity as compared to neighboring India and Bangladesh [20] however, to assess the impact of climate change on potato crop in Pakistan, especially in the areas which are rain fed, it is essential to analyze its distribution under future climate change scenarios. This can be achieved through the use of various modeling techniques. Species Distribution Models (SDMs) use a combination of numerical tools with species presence data along with various environmental factors. Maxent (Maximum Entropy) is one such species distribution model which forecasts the distribution of a given species using only its presence data [21], capable of modeling crop suitability [22] and provides the most precise distribution function based on best entropy [23,24]. The model has been widely used all over the world to predict changes in the geographical distribution of species under climate change [13,25]. Based on the wide use of Maxent for species distribution, this study was designed with the main objective to predict potato crop distribution across Pakistan by comparing two climate change scenarios RCP 4.5 and RCP 8.5 with two CMIP5 models.
Pakistan, with a latitude of 30.3753° N and longitude of 69.3451° E, is the study area of this research work. Pakistan is located in South Asia with an area of about 881,913 km2. It is bordered by countries like India, Afghanistan, Iran and China. It also has coastline along Arabian Sea in the South. Pakistan is a country with a diversified temperature and precipitation patterns. Very high altitudes modify the climate in the cold, snow-covered northern mountains; temperatures on the Balochistan plateau are somewhat higher (Figure 1a). Along the coastal strip, the climate is modified by sea breeze. In the rest of the country, temperatures reach great heights in the summers. Surface-water resources in Pakistan are based on the flows of the Indus River and its tributaries. Climate is not uniform over the Indus river basin. It varies from subtropical arid and semi-arid to temperate sub-humid on the plains of Sindh and Punjab provinces to alpine in the mountainous highlands of the north. Potatoes are mostly cultivated in Pakistan where irrigation is available. Punjab leads in potato production with 83% of total production followed by KPK with 10%. Baluchistan contributes 6% to the potato production whereas only 1% potatoes are cultivated in Sindh.
Collection of occurrence records for Solanum tuberosum was done primarily from report of Statistical Bureau of Pakistan [26]. Unpublished data was taken from Potato Research Centers in various cities, Sahiwal, Sialkot and Murree in Punjab province in December 2018. GPS points were gathered during field visits to Abbottabad, Mansehra and Feteh Jang in KPK province. The data gathered from these sources was further processed, where areas with high potato yield were considered and areas with low potato yields were discarded. This was done by calculating average of production/area for the years 2005–2012 and data points for locations with yield (production/area) above average were considered for input in the Maxent model. To correct for clustered occurrence records that affect SDM predictions, we used a raster grid size of 0.05 degrees of latitude, which are equivalent to about 5 kilometers square. This method randomly selects a single point from each grid square. The total occurrence data points selected were 58 (Figure 1b).
19 Bioclimatic variable layers (BIO 1 to BIO 19) (Table 1) for two models CCSM4 and MPI-ESM-LR were downloaded from WorldClim website (https://www.worldclim.org/) using R-program [27]. This website provides current (1960–1990), high resolution WorldClim climate data and future climate projections from global climate models (GCMs) for four representative concentration pathways (RCPs).
