Keywords

1 Background

The Mediterranean region is a bridge connecting Asia, Africa, and Europe, consisting of 21 countries around the Mediterranean Sea. The Mediterranean region is the birthplace of ancient Egyptian, Babylonian, Roman, and Greek civilizations. It is a global biodiversity hotspot (Médail and Quezél 1999), with up to 25,000 plant species (Cuttelod et al. 2009), making it a typical and representative area for studying dryland ecosystems. The aridity index (AI) is the ratio of potential evapotranspiration (PET) and annual precipitation (Budyko 1974). A region with an AI of less than 0.65 belongs to a drylands region (Hulme 1996). The Mediterranean region is classified as hyperarid (AI < 0.05), arid (0.05 < AI < 0.2), semiarid (0.2 < AI < 0.5), dry subhumid (0.5 < AI < 0.65), humid (0.65 < AI < 0.75) and hyperhumid (AI > 0.75) areas based on the AI. Drylands occupy 85.98% of the Mediterranean region, of which hyperarid, arid, semiarid, and dry-subhumid drylands account for 48.76%, 13.44%, 18.75%, and 5.03%, respectively. Spatially, hyperarid and arid regions occupy the largest part of the drylands of North African and West Asian countries, while semiarid and subhumid arid regions are mainly in Turkey, Greece, Italy, Spain, and Portugal, as well as in the coastal areas of Morocco, Algeria and West Asia (Fig. 8.1).

Fig. 8.1
A Mediterranean region's distribution map highlights drylands. North African and West Asian countries are hyper-arid, while semiarid and subhumid areas encompass Turkey, Greece, Italy, Spain, Portugal, and coastal regions in Morocco, Algeria, and West Asia.

Distribution of Mediterranean drylands based on the Global Drought Index Climate Database v2 (Trabucco et al. 2019)

The Mediterranean region faces challenges in achieving Sustainable Development Goals under a significant warming trend and complicated anthropogenic factors. The first challenge is water shortages, as 85.98% of the Mediterranean region is classified as drylands (Zeng et al. 2021). The second challenge is the pressure to feed a rapidly growing population, especially in the Middle East and North Africa, where the population will double from 2015 to 2080 (Waha et al. 2017). The third challenge is a significant trend of warming (Lionello and Scarascia 2018), which tends to increase arid conditions and lead to land degradation and biodiversity loss in the ecosystem. Different challenges interact with each other. Population growth will lead to the expansion of cultivated land, which in turn will increase the consumption of water resources, while climate warming will reduce the amount of water resources. The conflict between the shortage of water resources and the expansion of cultivated land will lead to the abandonment of cultivated land and land degradation. How to balance the conflict between water, food and ecological protection is an urgent issue that requires close attention in the sustainable management of Mediterranean dryland ecosystems.

2 Major Characteristics of Drylands in the Region

2.1 Climate and Distribution of Drylands

The climate characteristics of the drylands show significant spatial heterogeneity, with arid, desert, and hot climates in the south; arid, steppe, and cold climates in Turkey and Spain; and temperate, dry, and hot summer climates in other regions. According to updated Köppen–Geiger climate data (Beck et al. 2018), there are 16 climate types in the drylands of the Mediterranean region (Fig. 8.2). BWh (arid, desert, hot), Csa (temperate, dry summer, hot summer), BSk (arid, steppe, cold), and BWk (arid, desert, cold) are the dominant climate types, accounting for 64.65%, 11.88%, 9.67% and 3.17% of the area of drylands, respectively. The Csa climate occupies the central part of coastal areas of the Mediterranean region, which favors the growth of heat-tolerant crops such as olives, grapes, figs, and citrus, as well as the accumulation of sugar in fruit crops.

Fig. 8.2
A distribution map of the Mediterranean region categorizes climates using the updated Koppen–Geiger classification. It encompasses 16 climate types including B W h, B W k, B S h, B S k, C s a, C s b, C f a, C f b, D s a, D s b, D s c, D f a, D f b, D f c, and E T.

Distribution of climate patterns based on updated Köppen–Geiger climate

The climate type, geography, and topography govern the precipitation and temperature in drylands. The annual precipitation and annual mean temperature of dryland areas are shown in Fig. 8.3 according to the WorldClim 2 dataset from 1970 to 2000 (Fick and Hijmans 2017). The annual precipitation (Fig. 8.3a) of 51.48% of the drylands is less than 50 mm and is mainly distributed in the Nile delta and the desert areas of North Africa. A total of 15.02% of the drylands receive between 50 and 200 mm of annual precipitation, mainly in the desert periphery. The annual precipitation of 8.24% of the drylands ranges between 200 and 400 mm per year, mainly in Morocco and Algeria. A total of 23.55% of the drylands have an annual precipitation amount between 400 and 800 mm per year, mainly in Turkey, parts of Europe, and the coastal areas of North Africa. Temperatures in the drylands also show significant spatial heterogeneity (Fig. 8.3b), gradually decreasing from south to north. In particular, the main parts of North Africa and West Asia are dominated by hot weather, with annual mean temperatures varying between 17 and 29 °C. In contrast, the European part and the coastal areas of Morocco and Algeria are characterized by temperate and cold temperatures.

