Vulnerability of maize, barley, and wheat yields to growing season temperature and socioeconomic indicators in Morocco

Agriculture in Africa is increasingly being impacted by global climate change. The duration of the growing season is influenced by changes in climatic variability. Changes in precipitation in conjunction with a rise in average and severe temperatures will have consequences for water supplies upon which agriculture is dependent (Sonwa et al. 2017). Under the medium scenarios (RCP 4.5 and RCP 6) outlined in the IPCC Fifth Assessment Report (AR5), by the end of this century, several African countries will experience warming of more than 2 °C when compared to pre-industrial values. According to AR6, these pathways (1.5, 2, and 4 °C) predict increasing temperatures throughout Africa, including Morocco, as well as a rise in the frequency and severity of heavy but unevenly distributed precipitation events across Africa (IPCC 2022). Since 1901, temperatures have risen by more than 1 °C across most of Africa, causing a rise in the frequency and duration of heat waves and hot days (UNCC 2020). Studies have linked climate change to more frequent droughts in East Africa (Cook & Vizy 2013; Lott et al. 2013; IPCC 2022). Moreover, it is anticipated that droughts will become more frequent by the end of the 21st century throughout the continent (Dabo-Niang et al. 2014). Long-lasting and recurrent droughts as well as rising temperatures, dry out agricultural soils, make it difficult for seeds to germinate and crops to grow on time (Sonwa et al. 2017). According to other research, even a little rise in temperature may shorten the growing season for many crops, resulting in a lower ultimate harvest (Battisti & Naylor 2009; Lobell et al. 2011). As a consequence of warming and other variables, present agricultural systems are anticipated to see considerable yield decreases in the next decades (Qiao et al. 2010). As a result of these changes, primary cereal crops cultivated in Africa will be significantly affected by the middle of this century, albeit with geographical heterogeneity and variances in the vulnerability of crops (Nhamo et al. 2019). According to the most pessimistic projections on the effects of climate change, the average yield in western and central Africa is expected to decline by 13, and 11% in North Africa, and an 8% decrease in East and Southern Africa. While rice and wheat will be the most adversely impacted by heat stress and will experience production losses of 13 and 21%, millet and sorghum have been recognized as the most promising crops because they can handle heat stress better and are anticipated to suffer yield losses of 5 and 8%, respectively, by 2050 (UNCC 2020).

In recent years, North Africa has received a growing amount of attention as a ‘climate change hotspot.’ (Diffenbaugh & Giorgi 2012; Schilling et al. 2020). In the last several decades there has been a considerable warming trend in all parts of North Africa, with especially noticeable summertime readings (Donat et al. 2014). Along with the warming trend, the number of hot nights, warm days, and heat waves went up, while the number of cold waves went down (Elsharkawy & Elmallah 2016; Filahi et al. 2017; Nashwan et al. 2019; Zeroual et al. 2019). Every one-tenth of a degree increase in winter temperature reduces agricultural production in North Africa and Middle East (MENA) (Alboghdady & El-Hendawy 2016). Morocco is one of the nations that are most susceptible to the effects of climate change (Tramblay et al. 2013; Simonneaux et al. 2015). After the years 2010 and 2017, the year 2020 will go down in history as the warmest year ever recorded in Morocco. This is due to an annual national mean temperature anomaly that was +1.4 °C higher than the norm for the years 1981–2010 (Directorate General of Meteorology 2020).

The issue of vulnerability is very important to Morocco because its economy relies heavily on agriculture (Schilling et al. 2012) and its food security is based on cereals, which occupy 55% of the agricultural area (soft wheat 45%, barley 35%, and durum wheat 20%) (Dewi & Muhammad Amir Masruhim 2016). Phosphates, minerals, and tourism also play an important role but are not the subject of this current study. In most wheat-growing regions, it has been shown that warming is already causing production advances to slow down. It is anticipated that global wheat output will decrease by 6% for every additional °C rise in temperature, and it will become more (Asseng et al. 2015). So in order to prevent the deterioration of this scenario, it is crucial to comprehend how various climatic, yield and socio-economic conditions interact and influence food production before making judgments about how to adjust to the threats of climate change (Godfray et al. 2010). Identifying and quantifying the effects of various climatic factors on agricultural yields is necessary before implementing such tactics. An adaptation plan for a change in temperature is much different from one for a change in rainfall, for example (Climate Change and Crop Production). A rise in temperature has the ability to influence agricultural productivity significantly (Asseng et al. 2011).