Code | Bioclimatic variables name |
BIO1 | Annual Mean Temperature |
BIO2 | Mean Diurnal Range (Mean of monthly (max temp-min temp)) |
BIO3 | Isothermality (BIO2/BIO7) (*100) |
BIO4 | Temperature Seasonality (standard deviation *100) |
BIO5 | Max Temperature of Warmest Month |
BIO6 | Min Temperature of Coldest Month |
BIO7 | Temperature Annual Range (BIO5-BIO6) |
BIO8 | Mean Temperature of Wettest Quarter |
BIO9 | Mean Temperature of Driest Quarter |
BIO10 | Mean Temperature of Warmest Quarter |
BIO11 | Mean Temperature of Coldest Quarter |
BIO12 | Annual Precipitation |
BIO13 | Precipitation of Wettest Month |
BIO14 | Precipitation of Driest Month |
BIO15 | Precipitation Seasonality (Coefficient of Variation) |
BIO16 | Precipitation of Wettest Quarter |
BIO17 | Precipitation of Driest Quarter |
BIO18 | Precipitation of Warmest Quarter |
BIO19 | Precipitation of Coldest Quarter |
These models were selected to explain variation in model outputs, and any uncertainty in single model predictions. MPI-ESM-LR is low-bias and high-sensitivity model and has least interdependency to CCSM4 [28]. The future projections selected for the analysis along with current bioclimatic layers, included two climate change scenarios RCP 4.5 and RCP 8.5. RCP 4.5 represents a future in which some collective action is taken to limit GHG emissions, with global annual average temperatures increasing 2.4 ℃ by 2100. RCP 8.5 is closer to a scenario in which no actions are taken to reduce emissions, and global annual average temperatures increase 4.3 ℃ by 2100. The average prediction by these climate models is that annual average temperatures in South Asia will increase 1.6 ℃ by 2050 under RCP 4.5, and 2.2 ℃ under RCP 8.5. As these increases are relative to 1981–2010 conditions [29], hence these two RCPs were selected for the analysis. Each layer had a spatial resolution of 30sec and was converted to ASC format in R-program as required by Maxent [30]. Additional covariates were also downloaded from FAO website including Soil type and Irrigation (from FAO.org). Digital Elevation Model Data was downloaded from SRTM DEM 30 m resolution (https://gisgeography.com/free-global-dem-data-sources/). These covariates were incorporated in the analysis along with the bioclimatic layers to evaluate their impact on crop distribution.
Occurrence data file in "txt" format along with 19 current bioclimatic variable and covariates (soil type, irrigation and elevation) ASCII grid files with same resolution were incorporated in R-program The R package dismo [31] was used to run Maxent model. Maxent has two main modifiable parameters: (1) regularization multiplier and (2) feature classes. The regularization multiplier sets how focused or closely-fitted the output distribution is–a smaller value than the default of 1.0 will result in a more localized output distribution that is a closer fit to the given presence records. A larger regularization multiplier will give a more spread out, less localized prediction [32]. The "regularization multiplier" parameter was set to 1 for this study. "Feature class" is an expanded set of transformations of the original covariates [33] which constrains the computed probability distribution. The available feature types include linear, quadratic, product, threshold, and hinge [30]. The selection of feature class is related to the number of species occurrence points. By default, the program restricts the model to simple features if few samples are available [33]. Model generally uses linear feature; quadratic feature is used with at least 10 samples; hinge is used with at least 15 samples; threshold and product with at least 80. Combinations of all feature classes are usually used as default [30,33,34]. 1000 Background points were randomly selected all over the study area to generate the suitable area maps. 5-fold cross-validation method was used which means Maxent randomly partitioned the occurrence data into five independent sets and repeated the model building process five times. Thus, the maps for suitable habitat for current climatic conditions were generated. Same procedure was repeated for future scenarios (RCP 4.5 and RCP 8.5) with future bioclimatic variable layers but the covariates were kept constant. Maps for future scenarios were generated and suitable areas for potato production were calculated for current climate and future scenarios. Maps for suitable area for all provinces were generated. Model was evaluated on the basis of area under receiver operating characteristic curve (AUC) which is a threshold-independent statistic commonly used to evaluate species distribution models. Maxent calculated AUC values for both the training and test data.
Satellite images of Landsat 8, level 1 for October and November 2018 were downloaded from USGS website. The images used to validate the exact locations of potato production using NDVI range. For this reason, band 4 and band 5 from all the files were extracted and separated. After mosaicking and removal of cloud cover further atmospheric correction was done and thus, NDVI for the final product was calculated. The final mosaicked file was further clipped by masking using R-program. After processing of the files, classification was done to get the locations of potato using NDVI range 3–6 extracted from literature. Map for NDVI range for potato was generated and thus validation was done against the current distribution maps.
