Introduction
The physical, social, cultural, and spiritual well-being of humans is highly dependent on nature (Diaz et al., Reference Diaz2018; Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021). Nature encompasses not only organisms and their ecosystems, but also ecological and evolutionary processes on Earth, resulting in both positive and negative consequences for humans and their quality of life (IBPES, Reference Brondizio, Settele, Díaz and Ngo2019). Nature’s contribution to people (NCP) has been defined as ‘all the contributions, both positive and negative, of living nature’ (diversity of organisms, ecosystems, and their associated ecological and evolutionary processes) to people’s quality-of-life (Diaz et al., Reference Diaz2018). Nature’s contributions can be organised broadly into three categories: material contributions (e.g., food and energy), non-material contributions (e.g., recreation, spiritual services and experiences), and regulating contributions (e.g., clean air and good water quality). Material and regulating contributions are similar to the elements captured in the ecosystem services paradigm, whereas non-material contributions include attributes that relate to the quality of life, belief systems, or nature-based experiences (Diaz et al., Reference Diaz2018; Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021).
Implicit in the NCP concept therefore is not only what nature can provide, but also what people need; social justice, spiritual beliefs and their links to the natural environment (Pascua et al., Reference Pascua, McMillen, Ticktin, Vaughan and Winter2017), regardless of whether these needs are realised (Chaplin-Kramer et al., Reference Chaplin-Kramer2019). The magnitude of these contributions, therefore, would be expected to be greater where people have the strongest association with nature and the greatest needs (Chaplin-Kramer et al., Reference Chaplin-Kramer2019). Therefore, we would expect stronger contributions to human societies where people have a closer connection with nature or derive a living from the land. Such associations would be expected to be greater in drylands, where 90% of the human population has relatively low standards of living (Reynolds et al., Reference Reynolds, Stafford Smith, Lambin, Turner, Mortimore, Batterbury, Downing, Dowlatabadi, Fernández, Herrick, Huber-Sannwald, Jiang, Leemans, Lynam, Maestre, Ayarza and Walker2007). Yet, previous global assessments (e.g., Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023) suggest that it is in arid and semi-arid environments and developing countries, mostly drylands, where ecosystem degradation fails to sustain the needs of people, i.e., areas with the greatest benefit gaps (sensu Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021) compared with wealthy, less marginalised communities.
Despite global assessments of NCP (Chaplin-Kramer et al., Reference Chaplin-Kramer2019), we have a relatively limited understanding of how nature contributes to the lives and welfare of people in drylands. Drylands are important because they account for almost 40% of the terrestrial land area, and are home to about 38% of the world’s population (2.5 billion people; Huang et al., Reference Huang, Li, Fu, Chen, Fu, Dai, Shinoda, Ma, Guo, Li, Zhang, Liu, Yu, He, Xie, Guan, Ji, Lin, Wang, Yan and Wang2017). Moreover, large areas of drylands are devoted to primary production, particularly fodder production for livestock (Prăvălie, Reference Prăvălie2016). This makes them more vulnerable to environmental changes associated with increasing aridity (Huang et al., Reference Huang, Li, Fu, Chen, Fu, Dai, Shinoda, Ma, Guo, Li, Zhang, Liu, Yu, He, Xie, Guan, Ji, Lin, Wang, Yan and Wang2017) than other, more mesic environments (Berdugo et al., Reference Berdugo, Gaitán, Delgado-Baquerizo, Crowther and Dakos2022). Many people in drylands are marginalised, often in areas of political conflict (Global Conflict Tracker, 2023), with low standards of living and sometimes poor nutrition (Prăvălie, Reference Prăvălie2016). Areas where overall contributions are low have been shown to be associated with transitional climates (Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023), reflecting ongoing degradation and declines in nature itself, through, for example, land clearing, desertification, and atmospheric pollution. This suit of physical and environmental conditions likely places drylands at a greater risk of famine and global tragedy. Exploring how nature can contribute to people in drylands is critical if we are to balance the competing needs of people and the natural environment, a more equitable human society, and work towards achieving sustainable development of drylands.