To successfully adapt to climate change, we must comprehend how variations in agricultural yield correspond to climate change trends and how crop yields vary in response to climatic variation. Due to the significance of understanding the influence of climatic and non-climatic drivers of vulnerability, we choose to examine the crop yield vulnerability index in the context of growing season temperature and socio-economic variables using an approach developed by (Epule et al. 2017) and based on Ugandan maize output, and subsequently applied in other contexts (Epule & New 2019; Epule et al. 2021b). The present study's index is widely applicable in quantifying cropping system drought vulnerability in an African agricultural production environment by combining data on crop output with climate and adaptive capacity data based on two proxies that are particularly significant in Africa, namely literacy and poverty rates. On the other hand, research has used conventional approaches to illustrate this, for example, Mishra & Singh (2011) examined changes in precipitation and temperature, whereas Kamali et al. (2018) used physical and social indicators to examine the vulnerability of maize yields to water stress in Africa and (Profile et al. 2014) and (Mwaura & Okoboi 2014) then used a crop-climate interplay approach. Furthermore, the spatial and spatial-temporal aspects of drought risk and vulnerability were examined based on the Standardized Precipitation Index (SPI) and found that inter-annual precipitation variability is responsible for some of the vulnerability of the agricultural sector in Morocco (Fniguire et al. 2016).

This study aims at providing an assessment of the vulnerability of cereal crops, namely wheat, barley, and maize to changes in growing season temperature and the proxies of adaptive capacity at the national and sub-national scale. Therefore, vulnerability is assessed based on evidence of the stress imposed by climate on yield and also based on the ability of the people to adjust in the face of climate change through adaptive capacity; a robust approach (Burke et al. 2009). To the best of our knowledge, this study is the first of its kind to assess the exposure, sensitivity, and adaptive capacity of cropping systems based on growing season temperature as well as two socio-economic proxies, which are literacy rates and rates of poverty. Indeed, as far as we know there are no other studies in Morocco that evaluate the vulnerability of these crops to growing season temperature and socio-economic proxies such as poverty and literacy rates. The only study that comes close to this work focuses on vulnerability in the context of growing season precipitation for these same crops in Morocco (Achli et al. 2022). These findings will thus be used to advise and propose recommendations for strengthening the resilience of these three crops and provide a deeper understanding of the processes involved in designing better-targeted climate mitigation and adaptation strategies and programs in Morocco in the context of agriculture. Unfortunately, in Morocco, most climate-crop modeling studies have focused on the use of process-based models such as Aquacrop to assess the response of crops to various environmental conditions. Unfortunately, these approaches have not been able to integrate socio-economic conditions; challenges that this current study responds to adequately. Therefore, the lack of focus of previous research on the use of vulnerability indexes is partly anchored on the need to develop ways of integrating socio-economic variables.

The approach of framing vulnerability as a driver of sensitivity, exposure, and adaptive capacity goes beyond simply examining the climatic dimensions of vulnerability into also assessing the less frequently analyzed aspects that deal with the proxies of adaptive capacity (Schneider & Sarukhan 2001; Ford et al. 2013; Sherman et al. 2016). These proxies have been described as the ultimate determinants of the pattern of vulnerability and consequently, resilience. In fact, the scientific literature on vulnerability shows that the adaptive capacity component is often the most important indicator of vulnerability because it determines the ability of the farmers involved to cope and adjust to any climate stressors by using climate information through climate literacy or purchasing farm inputs such as high yielding drought tolerant planting materials (Simelton et al. 2009; Sharma & Ravindranath 2019). This approach is also important because it provides a framework from which yield, climate and socio-economic indicators can be used at the same scale through an index to assess vulnerability. In the subsequent parts of this work, we present the study area, deep dives into the composite concept of vulnerability, the methods, results, and discussion.