Fig. 8.3
2 distribution maps of the Mediterranean region categorize regions based on annual mean precipitation and temperature. Over 51.48% of drylands receive less than 50 mm of annual precipitation, concentrated in the Nile Delta and North African deserts. Temperature exhibits spatial heterogeneity, decreasing gradually from south to north.

a Annual mean precipitation; b temperature

2.2 Land Cover and Land Use

Land cover types in the drylands of the Mediterranean region are significantly affected by dry and hot climates, which determine the predominance of low-productivity land cover in the drylands of the Mediterranean region. Moreover, water availability, soil characteristics, wildfires, and land abandonment have profound impacts on vegetation cover and patterns in the Mediterranean region (Fenu et al. 2013; Gouveia et al. 2017; Satir et al. 2016). For instance, drought strongly affects dry and desert vegetation (Gouveia et al. 2017), and dune plants along the western Mediterranean coast are deeply affected by soil properties (Fenu et al. 2013). According to the 2019 land cover and land use data from Copernicus Global Land Cover (Buchhorn et al. 2020), the land cover types and land use patterns in the drylands of the Mediterranean region show strong spatial heterogeneity (Fig. 8.4). Low-productivity bare/sparse vegetation accounts for 63.11% of the dryland area in the Mediterranean region, with most of these areas distributed in North Africa and West Asia. Cultivated agriculture accounts for 13.14% of the region’s dryland area and is distributed in European countries, Turkey, and the coastal areas of North Africa and West Asia. Herbaceous vegetation, sparse forest, dense forest and shrubs account for 7.98%, 5.39%, 4.38%, and 4.46%, respectively. The highly productive land cover types are mainly distributed in the European part and Turkey, which have relatively high precipitation and mild temperatures. Although the desert climate dominates the Nile delta, it benefits from the rich water resources of the Nile River and is the central part of the cropland, vegetated, and built-up areas of Egypt.

Fig. 8.4
A land cover type distribution map of the Mediterranean region categorizes areas into shrubs, herbaceous vegetation, cultivated agriculture, urban, bare and sparse vegetation, snow and ice, permanent water bodies, herbaceous wetland, moss and lichen, closed forest, and open forest. Libya, Egypt, Israel, Jordan, and Algeria exhibit the highest area of bare and sparse vegetation.

Land cover type distribution in the dryland areas

2.3 Land Degradation and Its Signal

Land degradation or desertification indicates the transition from a productive vegetative state to unproductive bare land (Zelnik et al. 2013). The drylands of the Mediterranean region with high temperature and low annual precipitation are prone to land degradation and desertification. In particular, areas around deserts and barren land have a higher risk of desertification (Huang et al. 2020). Land degradation and desertification were widely observed in the southern part of the Mediterranean region (Safriel 2009) and even in the European part (Ferreira et al. 2022; Jucker Riva et al. 2017) due to the increasing trend of unsustainable land use. The results of previous studies suggest that drylands in the Mediterranean region may degrade in three ways under climate change. The first is soil erosion due to increased drought, intense rainfall, and other climatic extremes. The second is soil salinization due to increased drought, irrigation, and increased sea level. The third is the depletion of soil carbon stocks due to increased temperature and drought (Lagacherie et al. 2018). Land degradation and transboundary migration have occurred in Mediterranean drylands due to severe conflicts between population, water scarcity, and land (Mohamed and Squires 2018). This snowball effect, characterized by the reclamation, degradation, abandonment, and reclamation of arable land, has evolved land degradation in the Mediterranean region from an environmental biophysical phenomenon to a social security issue (Mohamed and Squires 2018). Land degradation drew the attention of the European Commission, which launched the Mediterranean Desertification and Land Use (MEDALUS) program to monitor the sensitivity of land to degradation and desertification during the period 1991–1999 (Kosmas et al. 1999). Different kinds of tools and methods are proposed for the assessment of soil erosion and desertification. The change in vegetation pattern (Zurlini et al. 2014) can be used to capture the early sign of land degradation. For example, spotted vegetation patterns can be considered a key signal of vegetation degradation from vegetation to bare land. Earth observations play an important role in land degradation at a large scale, while the signal of sparse vegetation in satellite images is susceptible to contamination by bare land due to low vegetation cover. The weakness of satellite data can be overcome by unmanned aerial vehicles (UAVs) that carry different sensors with a very fine spatial resolution and can capture overgrazing, aridity, and vegetation pattern changes in an easy manner (Kyriacos 2017). The abrupt change in vegetation time series can be used to identify land degradation (Smith et al. 2019). Time series segmented residual trends (TSS-RESTREND) for vegetation time series analysis (Burrell et al. 2017) is proposed to capture land degradation. With the development of remote sensing cloud computing platforms and the occurrence of mountain data, some cloud computing tools have been developed to estimate soil erosion and land degradation. Recently, Elnashar et al. developed the RUSLE-GEE for soil erosion assessment (Elnashar et al. 2021b) and MEDALUS-GEE for desertification (Elnashar et al. 2022). Both use public data to drive models to predict soil erosion and desertification, serving as tools to assess the risk of soil erosion and desertification in a developing country.