Here we report on a study where we used environmental, social and biological data as surrogates for 18 contributions that nature can make to people in drylands (Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021). These contributions comprise seven regulating, six material, and five non-material categories. Drylands are defined as areas where the ratio of evaporation to average annual precipitation exceeds 0.65 (MEA, 2005), including dry subhumid, semi-arid, arid and hyper-arid areas. Previous studies have focussed on nature’s contributions at global scales (Chaplin-Kramer et al., Reference Chaplin-Kramer2019; Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021) or explored the service provision of drylands rather than associating provision and people’s needs (Maestre et al., Reference Maestre, Le Bagousse-Pinguet, Delgado-Baquerizo, Eldridge, Saiz and Berdugo2022). Traditional ecosystem service approaches have focussed only on regulating and provisioning contributions such as primary production, carbon, and food, but neglect the actual non-material needs of dryland people.
Our study links the provision of tangible goods and services with human needs, endeavouring to provide insights into the connection between potential contributions from nature and the capacity of people in drylands to use these contributions. We asked the following two questions: First, does the average contribution to people in drylands differ from that in non-drylands, and if so, what is the nature of the difference in these contributions? We posit that drylands would have a lower overall (average) contribution. Our rationale is that drylands are less densely populated, and their environmental resources are more degraded and susceptible to global changes than non-drylands (Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021). Second, have drylands exhibited greater temporal declines in contributions over the past decades (1992–2018) than non-drylands? We would expect the affirmative, given the generally greater declines in environmental quality in drylands than non-drylands over the past half century, though this could be masked by a larger population size and therefore greater human need over that period.
Methods
We used the datasets of Liu et al. (Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023); see Supplementary Text S1), and the assessment is briefly described as follows. A general simplified flow chart illustrating the process of calculating nature’s contribution to people using air quality regulation (NCP3) as an example is presented in Figure 1.
Air quality regulation (NCP3) as an example of a general simplified process for calculating nature’s contribution to people.
Spatial datasets
We reclassified the European Space Agency Climate Change Initiative-Land Cover (ESA CCI-LC) product (European Space Agency 2018), as the core data indicating the change of nature for NCP assessment. Complex ecosystem classifications reduce the accuracy of some NCP calculations, so ecosystem classes need to be consolidated and reclassified to harmonize different terminologies. Reclassification and consolidation can simplify different terminologies prior to analyses, such as combining multiple forests into a single forest class. This process simplifies the computational steps and permits a more rapid assessment of NCP. We used 20 spatial datasets to make the 18 NCP assessments. Most raster datasets had spatial resolutions finer than 10 km, providing sufficient pixels for each sub-basin unit. Maps of the richness of mammals, birds and amphibians (Jenkins et al., Reference Jenkins, Pimm and Joppa2013) at a resolution of 10 km were downloaded from BiodiversityMapping.org. The Global Inventory Modelling and Mapping Studies (GIMMS) provided the vegetation leaf area index (LAI)3g product at a spatial resolution of 1/12 arc degrees (Zhu et al., Reference Zhu, Bi, Pan, Ganguly, Anav, Xu, Samanta, Piao, Nemani and Myneni2013). The global human settlement layer (GHSL) was downloaded from the Joint Research Centre (JRC) and included grids for built-up areas, populations, and settlements (Corbane et al., Reference Corbane, Pesaresi, Kemper, Politis, Florczyk, Syrris, Melchiorri, Sabo and Soille2019). The gross primary production (GPP) dataset was estimated using a revised light use efficiency model, with a spatial resolution of 0.05 arc degrees (Zheng et al., Reference Zheng, Shen, Wang, Li, Liu, Liang, Chen, Ju, Zhang and Yuan2020). The vectorized Global Mangrove Watch (GMW) datasets were transformed into 1 km spatial resolution data (Bunting et al., Reference Bunting, Rosenqvist, Hilarides, Lucas, Thomas, Tadono, Worthington, Spalding, Murray and Rebelo2022) and evapotranspiration (ET) was a synthesized product with a 1 km spatial resolution (Elnashar et al., Reference Elnashar, Wang, Wu, Zhu and Zeng2018). The MODIS (Terra Moderate Resolution Imaging Spectroradiometer Land Water Mask (MOD44W)) Version 6 data product was accessed from the Land Processes Distributed Active Archive Centre (LP DAAC) with a spatial resolution of 250 m (Carroll et al., Reference Carroll, DiMiceli, Townshend, Sohlberg, Elders, Devadiga, Sayer and Levy2017). Annual streamflow maps were obtained from the FLO1K dataset at a spatial resolution of 1 km (Barbarossa et al., Reference Barbarossa, Huijbregts, Beusen, Beck, King and Schipper2018).