Barley plays an important and unique role in Morocco's agriculture and development. Morocco is a major center for the cultivation of barley. This is seen through a large variability and the presence of numerous wild Hordeum species including wild progenitors of cultivated barley collected in the low Atlas region (Amri et al. 2005). Among other crop species, barley has the greatest range of adaptation across varying climatic conditions as well as to harsh environmental and low-input agricultural systems. It has a short growing cycle that is well adapted to Mediterranean dryland conditions where late droughts are frequently experienced. In Morocco, barley is grown in all agroclimatic zones. However, 70 and 10% of the average 2.3 million hectares sown annually are planted in the semi-arid/arid and mountainous zones, respectively (Amri et al. 2005). The cultivation is always reserved for harsh conditions and shallow, rocky soils. It frequently follows wheat and is considered by most farmers as a minimum-risk crop that will not require additional inputs. The importance of barley in Morocco is mainly due to the multiple functions it plays at the farm level. Most farmers practice grazing on barley fields close to their houses. Barley straw and most of the barley grain are used as the preferred animal feed with approximately 20% of the production used as human food, which makes the country a leader in the use of barley for food (Amri et al. 2005). However, with the spread of bread wheat cultivation in Morocco in the second half of the twentieth century, the consumption of barley as food decreased significantly because of wheat flour subsidies. Barley is consumed mainly in the arid zones and mountainous regions. Morocco has given top priority to the improvement of barley productivity through breeding, agronomy, and valorization of the research efforts. With Morocco's predominating pattern of dry seasons, barley can play a major role in stabilizing cereal production and supplying feed for animals. Also, the dietetic benefits of barley are beginning to gain favor in affluent urban households. If this trend continues, the increasing consumption of barley by the population will help to limit the imports of bread wheat (Amri et al. 2005).

This study responds to a need by academics and policymakers to have a better understanding of the vulnerability of wheat, barley, and maize yields to changes in growing season temperature and socio-economic proxies of adaptive capacity on a national and sub-national scale in Morocco. This method establishes a baseline against which comparisons can be made and will greatly contribute to agricultural adaptation efforts. The desired information on climate change vulnerability revolves around several sub-indexes such as the exposure index, the index of sensitivity, and the index of adaptive capability. These terms, as well as the methodologies used to fit the models, will be defined in the following sections.

According to the IPCC, the sensitivity index is the degree to which the cropping system responds either negatively or positively to a climate shock or stressor (IPCC 2022). The effects might be direct, such as a reduction in crop yield as a result of variations in temperature and precipitation, amplitude, or variability; or they can be indirect, such as damage caused by rising sea levels resulting in more frequent storms on the coast (Agard et al. 2014; Sharma & Ravindranath 2019). In this research, the sensitivity index was used to describe the effects of the climate forcing (temperature) on the three crop yields under consideration. This involves describing the quantitative loss in crop production caused by climate change, climatic fluctuations, and severe events (IPCC 2007; Ford et al. 2013; Sherman et al. 2016).