Potato climatic distribution is dependent on a certain low temperature range, availability of enough water, soil pH and fertile soil that supports its growth. These conditions might change in future due to changing climatic scenarios thus leading to the reduction in suitable area for potato cultivation.
Using accumulative output of the real outcomes from the Maxent model [35], raw maps were generated which showed highly suitable, moderately suitable, least suitable and unsuitable climate areas for potato production in Pakistan (Figure 2a).
NDVI based map (Figure 2b) validated the presence locations of the potato in field-based map (Figure 1) except for some area in Sindh Province. These areas might have some other crops or vegetation such as Maize (NDVI Range 0.25–0.75) which falls under the same broad range of potato and widely grown in Sindh under the same weather conditions [36].
Another approach used for the evaluation of model was by defining thresholds. In this approach, threshold selection is done for the establishment of the locations that are considered suitable and non-suitable for vegetation. Minimizing the specificity and maximizing the sensitivity are the key criteria that define the thresholds [37,38]. The threshold map (Figure 2c) was generated to simplify the output of Maxent model and present it in binary form (suitable and non-suitable climate locations). The area calculated as suitable climate for potato production under current scenario was about 265,016 km2. Punjab province was found to be the most suitable area to grow potato. The suitable area for Punjab was about 152,411 km2, followed by KPK with an area of 56,392 km2. The share of Baluchistan and Sindh was about 42,509 km2 and1239 km2 respectively. Suitable area under the territory of Gilgit Baltistan was calculated to be about 5,534 km2. Out of all the provinces, Sindh had the least suitable climate to grow potato as the temperature and climate of Sindh even in winters is not suitable for Potato cultivation.
Jackknife test was used to estimate the variable importance. The jackknife test is shown by a bar graph of pink and blue bars with red vertical dotted line (Figure 3). The size of pink bar denotes the score with one of the climatic indices; the length of the bar indicates the importance of that variable, more the length, more important is that variable. After running Jackknife for variable importance, the most important variables were found to be isothermality, soil type, mean temperature of the driest quarter, precipitation of wettest quarter and annual precipitation.
Assessment and evaluation of Maxent model can be done using area under the receiver operating characteristic (ROC) curves (AUC). ROC is plotted after plotting the sensitivity values (True positive rate) on y-axis and 1-specificity values (false positive rate) on x-axis (Figure 4). The area under the curve (AUC) measures the overall accuracy of the model. Higher AUC values depict better performance of the model and AUC lower than 0.5 suggests no discrimination [37]. If AUC values are more than 0.78, then the model has performed well [39]. In case of current distribution of potato using Maxent, the results of AUC 0.85 showed that the model performed well, and results showed strong likelihood of crop distribution.
Future CMIP5 models, CCSM4 (Model 1) and MPI-ESM-LR (Model 2) were used for this study. Future distribution of potato was predicted for 2070 under RCP 4.5 and RCP 8.5. The results for RCP 4.5 showed a slight decrease in the suitable area for potato cultivation in comparison with the area shown in current distribution. In model 1 (Figure 5a) the suitable area calculated was about 237,132 km2 where the area suitable in Punjab was found to be 149,193 km2 followed by KPK which was 43,053 km2. Suitable area in Baluchistan was predicted to be 35,300 km2 and in Sindh it was predicated to be about 518 km2. Gilgit Baltistan had about 3,556 km2 area suitable for potato production. In model 1 the suitable area reduction for RCP 4.5 was about 10.5% for whole Pakistan. For model 2 (Figure 5b) the suitable area calculated was about 252,449 km2 in which suitable area in Punjab was found to be 145,200 km2 followed by KPK which was 54,390 km2. Suitable area in Baluchistan was predicted to be 36,602 km2 and area in Sindh was about 27, 23 km2. Gilgit Baltistan had an area about 6,193 km2. In model 2 the suitable area reduction for RCP 4.5 was about 4.7% as compared to the current area.