2.4 Shrub Encroachment

Shrub encroachment is generally considered an important cause of dryland grassland degradation (Cao et al. 2019) and has been widely reported in the Mediterranean region. It may threaten livestock and pastoralists’ livelihoods (Belayneh and Tessema 2017; Nunes et al. 2019). However, whether shrub encroachment has a positive or negative impact on Mediterranean dryland ecosystems is controversial, depending on the function and traits of shrub species (Valencia et al. 2015). Several studies have shown that shrub encroachment negatively affects dryland biodiversity by altering soil bacterial communities (Stanton et al. 2018; Ubach et al. 2020; Xiang et al. 2018). Some studies have shown that shrub encroachment benefits ecosystem biodiversity. For example, Aleppo pine encroachment reduced the nesting success of Sardinian warblers and increased the activity of Eurasian jays (Ben-David et al. 2019), and soil soluble carbon and nitrogen mineralization in Mediterranean oak woodlands was higher in shrub-encroached areas than in nonencroached areas (Gómez-Rey et al. 2013). Shrub cover in agropastoral systems in southern Portugal increased above-ground biomass and net primary productivity (Castro and Freitas 2009). Shrub encroachment in degraded grasslands in Spain increased vascular plant abundance and the biomass of fungi, actinomycetes, and other bacteria (Maestre et al. 2009). The relationship between climate and encroachment is complex and controversial. Some studies have suggested that shrub encroachment might amplify the effects of climate, thus increasing the exposure of Mediterranean woody grasslands to drought (Rolo and Moreno 2019). Other studies have demonstrated that shrub encroachment has a positive effect on reversing desertification processes and improving ecosystem function (Maestre et al. 2009). However, the lack of fine-resolution shrub encroachment products hinders the ability to determine the impact of shrub encroachment on dryland ecosystems. The prediction and monitoring of shrub encroachment is essential to study its effects on dryland ecosystems. Related studies have revealed that topography and soil conditions are better predictors of shrub encroachment than climate (Nunes et al. 2019). Remote sensing has great potential for monitoring shrub encroachment, and recent studies have shown that unmanned aircraft systems and light detection and ranging (LiDAR) can be used to identify shrub encroachment (Madsen et al. 2020).

2.5 Loss of Biological Soil Crust

In the Mediterranean region, biological soil crusts (hereafter referred to as biocrusts) are mainly distributed in the southern and eastern arid regions, the Iberian Peninsula and Turkey (Fig. 8.5). The composition and distribution of cyanobacterial diversity in Mediterranean ecosystems are mainly governed by temperature and precipitation (Muñoz‐Martín et al. 2019).

Fig. 8.5
A distribution map of the Mediterranean region highlights that biocrusts are predominantly found in the southern and eastern arid regions, as well as in the Iberian Peninsula and Turkey.

Distribution of biocrusts in the Mediterranean region (redrawn based on Rodriguez-Caballero’s study)

Biocrusts play a crucial role in maintaining the function of Mediterranean dryland ecosystems (Morillas et al. 2017). Biocrusts have profoundly impacted erosion control and the regulation of soil moisture and air quality (Morillas et al. 2017; Rodríguez-Caballero et al. 2018). Biocrusts are an excellent indicator that can be used to trace soil processes by modifying or improving soil chemistry (Miralles et al. 2020). They positively influenced seed germination and grass growth conditions of Mediterranean perennial grasses in Spain by improving soil chemistry and leaf nutrient uptake (Ghiloufi et al. 2017) and effectively increased water infiltration and soil moisture, reduced soil evaporation, and ultimately increased plant water (Chamizo et al. 2016).

The dryland ecosystem in the Mediterranean region is sensitive to climate warming. The intrinsic link between climate change and biocrusts is complex. The development of biocrusts can buffer the effect of climate warming (Delgado‐Baquerizo et al. 2016; Lafuente et al. 2020) and mitigate the negative impacts of increasing aridity on the multifunctionality of dryland ecosystems (Delgado‐Baquerizo et al. 2016). In turn, climate warming reduces soil water availability, leading to the loss of cover, abundance and diversity of biocrusts (Benvenutto‐Vargas and Ochoa‐Hueso 2020; Rodriguez-Caballero et al. 2018). In addition, biocrusts are very sensitive to atmospheric nitrogen (N) deposition and animal activity. A decrease in soil water availability and an increase in animal activity can reduce the coverage, abundance and richness of biocrusts (Ladron de Guevara et al. 2018). The warming of drylands in the Mediterranean region is already evident under representative concentration pathways (RCPs) 4.5 and 8.5. The intensity of arable cultivation and grazing in the drylands of the Mediterranean region has increased significantly with the rapid population growth in the arid zone. Under the influences of climate warming and increased human activities, the biocrust cover and abundance in the Mediterranean region will continue to decrease (Ladron de Guevara et al. 2018; Maestre et al. 2015).

Studying the changes in biocrust distribution, cover and abundance is critical in assessing the situation of dryland ecosystems. The identification of biocrusts at a large scale is challenging. Remote sensing is regarded as an effective way to map the distribution of biocrusts. Regional or global biocrust products with a fine spatial resolution are missing due to the spectral similarity between biocrusts and bare ground and signal interference from shrubland. To date, only Rodriguez-Caballero et al. (2018) have produced a global map of biocrusts with a coarse spatial resolution. Recently, satellites have developed toward high spatial resolution and frequency and provide a good opportunity for biocrust mapping. For example, Sentinel-2 multispectral data were used to trace biocrust changes in the Negev Desert (Israel) (Panigada et al. 2019). Hyperspectral airborne data were found to perform better than multispectral data in biocrust mapping (Rodríguez-Caballero et al. 2014, 2017).