This term is defined as the ability of a farming system to adjust and cope with potential damage, capitalize on opportunities, or respond to consequences (IPCC 2022). For example, the availability of irrigation facilities or increased crop diversification strengthens agricultural systems, while the provision of insurance protects may build resilience to crop failure. However, reduced adaptive capacity is linked to increased vulnerability, thus the indicators are parameterized accordingly (Sharma & Ravindranath 2019). In this study, the adaptive capacity of wheat, barley, and maize yields to adjust or cope with a variety of stressors and shocks, including catastrophic events, environmental issues, and climatic variability is dependent on proxies of adaptive capacity such as literacy and poverty rates (Climate Change 2007-IPCC 2007; Ford et al. 2010,,2013; Epule & New 2019). Since these two proxies are representative of adaptive capability, they were chosen as the key socio-economic proxies in Morocco as they impact most aspects of national life. Farmers, for example, tend to have limited access to social services and basic infrastructure where poverty rates are high, notably in mountainous areas, semi-arid plains, and highlands, as well as a lack of access to financial services, which impedes development. Similarly, when there is a lack of diversification of agricultural activities that provide alternative livelihoods during droughts and a lack of investment in irrigation and fertilizers, the population concerned becomes less adaptive in the face of climate shock (Ghanem 2016; Climate Action Report 2018). In addition, if the population's literacy rate is low, farmers may be unable to perceive and comprehend information connected to climate change, such as when to plant crops or which varieties are drought-resistant. They may also be unable to find other means to support their livelihoods (Epule et al. 2017). It is important to mention that technical and equipment assistance provided under the Green Morocco Program to medium and low-income producers has reduced the incidence of pesticide use in Morocco's farming sector. Therefore, lowering the level of poverty could lead to greater irrigation efficiency through easier access to information and the use of new practices in the agricultural industry. Reducing the rate of poor people can also foster literacy and resilience to climate shocks by improving people's ability to access climate change mitigation information and by building more sustainable systems of farming (Epule et al. 2021b).

This research is based on data gathered from a diverse variety of sources at the national and sub-national scale. The national annual yield data for all three crops (wheat, barley, and maize) for the period 1991–2016 were collected from the FAOSTAT database (FAOSTAT 2022) and used to compute the national sensitivity index. Data on crop yields at the sub-national scale were collected for the same period 1991–2016 from the Global Yield Gap Atlas (Global Yield Gap Atlas 2022) for nine sites across Morocco where these crops are mainly grown: Agadir, Beni Mellal, Fez, Marrakech, Meknes, Nador Aroui, Nouaceur, Safi and Tangiers. These sub-national sites were based on the availability of data.

We used the average annual temperature and average crop growing season temperatures of each crop during the period spanning 1991–2016. This data permitted us to compute the exposure index at the national and sub-national scales. These data were downloaded from the World Bank's Climate Change Knowledge Portal (World Bank 2020). As illustrated in Figure 3, the growing seasons for both barley and wheat are the same but different from the maize growing season. Also, it is noticeable that the three crops have a single growth season (uni-modal). While Figures 4 and 5 show a difference in the growing season of wheat and barley at the sub-national level due to several factors such as altitude and climate. The index of national-level exposure was then validated at the sub-national scale by using mean annual crop growing season temperature and yearly mean temperatures gathered from the World Bank's Climate Change Knowledge Portal (World Bank 2020).

To compute the adaptive capacity index, proxies of adaptive capacity such as literacy and poverty rates were employed. Both literacy and poverty rates are essential proxies of adaptation and resilience to climate disturbances in Morocco. Data on these two rates were collected for the national scale from Figshare (Epule et al. 2021a) from 1991 to 2016. The sub-national scale data were collected from regional disparities in development in Morocco (Yassine 2019).

In this study, we employed the vulnerability index created by (Epule et al. 2017; Epule & New 2019; Epule et al. 2021b) and tailored it for usage at the national and sub-national scales in an African crop farming context and predicated on the function of time series crop yields, growing season temperature, and socio-economic indicators that represent adaptive capacity (poverty and literacy rates) to assess the level of vulnerability of wheat, barley and maize in Morocco (Figure 2). The way that this research approaches the problem of conceptualizing vulnerability enhances the measurement of elements like adaptive capacity, which is notoriously difficult to quantify. Because gathering data on adaptive capacity across most of Africa is challenging, scientists have had to rely on proxies. In the case of Morocco two representative proxies of adaptive capacity are literacy and poverty rates. The vulnerability index and the sub-indexes are assessed on a scale of −2 and 2.