In RCP 8.5 there was a drastic decrease in the suitable area for potato cultivation. The suitable area calculated was about 76,608 km2 in model 1 (Figure 5c) in which suitable area in Punjab was found to be 47,411 km2 followed by KPK which was 210,257 km2. Suitable area in Baluchistan was predicted to be 46, 89 km2 and in Sindh, it was found to be only 29 km2. Gilgit Baltistan had about 1,032 km2 area suitable for potato production. For RCP 8.5 the suitable area reduced was about 71% with respect to the current suitable area. In model 2 (Figure 5d) the suitable area for whole country was about 154,598 km2. Area found suitable in Punjab was 106,086 km2 followed by KPK which was 31,440 km2. Suitable area in Baluchistan was predicted to be 11,183 km2 and area in Sindh was predicated be only 58 km2. Gilgit Baltistan had about 3,460 km2 area suitable for potato cultivation. For RCP 8.5 the suitable area reduced was about 55.5% with respect to the current suitable area.
Potato is one of the basic food crops grown in Pakistan. It is very important to see its future distribution in order to cope with future food insecurity. Maxent modeling was used to analyze its future distribution with two future models and both models predicted decrease in the suitable area for Potato cultivation in 2070. The predicted highly suitable cultivation regions were located mainly in Punjab and KPK. In model 1 the suitable area reduction for RCP 4.5 was about 10.5% and for RCP 8.5 the suitable area was reduced was about 71% with respect to the current suitable area. In model 2 the suitable area reduction for RCP 4.5 was about 4.7% and for RCP 8.5 the suitable area reduced was about 55.5% with respect to the current suitable area. Important environmental factors shaping suitable habitat of Potato for current climate included isothermality, soil type, temperature of driest quarter, precipitation of wettest quarter and annual precipitation.
In this research, modeling of the current and future distribution of Potato cultivation in Pakistan was done. Area Under the Curve value of 0.85 for current distribution indicated the robustness of the model [40]. Under the current and future scenarios, classification based on potato production for the highly suitable, moderately suitable, less suitable and non-suitable area was done. The findings for the current scenario are supported by the other studies [23] for the cultivation of Ceylon Spinach and by [24] for the cultivation of Vegetable Roselle in which the results showed that Maxent Modeling Approach can be used as a tool to study the climate suitability of different vegetable crop species.
The results for the future distribution of Solanum Tuberosum coincided with the findings in the literature [13,41,42,43,44] in which a temperature sensitive species underwent drastic decrease in the suitable habitat under the RCP 8.5 for year 2070 and suggested that one of the key factors shaping the distribution of the species was annual precipitation [43]. Another factor that was very important in the growth of potato was soil type, without suitable soil type and pH range, the growth of the species was not possible in any scenario. In a study on one of the mushroom species, Tricholoma matsutake was found to be highly dependent on the suitable soil type and no yield was predicated without favorable soil conditions even though the climate and topography were in the favor of its production [13].
With sufficient water supply, studies through free-air CO2 enrichment systems (FACE) have reported a yield increase in potatoes [45,46,47]. Many studies are going on to analyze the interaction between CO2 increase and warmer temperatures in field crops [46]. Currently, only modeling provides some insight about their combined effect on potato growth and environment. However, similar results were found as this study with drastic decrease in potato yield using RCP8.5 [48].
Based on current and future potato cultivation distribution model predictions, the study showed a decline in Potato yield in both RCP 4.5 and RCP 8.5 climate change scenarios. The magnitude of uncertainty for RCPs and GCMs increases towards the end of the century. Pakistan is going to experience large year-to-year climate variability causing low yields in some seasons. Climate change enhances variability; however, uncertainty is also large, partly due to GCM scenarios used for this region and partly due to variability in growing conditions. Pakistan, that most depend on potato production for food security, is also a region least able to invest in agriculture and is the hardest hit by climate change impacts. Developing strategies and tools to comprehensively understand the impact of climate change and evolve possible adaptation measures in horticultural crops is less understood. Present research can be used for the future assessment of other non-grain crops and thus the estimates can help to generate more yields by adopting ways to enhance growth and cultivation of similar crops in the future when food insecurity will be rather challenging. Without better projections of crops in the future, there would be more chances of food shortage and famine.