2.6 Social and Economic Development

The populations of Mediterranean countries have shown diverse spatial changes. Populations in the southern and eastern Mediterranean region continue to grow rapidly, while European countries suffer population aging (Doignon 2020). Population aging in European countries accelerates land abandonment and shrub encroachment due to a lack of labor, which leads to a decrease in long-term soil erosion rates (Cerdà et al. 2018). The population boom in arid regions has significantly increased the pressure on the regional food supply. It has exacerbated the overexploitation of land and the overpumping of water resources in the arid region. Water limitation is prone to cause land abandonment and accelerate land degradation and desertification (Mohamed and Squires 2018). There is a very large gap in the economy between the European region and the North African and West Asian regions. Almost all North African and West Asian countries are in the lower-middle-income category, well below the world average, while most countries in Europe are in the high-income category (Fig. 8.6).

Fig. 8.6
A distribution map of the Mediterranean region highlights the G D P per capita, with the highest G D P observed in Israel and France, followed by Cyprus, Slovenia, and Portugal.

GDP per capita (current US$) of the Mediterranean countries in 2020, Data source World Bank, https://ourworldindata.org/grapher/population-density-vs-prosperity?time=2020

To pursue a better livelihood and higher income, large-scale population migration has been observed in the Mediterranean region. First, a large population migrated from rural areas into towns and cities to pursue better livelihoods along the Mediterranean coast (Wolff et al. 2020). Second, cross-border migration from the eastern and southern Mediterranean regions to European countries has been widely observed in the Mediterranean region (Crawley et al. 2016). In 2015, a Syrian refugee crisis occurred that led to more than 1 million people crossing the Mediterranean Sea to Europe along the eastern Mediterranean (Fig. 8.7). Cross-border migration has become a significant social issue affecting the sustainable development of Mediterranean drylands (Perkowski 2016; van Reekum 2016).

Fig. 8.7
A bar graph illustrates the variation in the number of people over the years. The peak, occurring in October 2015, records 230 individuals, while the trough, observed in October 2020 exhibits a minimum of 5 people.

Monthly demography of sea and land arrivals between 2015 and 2021. Data source UNHCR, https://data2.unhcr.org/en/situations/mediterranean

3 Change in Drylands in the Region

3.1 Climate Change

Dryland areas in the Mediterranean region have experienced a significant warming trend (Fig. 8.8), and it has been well documented that the Mediterranean region has experienced a significant decrease in precipitation and warming in recent decades (Prăvălie 2016). The warming trend in the Mediterranean region is more prominent than the global average, and the temperature in the Mediterranean region will be 20% higher than the global average in the twenty-first century (Lionello and Scarascia 2018). According to the CMIP5 climate change dataset (Thrasher et al. 2012), the annual growth rates of the minimum temperature (Tmin) from 2020 to 2099 are 0.014 °C to 0.037 °C (Fig. 8.9a) and 0.034 °C to 0.075 °C (Fig. 8.10a) for RCPs 4.5 and 8.5, respectively; the annual growth rates of the maximum temperature (Tmax) are 0.012 °C to 0.038 °C (Fig. 8.9b) and 0.033 °C to 0.079 °C (Fig. 8.10b), respectively, for RCPs 4.5 and 8.5. The most significant increase in Tmin will occur in Turkey under RCP4.5 and RCP8.5, while the most significant increase in Tmax will occur in Turkey and Morocco under RCP4.5 and RCP8.5.

Fig. 8.8
2 double line graph plots temperature versus the years. The lines are R C P 4.5 and R C P 8.5. The lines have a fluctuating trend followed by a gradually increasing trend.

a Tmin and b Tmax change trends under RCP4.5 and RCP8.5

Fig. 8.9
2 distribution maps of the Mediterranean region categorize regions into T min and T max from 2020 to 2010. The annual growth rates of the minimum temperature are 0.014 to 0.037 degree Celsius for R C P 4.5 and the annual growth rates of the maximum temperature are 0.012 to 0.038 degree Celsius.

a Tmin and b Tmax change trends from 2020 to 2010 at RCP4.5

Fig. 8.10
2 distribution maps of the Mediterranean region categorize regions into T min and T max from 2020 to 2010. The annual growth rates of the minimum temperature are 0.034 to 0.075 degree Celsius for R C Ps 8.5 and the annual growth rates of the maximum temperature are 0.033 to 0.079 degree Celsius.