In addition to the above components of vulnerability presented in the equations above, this study further explored the data using linear regression of yield over time for the three crops, temperature over time and scatter plots of the linear relationship between yield and growing season temperature. In addition, the p-values were calculated to determine the level of significance of the above relationships based on a significance threshold of p value less than or equal to 0.05. The coefficient of determination or R² was also used to determine the contribution of any independent variable to the dependent variable. RCPs have been used to determine the likely potential impacts of different climate scenarios in the future. This helped us better understand the spatial distribution of crop yields across Morocco under different scenarios for (a) wheat at RCP 4.5, (b) wheat at RCP 8.5, (c) barley at RCP 4.5, and (d) barley at RCP 8.5. (1 cm to 200 km). The computations were done within the MOSAICC calculator and scenarios based on RCP 4.5 and 8.5. The calculator provides opportunities for the response of different crops to different scenarios based on different climate models. Our use was mainly aimed at validating the spatial dynamics in growing season temperature and the relative impacts on crop suitability and yield. The MOSAICC interface is available on http://196.200.148.123/mosaicc/.

Also, the time series of yield over the period 1991–2016 shows that though highly fluctuating for all three crops, wheat consistently records higher yields throughout the data series (Figure 10(b)). Maize records the lowest yields while barley is in between except in the years 1997, 2000, 2012, and 2016 where barley was the lowest. These trends go a long way to support the vulnerability index (Figure 7(a)–7(e)) analysis, which illustrates that despite having the highest vulnerability and adaptive capacity indices, all the way through the period 1991–2016, the highest vulnerability and the lowest adaptive capacity indexes are exhibited by the crop species barley; the median ratings for the crop species in between are shown by maize. The accompanying R² for all crops further solidifies our finding. This is seen as wheat records the highest yields, lower vulnerability, and higher adaptive capacity indexes record an R² of 17% and a p-value = 0.03 which is significant. Maize on the other hand records a second median level of vulnerability index, an R² of 12% and a non-significant p-value = 0.07. Also, barley, which has the weakest index of adaptive capacity and the highest index of vulnerability, records the lowest R² of just 4%, making it less significant (p-value = 0.30).

As concerns growing season temperature, this may be taken into consideration as the normalized growth season temperature reduces northward as vulnerability also reduces. The northward decrease in vulnerability is higher for wheat, followed by maize and barley. This demonstrates that wheat is overall less vulnerable and most adaptive. In terms of these remaining crops, maize shows the second to last lowest vulnerability and second to last highest adaptive capacity index, while barley has a higher index of vulnerability and a lower index of adaptive capacity (Figure 10). All these crops, however, have a decreasing vulnerability rating and an increased adaptive capacity northward. The normalized growing season temperature also decreases from the south to the north. At each specific latitude, wheat has the strongest index of adaptive capacity plus the lowest vulnerability index, as well as the second highest level of normalized growing season temperatures of all the crops tested. The resulting data at the sub-national levels are internally consistent with national results and provide an improved overall basis for understanding spatial and sub-national changes in vulnerability, resilience, and growing season standardized temperature.