The authors gratefully acknowledge the research support funding from the German Federal Ministry of Education and Research (BMBF) for the SUSFOOD project (grant agreement No. 01DP17035).
All authors declare no conflicts of interest in this paper.
[1] | IPCC (2019) Climate change and land: An IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems. Available from: https://www.ipcc.ch/srccl/download. |
[2] |
Hertel TW, Burke MB, Lobell DB (2010) The poverty implications of climate-induced crop yield changes by 2030. Global Environ Change 20: 577-585. doi: 10.1016/j.gloenvcha.2010.07.001
![]() |
[3] |
Rosenzweig C, Parry ML (1994) Potential impact of climate change on world food supply. Nature 367: 133-138. doi: 10.1038/367133a0
![]() |
[4] | Mbow C, Rosenzweig C, Barioni LG, et al. (2019) Food security. In: Climate Change and Land: an IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems. |
[5] | Stocker TF, Qin D, Plattner GK, et al. (2014) Technical Summary. In: Climate change 2013: The physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. United Kingdom, New York: Cambridge University Press. |
[6] |
van Meijl H, Havlik P, Lotze-Campen H, et al. (2018) Comparing impacts of climate change and mitigation on global agriculture by 2050. Environ Res Lett 13: 064021. doi: 10.1088/1748-9326/aabdc4
![]() |
[7] | Schmidhuber J (2015) "The Food Equation": Taking a long/term view on world agriculture, climate change and food security. Avaliable from: https://ojs.macsur.eu/index.php/Reports/article/view/SP4-15/76. |
[8] | Chaudhri T (2017) The China-Pakistan Economic Corridor (CPEC). Ohio USA: The Ohio State University. Avaliable from: http://cpec.gov.pk. |
[9] | Pradhan P, Fischer G, van Velthuizen H, et al. (2015) Closing yield gaps: How sustainable can we be? PLoS ONE 10: e0129487. |
[10] |
Baig N, He C, Khan S, et al. (2019) CPEC and food security: Empirical evidence from Pakistan. J Public Adm Governance 9: 191-208. doi: 10.5296/jpag.v9i1.14309
![]() |
[11] | Von Grebmer K, Bernstein J, Patterson F, et al. (2019) 2019 global hunger index: The challenge of hunger and climate change. Dubli, Bonn: Welt hungerh ilfen. |
[12] | World Food Programme (2017) Year in review 2018, WFP: The last resort, the first defence. Avaliable from: https://www.wfp.org/publications/wfp-year-review-2018. |
[13] |
Guo YL, Li X, Zhao ZF, et al. (2017) Prediction of the potential geographic distribution of the ectomycorrhizal mushroom Tricholoma matsutake under multiple climate change scenarios. Sci Rep 7: 46221. doi: 10.1038/srep46221
![]() |
[14] |
Zaheer K, Akhtar MH (2016) Potato production, usage, and nutrition-A review. Crit Rev Food Sci Nutr 56: 711-721. doi: 10.1080/10408398.2012.724479
![]() |
[15] |
Devaux A, Kromann P, Ortiz O (2014) Potatoes for sustainable global food security. Potato Res 57: 185-199. doi: 10.1007/s11540-014-9265-1
![]() |
[16] |
Cotton PA, Subar AF, Friday JE, et al. (2004) Dietary sources of nutrients among US adults, 1994 to 1996. J Am Diet Assoc 104: 921-930. doi: 10.1016/j.jada.2004.03.019
![]() |
[17] |
Li Q, Zhang SQ (2020) Impacts of recent climate change on potato yields at a provincial scale in Northwest China. Agronomy 10: 426. doi: 10.3390/agronomy10030426
![]() |
[18] |