a Tmin and b Tmax change trends from 2020 to 2010 at RCP8.5

Warming has had a significant negative impact on the dryland ecosystem in the Mediterranean region. Adverse effects have been widely found in regional ecology and sustainable development. For example, warming has led to the northward expansion of semiarid areas in the Mediterranean region (Feng and Fu 2013) and a decline in productivity, mediating the relationship between biodiversity and dryland ecosystem stability (García-Palacios et al. 2018). The warming trend has increased the frequency of droughts and heavy rainfall, aggravated soil erosion and salinization, and led to the depletion of soil carbon stocks (Lagacherie et al. 2018). It has reduced the coverage of biocrusts (Rodriguez-Caballero et al. 2018), accelerated the loss of biodiversity (Verdura et al. 2019), and decreased species richness (Newbold et al. 2020). It has increased the risks of fire (Turco et al. 2018), land degradation and desertification (Yao et al. 2020) and exacerbated environmental problems. It has also increased the risks to water, ecosystems, food and health (Cramer et al. 2018). Warming and drying trends in the Mediterranean region have severely affected crop yields, leading to decreases in barley (Cammarano et al. 2019), olive in Western Europe (Fraga et al. 2020), and sunflower and wheat in the Mediterranean region (Abd-Elmabod et al. 2020). A recent study indicated that the wheat yield would decrease and the wheat price would increase in Egypt under the 2 °C warming scenario (Zhang et al. 2022). There are also some negative impacts on livestock, with shifts and reductions in livestock production in the Mediterranean region due to frequent and intensified droughts resulting from warming (Daliakopoulos et al. 2017).

3.2 NPP Change Trends

Vegetation indices based on remote sensing can reflect the dynamics, greenness, and biomass of vegetation, so they are widely used to assess changes in dryland ecosystems. The net primary productivity (NPP) of vegetation plays a vital role in the carbon cycle by indicating the amount of plant carbon fixed in the atmosphere minus the carbon released by respiration (Ji et al. 2020). According to the global 500-m Terra NPP gap-filling annual data from 2000 to 2020 (Running and Zhao 2021), the annual NPP in the Mediterranean region varied between 0 and 2.1 kg*c/m2. The spatial distribution of NPP (Fig. 8.11a) is closely related to the intensity of aridity and water availability. The areas with a higher NPP are mainly located in the relatively humid European region and the coastal area of North Africa, as well as in the Nile delta where irrigated agriculture is well developed. In contrast, areas with a low NPP are mainly located in the dry Sahara Desert and the central and eastern Anatolian Plateau. The NPP in the Mediterranean drylands showed significant spatial variation. From 2000 to 2020, almost all dryland regions, such as Turkey, Greece, Italy, and northeastern Spain, showed a significant increase in the annual NPP, while the Nile delta had a decreasing trend in the annual NPP (Fig. 8.11b).

Fig. 8.11
2 distribution maps of the Mediterranean region categorize regions into Annual mean N P P and its trend. The annual N P P varied between 0 and 2.1 kilograms of carbon per square meter and its trend varied between negative 100 to 10 grams of carbon per square meter.

Annual mean NPP and its trend in the dryland areas in the Mediterranean region

3.3 Land Cover and Vegetation Changes

We used a chord diagram to describe the land cover and land use conversion from 2000 to 2020 in the drylands of the Mediterranean region (Fig. 8.12). Agricultural land and bare land are the major sources of settlement and indicate the rapid urbanization process in the Mediterranean region. Bareland is converted into agricultural land, which reflects the process of cropland expansion. Agricultural land change is complex, bare land is the primary source of agricultural land, and agricultural land and forest are converted from each other. Moreover, the vegetation significantly changed in the Mediterranean region in recent decades. Between 1999 and 2012, the vegetation cover in the Middle East and North Africa decreased significantly, except in sporadic areas of Algeria and Egypt. Forest extent, structure, and composition in the northern Mediterranean have experienced dramatic changes and have become fragmented (Doblas-Miranda et al. 2017). Vegetation cover and the size and spatial pattern of vegetation patches have a direct impact on the health of dryland ecosystems (Meloni et al. 2020; Meron 2016). Dryland vegetation in the Mediterranean has a unique spatial pattern, ranging from alternating regular bands of vegetation and bare ground to regular gaps of bare ground within a continuous vegetation cover and scattered vegetated spots (Mander et al. 2017). Declining vegetation cover in small and overdispersed patches can lead to a rapid and significant loss of ground arthropod diversity (Meloni et al. 2020).

Fig. 8.12
A chord diagram of land use land cover changes from 2000 to 2020. It exhibits agricultural land, settlement, water, bare land, grassland, wetland, and forest. The highest land is covered by agricultural and bare.

Chord diagram of land use/land cover change (LULCC) changes from 2000 to 2020 in the drylands of the Mediterranean

Egypt is the country with the largest population in the Mediterranean region. At the current stage, the population in Egypt has surpassed 100 million and is experiencing very large pressure on the food supply. Under the pressure of the rapidly growing population, the agricultural area and settlement area of Egypt significantly expanded from 2000 to 2020. The conversion matrix of land cover in Egypt from 2000 to 2020 is shown in Table 8.1. From 2000 to 2020, the agricultural land and settlement land expanded rapidly; in return, the area of bare land decreased by 3941.6 km2. To feed the increasing population, Egypt tried to expand the area of agriculture to improve food production. The agricultural area increased from 54207.4 km2 in 2000 to 55476.2 km2 in 2020, an increase of 2.3%. A total of 3100.9 km2 of bare land was converted into agricultural land. This conversion matrix also reflected the rapid urbanization process in Egypt. From 2000 to 2020, the area of settlement increased from 1506.4 km2 to 3955.8 km2, an increase of 162.6%. Agricultural land and bare land were the primary sources of increased settlement, and 1729.4 km2 of agricultural land and 645.0 km2 of bare land were converted into settlement. Water was the major limiting factor to agricultural development, and a water deficit leads to cropland abandonment. This phenomenon has occurred in Egypt. From 2000 to 2020, a total of 75.9 km2 of agricultural land was converted into bare land in Egypt.