Temperature is an important driver of crop yield (Ciaffi et al. 1996), as well as a significant determining factor for germination in dry and semi-arid regions (Huang et al. 2003; Tlig et al. 2008). In the absence of water, temperature has a greater impact on plant growth and development than other environmental elements, notably in reproductive activities (Thuzar et al. 2010; Iloh et al. 2014). According to our RCP findings from this study, when we move from RCP 4.5 to RCP 8.5, the yield tends to fall as in the case of wheat as seen in Figure 11. According to previous research, the production of cereal crops might be reduced by up to 40% if the average temperature rises by 2.0 °C, notably in Africa (Fatima et al. 2020). Furthermore, depending on the region (Hu et al. 2005; García-Mozo et al. 2010; Tsimba et al. 2013; Bai et al. 2016; Shim et al. 2017; Wu et al. 2019; He et al. 2020), temperature increase can have either a negative or favorable impact on crop yield. The majority of studies anticipate that there would be a reduction in yields of rainfed wheat and rice for a temperature rise of 2–3.5 °C, as well as a loss of agricultural revenue of between 9 and 25% at the farm level, (Varga 2021). In this study, our findings show that increased temperature from RCP 4.5 to RCP 8.5 will result in reduced wheat production while barley production is unchanged. Therefore, the assertion that the impacts of temperature change might vary from one region to another and for different crops is highly consistent with our findings. This is clearly consistent with other studies which note that wheat crop yields have the potential to decrease by 32% if the average yearly temperature rises by only 1% (Mourad et al. 2017). More research evidence conducted to study the effects of climate change on the growing seasons of many types of crops including the shortening of the crop and its maturation at an earlier age does exist (Brown & Rosenberg 1999; Chmielewski et al. 2004; Schwartz et al. 2006). Islam (Karim et al. 2012) explained the yield loss as a function of a shorter growing season as a result of the greater number of birds in the pen due to increased temperature. In addition, during the maize growing season, the majority of northern African countries, notably Morocco, Algeria, and Sudan, had greater rates of warming, with Tmean trends of more than 0.3 °C per decade between 1961 and 2010 (Shi & Tao 2014). In these countries, this indicates that a little difference in Tmoy can cause a great difference in Tmoy variation in maize yields. Kazakhstan's crop yields are weak not only because of the country's harsh climatic circumstances (a short growth season with elevated heat and limited water supply), but also because of the minimal amount of inputs that are used (fertilizers, pesticides, and water) (Schierhorn et al. 2020). In this study, the R² are generally below 50%, indicating that the influence of temperature on yield is weak, for example and not as impactful as precipitation, soils, and socio-demographics. In a country where these crops are essentially rainfed, it is without doubt that precipitation might be playing a more significant role. Crop production is therefore a complex process that is often driven by several rather than any single factor. It can be remarked here that in addition to the effects of growing season temperature, other factors that could affect cereal production were not considered in this study, such as irrigation, fertilization, land management, and soil properties. The weak R² is an indication that the yield of the concerned crops can be better understood when several variables are computed. However, the objective of this current work is to investigate the linear relationship between yield and annual temperature as well as growing season temperatures for concerned crops. Introducing precipitation in the equation will yield high correlations (R) and coefficients of determination (R²). For example, a study by Achli et al. (2022) that investigated the vulnerability of barley, maize and wheat production in Morocco observed that the scatter plots of barley yield vs growing season precipitation produce R of 55% and R² of 30, the scatter plot of maize vs growing season precipitation produces R of 32% and R² of 13% while the scatter plot of wheat vs growing season precipitation produces R of 57% and R² of 26%. This shows that precipitation data evidently enhances the outputs and the performance of the models. However, in terms of vulnerability, the pattern is observed for both studies.

The findings of this study reveal that there is a spatial variation of the vulnerability of cereal yields across Morocco. Morocco's annual mean temperature fluctuates between 11.8 and 21 °C, with the lowest readings coming from hilly areas (such as Ifrane station) and the highest coming from southern locations (such as Lâayoune station). Typically, temperatures will increase as we get inland from the coastlines and as we travel southward from the northern reaches of the country (Driouech et al. 2021). This shows increased yields of wheat grown on the best soils and the highest rainfall in the country. In general, the nicest sites are found around or close to the coast (Global yield gap atlas 2022). It is also good for semi-arid regions since the wheat cultivated in Northwest Africa is a short-season cultivar with minimal water needs, making it perfect for the region. Though wheat is sown in the autumn, it doesn't go totally dormant in Morocco's warmer winters (Global yield gap atlas 2022). While barley is mostly seeded under a traditional agricultural technique defined by the use of a few chosen seeds, fertilizers, and pesticides, it is cultivated alongside maize located in the central highlands as well as on plateaus and poorer soils toward the south (Karaky et al. 2004; Global yield gap atlas 2022). So, farmers are less inclined to spend money in areas that are more marginal, and lower yields are generated. In expansion, the drought in Morocco's Maize Growing Season exhibited increasing patterns, which had a significant impact on maize yield trends (Schilling et al. 2012).