Hijmans RJ (2003) The effect of climate change on global potato production. Am J Pot Res 80: 271-279. doi: 10.1007/BF02855363
![]() |
[19] | Hussain T (2016) Potatoes: Ensuring food for the future. Adv Plants Agric Res 3: 00117. |
[20] | Majeed A, Muhammad Z (2018) Potato production in Pakistan: Challenges and prospective management strategies-a review. Pakistan J Bot 50: 2077-2084. |
[21] |
Byeon D, Jung S, Lee WH (2018) Review of CLIMEX and MaxEnt for studying species distribution in South Korea. J Asia-Pac Biodivers 11: 325-333. doi: 10.1016/j.japb.2018.06.002
![]() |
[22] |
Heumann BW, Walsh SJ, McDaniel PM (2011) Assessing the application of a geographic presence-only model for land suitability mapping. Ecol Inform 6: 257-269. doi: 10.1016/j.ecoinf.2011.04.004
![]() |
[23] | Reddy M, Sunil N, Pandravada S, et al. (2015) Assessing climate suitability for sustainable vegetable roselle (Hibiscus sabdariffa var. sabdariffa L.) cultivation in India using MaxEnt model. Agr Biol Sci J 1: 62-70. |
[24] | Reddy MT, Begum H, Sunil N, et al. (2015) Mapping the climate suitability using MaxEnt modeling approach for Ceylon spinach (Basella alba L.) cultivation in India. J Agr Sci 10: 87-97. |
[25] |
Singh M (2020) Evaluating the impact of future climate and forest cover change on the ability of Southeast (SE) Asia's protected areas to provide coverage to the habitats of threatened avian species. Ecol Indic 114: 106307. doi: 10.1016/j.ecolind.2020.106307
![]() |
[26] | Agricultural Statistics of Pakistan (2015) Government of Pakistan Statistics Division Pakistan Bureau of Statistics. Available from: https://www.pbs.gov.pk/content/agriculture-statistics. |
[27] | R Core Team (2019) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from: https://www.R-project.org. |
[28] |
Sanderson BM, Knutti R, Caldwell P (2015) Addressing interdependency in a multimodel ensemble by interpolation of model properties. J Climate 28: 5150-5170. doi: 10.1175/JCLI-D-14-00361.1
![]() |
[29] | Mani M, S. Bandyopadhyay, S. Chonabayashi, et al. (2019) South Asia's hotspots: Impact of temperature and precipitation changes on living standards. Washington, DC: World Bank. |
[30] |
Phillips SJ, Dudik M (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31: 161-175. doi: 10.1111/j.0906-7590.2008.5203.x
![]() |
[31] | Hijmans RJ, Phillips S, Leathwick J, et al. (2015) Dismo: Species distribution modelin, Version 1.3-3. Avaliable from: http://CRAN.R-project.org/package=dismo. |
[32] | Phillips SJ (2017) A brief tutorial on Maxent. Avaliable from: https://biodiversityinformatics.amnh.org/open_source/maxent. |
[33] |
Elith J, Phillips SJ, Hastie T, et al. (2011) A statistical explanation of MaxEnt for ecologists. Divers Distrib 17: 43-57. doi: 10.1111/j.1472-4642.2010.00725.x
![]() |
[34] |
Tourne DCM, Ballester MVR, James PMA, et al. (2019) Strategies to optimize modeling habitat suitability of Bertholletia excelsa in the Pan-Amazonia. Ecol Evol 9: 12623-12638. doi: 10.1002/ece3.5726
![]() |
[35] |
Merow C, Smith MJ, Silander JA (2013) A practical guide to MaxEnt for modeling species' distributions: what it does, and why inputs and settings matter. Ecography 36: 1058-1069. doi: 10.1111/j.1600-0587.2013.07872.x
![]() |
[36] | Arain GN (2013) Maize (corn) cultivation in Pakistan. Available from: http://www.valleyirrigationpakistan.com/wp-content/uploads/2012/09/Maize-Cultivation-in-Pakistan1.pdf. |