Table 8.1 Land cover conversion from 2000 to 2020 in Egypt

3.4 Crop Structure and Food Production Per Capita Change

A mixture of drought-tolerant crops, cash crops, and livestock dominates the agriculture of the Mediterranean region. The planting area, fruit area, and pastoral production are spatially separated. Wheat, barley, and maize are the main staple crops in the Mediterranean region, and they are mainly grown on flat terrain or in areas with low slopes. The share of food crops (wheat, maize, barley, rice, oats, potatoes, etc.) in Mediterranean countries ranges from 23.0 to 70.9%. Morocco, France, Turkey, and Egypt have 70.9%, 63.8%, 62.1%, and 58.4% of their agricultural land dedicated to food crops, respectively. The other countries have less than 50%, especially Israel, Greece, Portugal, and Lebanon, with only 29.8%, 31.4%, 23.0%, and 28.3% of their area dedicated to food crops, respectively. Olives are the most important cash crop in the Mediterranean region, with shares of 39.6%, 39.3%, 36.9%, 31.5%, 35.5%, and 32.1% in Tunisia, Israel, Greece, Libya, Portugal, and Jordan, respectively. Grapes and vegetables are also significant cash crops in the Mediterranean region. In Israel, for example, vegetables are grown on 10.7% of agricultural land. Cash crops occupy a large proportion of arable land, limiting the cultivation of staple foods and leading to insufficient crop production in the countries surrounding the Mediterranean.

Using crop production data from the FAO and CropWatch monitoring platforms, the change trends of crop production (Fig. 8.13a) and crop production per capita (Fig. 8.13b) were estimated in the countries surrounding the Mediterranean region from 2010 to 2020. Crop production in Egypt, Lebanon, and Algeria showed a significant upward trend from 2010 to 2020, while crop production in Italy, Greece, and Libya showed a significant downward trend from 2010 to 2020, and crop production in the other countries showed a variable trend. As 400 kg per capita per year is a criterion to eliminate food insecurity, all North African and West Asian countries were below this level and have not achieved food self-sufficiency. In contrast, the crop production per capita in France, Spain, Croatia, Bosnia, and Turkey was over 400 kg. The food produced by these countries is enough to meet their own needs and even export to other countries, e.g., France is one of the largest food exporters in the world.

Fig. 8.13
2 distribution maps of the Mediterranean region reveal trends in crop production. Egypt, Lebanon, and Algeria experienced a notable increase from 2010 to 2020, while Italy, Greece, and Libya saw a decline. Spain and France exhibit the highest food production per capita.

a Change in food production and b food production per capita in 2020 and its change from 2010 to 2020

3.5 Water Resource Analysis

Water limitation and scarcity are the greatest challenges to agricultural development in North and West Africa. Significant changes in agro-fruit and pastoral production in the Mediterranean have occurred in the past half-century. Large-scale farms and plantations have gradually replaced traditional small farms and plantations. The development of agriculture has significantly increased water consumption and poses a significant challenge to sustainable development. Based on NASA’s Gravity Recovery and Climate Experiment (GRACE) (Landerer and Swenson 2012), the change in liquid water thickness from 2003 to 2016 is shown in Fig. 8.14. West Asia, Egypt, Tunis, Algeria, and Libya have experienced the issue of declining liquid water thickness. The decline in the water table characterizes this region’s climate, indicating a severe water resource crisis.

Fig. 8.14
A distribution map of the Mediterranean region illustrates changes in liquid water thickness from 2003 to 2016. Regions including West Asia, Egypt, Tunisia, Algeria, and Libya have faced a noticeable decline in liquid water thickness.

Change in equivalent water thickness between 2003 and 2016

Figure 8.15 shows the distribution of the annual ET and its trends from 2003 to 2019. ET intensity is determined by land cover type, with a higher ET intensity in forests, shrubs, and arable land and a lower ET intensity in bare land and deserts due to the arid climate (Fig. 8.15a). Due to developed irrigated agriculture, the ET intensity is higher in the Nile basin and its delta. Figure 8.15b shows the different patterns of ET variation, with a strong increasing trend in the Nile delta and coastal areas of West Asia, Turkey, and Greece and a significant decreasing trend in Spain and the mountainous regions from Morocco to Tunisia.

Fig. 8.15
2 distribution maps of the Mediterranean region categorize areas based on Annual Mean E T and its trend. The annual E T varies between 0 and 3000 millimeters per year, while its trend ranges from negative 245.5 to 108.1 millimeters per year.

a Annual ET spatial distribution and b its trend from 2003 to 2019

Agriculture is the largest user of water in the Mediterranean region. ET represents the actual water loss due to climate change and anthropogenic factors, and separating the contributions of natural and anthropogenic factors to ET variability can provide valuable information for water resource management. This study used a data-driven approach (Zeng et al. 2022) to quantify the impact of natural and anthropogenic factors on ET changes in the Nile basin, Tunisian agro-pastoral, Algerian agro-pastoral, Moroccan agro-pastoral, Libyan agro-pastoral, West Asian agro-pastoral, and Turkish agro-pastoral regions (Fig. 8.16).