Further, the average highest and lowest temperatures play a role in figuring out how long the growth season is, as well as the optimum sowing and the window of opportunity for doing so; both factors can influence the possible yield, the quality of the produce, and their overall productivity. In general, the length of time needed for maximum crop yields corresponds directly to the length of the optimal growing season (Sullivan 2002; Fatima et al. 2020). According to the findings of Srivastava et al.'s research (Srivastava et al. 2021), the highest temperature has a greater impact on maize grain production than the lowest temperature does, given that CO₂ levels change depending on whether there is irrigation or rain. The research carried out by (Hatfield & Prueger 2015) shows that maize grain yields may be lowered by as much as 90% if temperatures during the reproductive season are extreme, causing pollen to become infertile and the grain to become smaller. The essential threshold temperature for maize in the United States is 29° C (Meerburg et al. 2009). Besides, the temperature may affect the grain filling rate in wheat, however, this effect is only noticeable up to around 20 degrees Celsius (Gooding et al. 2003; Reynolds 2010). Wheat germination was absolutely restricted under extremely high thermal stress (45 °C), which caused the loss of some cells and other embryos, and the rapidity with which young plants were developed was also diminished (Cheng et al. 2009). In another important aspect, the enhanced climate susceptibility of barley in Europe is an essential element (Moore & Lobell 2015), that is most probably connected to the lower temperature at which barley performs at its optimal level compared to wheat (Lobell & Gourdji 2012). The 2010 wheat crop in Morocco was shorter than typical due to cold temperatures in the first part of March which inhibited good flower propagation. These cold conditions caused flower destruction and speeded up the completion of the pollen stage. The seeds that later emerged were unable to mature. By the time the grain was in the flowering stage, the average temperature in this area of Morocco had already dropped to two to four degrees Celsius, which was sufficient to lower the number of grains per year, causing a decrease in grain yields in the area affected, which might aggravate low soil moisture conditions and affect crop yields (Directorate General of Meteorology 2020). Hot seasons, on the other hand, are more common in Morocco and are frequently connected with droughts. Consider the 1980s drought, which was a climate phenomenon that had a significant impact not only on cereals but also on livestock (El Harraki et al. 2020). A study by Shew et al. (2020) used complex regression models to analyze 71 different South African cultivars for their susceptibility to climate change from 1998 to 2014. Heat was found to be the primary cause of wheat production reductions, according to the findings. For warming scenarios with a uniform increase of +1, +2, and +3 °C, yield reductions of 8.5, 18.4, and 28.5%, respectively, were recorded. Another study made in Nigeria indicates clearly that increased temperatures seem to have slowed or halted seedling development in the three major cereal crops studied which are maize, rice and sorghum (Iloh et al. 2014).

In general, African farmers seem to be more susceptible to the effects of rising temperatures, erratic rainfall patterns, and unpredictable crop yields than farmers in advanced nations. In Morocco, poverty inhibits local populations' capacity to cope with these developments. In certain places, the average farm household income is as low as USD 540$ per year (Morocco | UNDP Climate Change Adaptation; UNDP Morocco 2022). Most rural residents are primarily dependent on agriculture to make money, with many engaged in informal agricultural labor. Inadequate access to essential social and infrastructural facilities mainly in semi-arid plains, mountainous zones, and highlands, as well as inadequate financial services impede development. A lack of diversification of agricultural operations affording alternative employment, particularly during times of drought, also contributes to poverty (Ghanem 2016; MOPAN 2019). Also due to the poor abilities of Moroccan farmers, agricultural extension services (AESs) are extremely scarce in the nation. The AES refers to the services that farmers receive in the form of agricultural advice. It is an important application of ‘scientific research and knowledge’ in farmer education in which information inputs are delivered to farmers in order for them to improve agricultural productivity, improve food security, promote agriculture, and raise their standard of living in order to generate economic growth (Benin et al. 2011; Labarthe & Laurent 2013). Such inequalities in the rates of poverty and readership are quite common in Africa in the main and in Morocco in the secondary and therefore are a very good proxy for adaptive capacity, demonstrating that climate change vulnerable crops are closely linked to poverty and literacy rates (Burton et al. 2002; Abeygunawardena et al. 2009). Invariably, this work has shown that the adaptive capacity index plays a key role in driving the vulnerability index because no matter the strength of the climate forcing, the ultimate pattern of vulnerability will be determined by the proxies of adaptive capacity. This has been made evident by several previous studies including (Achli et al. 2022).