[37] |
Fielding AH, Bell JF (1997) A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ Conserv 24: 38-49. doi: 10.1017/S0376892997000088
![]() |
[38] |
Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modeling of species geographic distributions. Ecol Model 190: 231-259. doi: 10.1016/j.ecolmodel.2005.03.026
![]() |
[39] | Peavey L (2010) Predicting pelagic habitat with presence-only data using maximum entropy for olive ridley sea turtles in the Eastern Tropical Pacific. Available from: https://dukespace.lib.duke.edu/dspace/bitstream/handle/10161/2247/Peavey_MP_4_30_10-1?sequence=1. |
[40] | Yackulic CB, Chandler R, Zipkin EF, et al. (2013) Presence-only modelling using MAXENT: when can we trust the inferences? Methods Ecol Evol 4: 236-243. |
[41] |
Kumar P (2012) Assessment of impact of climate change on Rhododendrons in Sikkim Himalayas using Maxent modelling: limitations and challenges. Biodivers Conserv 21: 1251-1266. doi: 10.1007/s10531-012-0279-1
![]() |
[42] |
Zhang KL, Yao LJ, Meng JS, et al. (2018) Maxent modeling for predicting the potential geographical distribution of two peony species under climate change. Sci Total Enviro 634: 1326-1334. doi: 10.1016/j.scitotenv.2018.04.112
![]() |
[43] | Abdelaal M, Fois M, Fenu G, et al. (2019) Using MaxEnt modeling to predict the potential distribution of the endemic plant Rosa arabica Crep. in Egypt. Ecol Inform 50: 68-75. |
[44] | Wei B, Wang RL, Hou K, et al. (2018) Predicting the current and future cultivation regions of Carthamus tinctorius L. using MaxEnt model under climate change in China. Glob Ecol Conserv 16: e00447. |
[45] |
Finnan JM, Donnelly A, Jones MB, et al. (2005) The effect of elevated levels of carbon dioxide on potato crops. J Crop Improv 13: 91-111. doi: 10.1300/J411v13n01_06
![]() |
[46] |
Kimball BA (2016) Crop responses to elevated CO2 and interactions with H2O, N, and temperature. Curr Opin Plant Biol 31: 36-43. doi: 10.1016/j.pbi.2016.03.006
![]() |
[47] |
Quiroz R, Ramirez DA, Kroschel J, et al. (2018) Impact of climate change on the potato crop and biodiversity in its center of origin. Open Agric 3: 273-283. doi: 10.1515/opag-2018-0029
![]() |
[48] |
Raymundo R, Asseng S, Robertson R, et al. (2018) Climate change impact on global potato production. Eur J Agron 100: 87-98. doi: 10.1016/j.eja.2017.11.008
![]() |
1. | Shiliang Yang, Huimin Wang, Jinping Tong, Yang Bai, Juha M. Alatalo, Gang Liu, Zhou Fang, Fan Zhang, Impacts of environment and human activity on grid-scale land cropping suitability and optimization of planting structure, measured based on the MaxEnt model, 2022, 836, 00489697, 155356, 10.1016/j.scitotenv.2022.155356 | |
2. | Zhengyu Lin, Chunyan Chen, Yuanli Liu, Guanghui Liu, Peng He, Guitang Liao, Wenbo Gao, Ji Cao, Zhouling Shao, Simulation of citrus production space based on MaxEnt, 2022, 10, 2296-665X, 10.3389/fenvs.2022.993920 | |
3. | Jinyao Lin, Hua Li, Yijuan Zeng, Xiaoyu He, Yaye Zhuang, Yingran Liang, Siyan Lu, Estimating potential illegal land development in conservation areas based on a presence-only model, 2022, 321, 03014797, 115994, 10.1016/j.jenvman.2022.115994 | |
4. | Adeel Ahmad, Sajid Rashid Ahmad, Hammad Gilani, 2022, Chapter 7, 978-981-19-4199-3, 135, 10.1007/978-981-19-4200-6_7 | |