Fig. 8.16
A map displays seven major agricultural regions in North Africa and West Asia: Nile Basin, Tunisian Agro-Pastoral Area, Algerian Agro-Pastoral Area, Moroccan Agro-Pastoral Area, Libyan Region, West Asian, and Turkish Agro-Pastoral Area.

Seven major agricultural regions in North Africa and West Asia

The ET separation method first divided ET and environmental factors into a natural group (ET, Xi)n and an anthropogenic group (ET, Xi)a according to the natural and human-managed features of land cover types. Here, Xi included precipitation (P), air temperature (Tair), wind speed (Wind), downward longwave radiation flux (LWdown), downward shortwave radiation flux (DWdown), pressure (Psurf), specific humidity (Qair), elevation (Ele), slope (Slo), aspect (Asp), latitude (Lat), and longitude (Lon). Second, a random forest regressor that optimized the parameters by the grid search algorithm was employed to build the ETn prediction model that explored the linkage between the ET and Xi of the natural group. Third, the ETn prediction model was transferred to predict the ET of agricultural land cover caused by natural factors as Xi of the anthropogenic group as input. Finally, the anthropogenic ET (ETa) of agricultural land cover was separated by calculating the difference between ET and ETn. This approach is explained in Fig. 8.17.

Fig. 8.17
An illustration of natural and anthropogenic Evapotranspiration separation includes predictors, E T, Random Forest Regressor, natural land-use types, agricultural land cover, and anthropogenic E T from natural land cover to agricultural land cover.

The framework of natural ET and anthropogenic ET separation of agricultural land

Here, ET and environmental data came from different sources. The synthesized ET product at a 1-km spatial resolution for 2019 was used in this study (Elnashar et al. 2021a). Synthesized ET data were generated by using the simple average of the Penman‒Monteith-Leuning (PML) (Zhang et al. 2019) and the Operational Simplified Surface Energy Balance (SSEBop) (Senay et al. 2013) remote sensing ET products. Precipitation data were collected from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) (Funk et al. 2015). The CHIRPS is a daily 0.05° × 0.05° (≈5 km) grid cell quasi-global rainfall dataset from 1981. It creates a gridded rainfall product by incorporating remotely sensed precipitation with in situ station data. The LWdown, DWdown, Tair, Wind, and Psurf with a spatial resolution of 0.25° (≈25 km) were provided by the Global Land Data Assimilation System (GLDAS-2.1) (Rodell et al. 2004). Elevation (Ele) information was extracted from the Shuttle Radar Topography Mission (SRTM) version 4 data (Jarvis et al. 2008) with a spatial resolution of 90 m. Slope (Slo) and aspect (Asp) data were calculated from SRTM data using spatial analysis. Land cover data were collected from the European Space Agency (ESA) Climate Change Initiative Land Cover (ESA-CCI-LC) (Bontemps et al. 2012) with a spatial resolution of 300 m. ESA-CCI-LC data extend from 1992 to 2019 with 37 land cover classes (ESA 2015, 2018). The longitude and latitude datasets were generated at a 1-km resolution by spatial analysis.

The ETn prediction model was built independently in B01, B02, B03, B04, B05, B06, and B07. Good performance was found in 7 basins. The R2 values were 0.83, 0.95, 0.98, 0.95, 0.99, 0.98, and 0.93, respectively; the Nash coefficients reached 0.83, 0.95, 0.98, 0.95, 0.99, 0.98, and 0.93, respectively; the mean absolute errors (MAEs) were 16.1 mm, 3.0 mm, 22.9 mm, 20.0 mm, 3.7 mm, and 21.7 mm; and the root mean square errors (RMSEs) were 58.0 mm, 7.6 mm, 34.2 mm, 30.2 mm, 10.9 mm, 35.2 mm, and 51.0 mm, respectively (Table 8.2).

Table 8.2 Performance summary of the ETn prediction model in 7 basins

The increase in ET intensity by agricultural activities was 453.9 mm, 68.0 mm, 56.1 mm, 64.4 mm, 137.8 mm, 93.3 mm, and 105.1 mm for the Nile basin, Tunisian agro-pastoral area, Algerian agro-pastoral area, Moroccan agro-pastoral area, Libyan agro-pastoral area, West Asian agro-pastoral area, and Turkish agro-pastoral area, respectively. The results indicated that ET will increase by 4539 m3, 680 m3, 561 m3, 644 m3, 1378 m3, 933 m3, and 1051 m3 for B01, B02, B03, B04, B05, B06, and B07, respectively, if agricultural land increases by one hectare (Table 8.3). Taking Egypt as an example, from 2000 to 2020, the agricultural land area increased by 1508.12 km2, this meant that the water consumption increased by 684.54 × 106 m3.