In terms of broader policy implications, it is obvious that Morocco needs an effective planning procedure to retain good agricultural land and farm operations that provide the food needs of the population. The strengthening of adaptation capacities therefore appears to be a determining factor, particularly for rainfed crops, to cope with the decrease in water resources and the evolution of areas suitable for cereal crops. Soil conservation and fertility preservation techniques (e.g., sowing under plant cover, agroforestry, agroecology) could be of increased interest in this context. Experiments with sowing under plant cover in the Fez-Meknes region, for example, have shown that yields are improved in dry years. The increase in crop areas favorable for cereal crops in mountainous areas shows the interest in agroforestry to fight against soil erosion (Balaghi 2017). Crop diversification, the selection of adapted varieties and the development of agro-ecological practices which are production systems that respect ecosystems and are labor intensive, will be supported as a solution for the resilience of the agricultural sector to the threat of climate change and the production of new employment opportunities. Furthermore, CA (conservation agriculture) is proven to be a feasible alternative for satisfying the Moroccan population's food demands sustainably in the next decades (Solh & Saxena 2010). For example, the implementation of CA in Morocco is related to a rise of 28% in the total amount of wheat produced nationally, saving Morocco around USD172$ million in yearly import expenses (Yigezu et al. 2021). Irrigated crops, which are less impacted by climate change, could, however, see their yields or production decrease in the event of irrigation restrictions imposed by shortages. The adaptive capacities of irrigated agriculture, in particular the improvement of water resource use efficiency, therefore also appear crucial.

Some of the weaknesses of this work are principally linked to the availability of data. For example, we did not account for the effects of other proxies of the adaptive capacity index, so we were confined to using poverty and literacy rates which are representative proxies. However, as far as we can tell, research on the influence of historical and contemporary changes in climatic conditions paired with socio-economic indices on cereal yields is uncommon in Morocco, even though the African cereal market depends on this area. Even more so, it is difficult to distinguish the effects of expected climatic change on agricultural production from the repercussions of corresponding socio-economic and the associated institutional shifts. Our research results offer the first fine-scale evaluation of the effect of the growing season temperature on the most common cereal harvests in Morocco by considering the three elements of vulnerability (exposure, sensibility, and coping capacities) to support the agricultural community, policymakers, and investors to formulate regionally tailored adaptation plans for climate change.

According to this work, and in relative terms, wheat records the lowest vulnerability indexes at both the national and sub-national levels. It records also the highest adaptive capacity indexes as well as the second highest normalized growing season temperature. In expansion, as we travel northward, both the vulnerability of wheat, barley, and maize and the normalized growing season temperature decrease while adaptive capacity improves. The development of more nutrient-dense crops will be made easier with a greater knowledge of how environmental growing circumstances (such as precipitation, temperature, soil qualities management, etc.) impact grain production and nutritional characteristics of cereals. Indeed, including climate considerations in larger planning operations would inspire policymakers to encourage diversification toward more climate-resilient food crops in vulnerable locations. Furthermore, the Moroccan government may employ this data to enhance state subsidies and design agricultural policies that assist areas based on their comparative advantage in cereal production, considering climate change and adaptive capacity indicators. Policymakers and investors must also carefully consider whether to help sustain cereal farming or to look for other farming activities for places that are already badly impacted by the change in climate. It would be worthwhile now to make other spatial studies for assessing vulnerability, adaptive capacity, and normalized growing season temperature of other crops in Morocco if more data becomes available. Similarly, developing scenarios and assessing the effects of future temperature changes could be a useful tool to analyze and anticipate projected hazards on crop yields.

This work is funded by the Pan Moroccan crop precipitation platform (PAMOCPP) project funded by OCP Foundation and awarded to Terence Epule.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.