5. | Qurat-ul ain-Ali, Nida Mushtaq, Rabia Amir, Alvina Gul, Muhammad Tahir, Faiza Munir, Farrukh Azeem, Genome-wide promoter analysis, homology modeling and protein interaction network of Dehydration Responsive Element Binding (DREB) gene family in Solanum tuberosum, 2021, 16, 1932-6203, e0261215, 10.1371/journal.pone.0261215 | |
6. | Marco O. V. Montero, Aurora A. Ariza, Nancy M. Barreno, Hilter F. Figueroa-Saavedra, Andrea F. Porras, Yul Aguilar, Wladimir Moya, Land suitability of coffee cultivation under climate change influence in the Ecuadorian Amazon, 2024, 17, 2542-1565, 49, 10.24057/2071-9388-2024-2969 | |
7. | Jocelyn M. Velázquez-Hernández, José Ariel Ruíz-Corral, Noé Durán-Puga, Diego R. González-Eguiarte, Fernando Santacruz-Ruvalcaba, Giovanni Emmanuel García-Romero, Jesús Germán de la Mora-Castañeda, Carlos Félix Barrera-Sánchez, Agustín Gallegos-Rodríguez, Eco-Geography of Dioscorea composita (Hemsl.) in México and Central America under the Influence of Climate Change, 2023, 15, 2071-1050, 12320, 10.3390/su151612320 | |
8. | Qingxiang Han, Ye Liu, Hongsheng Jiang, Xietian Chen, Huizhe Feng, Evaluation of Climate Change Impacts on the Potential Distribution of Wild Radish in East Asia, 2023, 12, 2223-7747, 3187, 10.3390/plants12183187 | |
9. | Manuela Hermeziu, Sorina Nițu, Radu Hermeziu, Correlations Between Genetic Characteristics and Agronomic Traits of Some Genotypes in the Selection Process of New Potato Varieties, 2023, 40, 12224227, 151, 10.59665/rar4014 |
Code | Bioclimatic variables name |
BIO1 | Annual Mean Temperature |
BIO2 | Mean Diurnal Range (Mean of monthly (max temp-min temp)) |
BIO3 | Isothermality (BIO2/BIO7) (*100) |
BIO4 | Temperature Seasonality (standard deviation *100) |
BIO5 | Max Temperature of Warmest Month |
BIO6 | Min Temperature of Coldest Month |
BIO7 | Temperature Annual Range (BIO5-BIO6) |
BIO8 | Mean Temperature of Wettest Quarter |
BIO9 | Mean Temperature of Driest Quarter |
BIO10 | Mean Temperature of Warmest Quarter |
BIO11 | Mean Temperature of Coldest Quarter |
BIO12 | Annual Precipitation |
BIO13 | Precipitation of Wettest Month |
BIO14 | Precipitation of Driest Month |
BIO15 | Precipitation Seasonality (Coefficient of Variation) |
BIO16 | Precipitation of Wettest Quarter |
BIO17 | Precipitation of Driest Quarter |
BIO18 | Precipitation of Warmest Quarter |
BIO19 | Precipitation of Coldest Quarter |
Code | Bioclimatic variables name |
BIO1 | Annual Mean Temperature |
BIO2 | Mean Diurnal Range (Mean of monthly (max temp-min temp)) |
BIO3 | Isothermality (BIO2/BIO7) (*100) |
BIO4 | Temperature Seasonality (standard deviation *100) |
BIO5 | Max Temperature of Warmest Month |
BIO6 | Min Temperature of Coldest Month |
BIO7 | Temperature Annual Range (BIO5-BIO6) |
BIO8 | Mean Temperature of Wettest Quarter |
BIO9 | Mean Temperature of Driest Quarter |
BIO10 | Mean Temperature of Warmest Quarter |
BIO11 | Mean Temperature of Coldest Quarter |
BIO12 | Annual Precipitation |
BIO13 | Precipitation of Wettest Month |
BIO14 | Precipitation of Driest Month |
BIO15 | Precipitation Seasonality (Coefficient of Variation) |
BIO16 | Precipitation of Wettest Quarter |
BIO17 | Precipitation of Driest Quarter |
BIO18 | Precipitation of Warmest Quarter |
BIO19 | Precipitation of Coldest Quarter |