Table 8.3 Summary of increased ET consumption by agriculture in 2019

4 Driving Forces of Dryland Change

4.1 Climate Change

The significant warming trend has had a strong negative impact on the extent of drylands, vegetation productivity, biodiversity, and stability of the dryland ecosystem. For example, warming may reduce soil water availability (Schlaepfer et al. 2017), soil fertility (Berdugo et al. 2020), plant productivity (Berdugo et al. 2020; Yao et al. 2020), leaf abundance, and species diversity (Maestre et al. 2016). It may also affect nutrient cycling and soil microbial communities (Maestre et al. 2016) and increase the risks of drought, land degradation, and desertification in dryland regions (Huang et al. 2017; Tietjen et al. 2017). Food security in the Mediterranean region has also been affected by a warming trend. Many studies have reported that warming has a negative impact on crop yield and livestock production due to increased frequency and intensified drought (Abd-Elmabod et al. 2020; Cammarano et al. 2019; Fraga et al. 2020; Mohamed and Squires 2018). New studies have indicated that the warming trend is critical for Syria’s civil war (Selby et al. 2017). First, severe drought caused a significant decline in wheat production and resulted in large-scale migration. The latter exacerbated the socioeconomic stresses that underpinned Syria’s descent into war.

Different climate scenarios have indicated that a significant warming trend would occur in the Mediterranean region (Fig. 8.10). A series of negative impacts of warming on the ecosystem would occur in the future if no reasonable human intervention occurred. As mentioned in Sect. 8.3.1, warming will increase the intensity of aridity and lead to a northward expansion of drylands (Feng and Fu 2013). Warming will increase the frequency of extreme climate events (droughts and heavy rains) and the risk of fire (Turco et al. 2018). Warming may lead to regime shifts and mediate the relationship between the biodiversity and stability of dryland ecosystems (García-Palacios et al. 2018). Warming trends will exacerbate environmental problems and increase the risks to water, ecosystems, food, and health (Cramer et al. 2018). For example, warming will aggravate soil erosion, salinization, soil carbon depletion (Lagacherie et al. 2018), land degradation (Yao et al. 2020), biodiversity loss (Verdura et al. 2019), and species richness loss (Newbold et al. 2020). In the future, the sustainable development of drylands in the Mediterranean region should pay close attention to warming trends and monitor and simulate the consequences caused by warming trends. More importantly, policy-makers should take suitable adaptation measures to reduce or even dismiss the negative impact of warming trends.

4.2 Anthropogenic Drivers

Population, wildfire, overgrazing, grazing abandonment, land intensity, land abandonment, and urban expansion are the main driving forces of dynamic changes in dryland ecosystems in the Mediterranean. North Africa and West Asia are experiencing rapid population growth. The populations of Egypt, Turkey, France, and Italy exceed 50 million, accounting for 21.1%, 17.3%, 14.1%, and 12.7% of the population of the Mediterranean region, respectively. The rapid population growth has led to the massive migration of the population from rural areas to towns and cities (Wolff et al. 2020). The population boom in the dry southern and eastern Mediterranean has put considerable pressure on the food supply that has aggravated the overexploitation of land and water resources (Mohamed and Squires 2018). Warming and water constraints reduce the productivity of cropland, leading to land degradation and abandonment. Land degradation and desertification would significantly reduce crop production, forcing population migration to more productive areas and even causing cross-border migration. Wildfires are another important anthropogenic driving force of dryland ecosystem change in the Mediterranean region. Studies have found that wildfire in abandoned terraces has resulted in significant soil degradation in the Mediterranean region (Lucas-Borja et al. 2018), and large fires led to the transition from Mediterranean oak woodlands to shrubland (Guiomar et al. 2015). Grazing and grazing abandonment also play a crucial role in modifying Mediterranean dryland ecosystems. Overgrazing could significantly reduce vegetation and biocrust cover and increase the risk of bare land exposure. Overgrazing is the main driver of land degradation in Spain, Greece, and Cyprus (Riva et al. 2017). The consequences of grazing on dryland ecosystems are controversial. Some studies have suggested that grazing abandonment could reduce soil fertility, carbon storage, soil multifunctionality, and soil microbial activity (Peco et al. 2017), while other studies found that grazing abandonment could increase the cover of biocrusts and benefit the stability of dryland ecosystems (Rodríguez-Caballero et al. 2018).

5 Summary and Perspectives

This chapter summarizes the characteristics and dynamic trends of Mediterranean dryland ecosystems and the impacts of climate change and anthropogenic drivers on dryland ecosystems. Mediterranean dryland is dominated by low productivity bare land and is experiencing a significant warming trend. Biodiversity, soil nutrients, carbon stocks, and microbial community viability are experiencing harm due to the warming trend. The sustainable development of drylands should pay more attention to the warming trend and predict the consequence of the warming trend. Due to the impact of the warming trend, land and water resources have uneven spatial distributions. North Africa and West Asia face extremely dry climate conditions and deeply suffer from water limitation and pressure from rapid population growth. The phenomena of cropland cultivation, degradation, and abandonment widely exist in the dry regions of North Africa and West Asia, even causing large-scale cross-border migration. Extreme climate events will become more frequent, widespread, and intense under the warming trend. The warming phenomenon may trigger population migration and social unrest. The lack of data and models are major issues for Mediterranean dryland ecosystems. In the future, more models should be developed to simulate the dynamic change in Mediterranean drylands and predict the consequences of dynamic change. Reasonable measures should be taken in case catastrophic consequences occur. Moreover, to understand the changes in the Mediterranean dryland ecosystem, a series of critical products should be produced, such as biological soil crusts, shrub encroachment, and land cover, with a finer spatial resolution.