A pesticide risk score, based on the most popular active pesticide ingredient, was at a spatial resolution of 1/12 arc degrees (Tang et al., Reference Tang, Lenzen, McBratney and Maggi2021). The soil erosion score was evaluated based on studies by Liu et al., (Reference Liu, Fu, Liu, Zhao and Wang2019) at a spatial resolution of 1/12 arc degrees. The Harmonized World Soil Database was at a spatial resolution of 1 km (Fischer et al., Reference Fischer, Nachtergaele, Prieler, van Velthuizen, Verelst and Wiberg2008). Slope and elevation data were obtained from the Shuttle Radar Topography Mission digital elevation model at a resolution of 3 arc seconds (Jarvis et al., Reference Jarvis, Reuter, Nelson and Guevara2008). The aridity index (AI) was determined as the relationship between precipitation and evapotranspiration and mapped at a resolution of 30 arc seconds (Trabucco and Zomer, Reference Trabucco and Zomer2019). Floodplain data were at a 250 m resolution (Nardi et al., Reference Nardi, Annis, Di Baldassarre, Vivoni and Grimaldi2019). Data on the yield and aggregated value of crop production were derived from the Spatial Production Allocation Model dataset in 2010 (SPAM 2010) at a spatial resolution of 1/12 arc degrees (Yu et al., Reference Yu, You, Wood-Sichra, Ru, Joglekar, Fritz, Xiong, Lu, Wu and Yang2020). The “best crop” map that indicated the maximum achievable bioenergy yields was derived from the dataset of lignocellulosic bioenergy crops at a spatial resolution of 0.5 arc degrees (Li et al., Reference Li, Ciais, Stehfest, van Vuuren, Popp, Arneth, Di Fulvio, Doelman, Humpenöder, Harper, Park, Makowski, Havlik, Obersteiner, Wang, Krause and Liu2020). Aboveground carbon biomass density data were derived from a 2010 harmonized map at a spatial resolution of 300 m (Spawn et al., Reference Spawn, Sullivan, Lark and Gibbs2020). Nighttime light data were obtained from a harmonized dataset from two satellites at a spatial resolution of 30 arc seconds (Li et al., Reference Li, Ciais, Stehfest, van Vuuren, Popp, Arneth, Di Fulvio, Doelman, Humpenöder, Harper, Park, Makowski, Havlik, Obersteiner, Wang, Krause and Liu2020). The locations of natural and mixed world heritage sites were obtained from WHC.UNESCO.org. The vector road dataset was downloaded from the Socioeconomic Data and Applications Centre (SEDAC) and named Global Roads Open Access Data Set, Version 1 (gROADSv1; SEDAC 2013). We applied the HydroBasin level 06 in the HydroATLAS database to take advantage of the nested sub-basins at multiple scales for regionalization (Linke et al., Reference Linke, Lehner, Ouellet Dallaire, Ariwi, Grill, Anand, Beames, Burchard-Levine, Maxwell, Moidu, Tan and Thieme2019). To accommodate the spatial resolution of the various spatial datasets described above, the units smaller than 500 km2 were merged into adjacent largest units to include more than four pixels of 1/12 arc degree raster data in a basin. This resulted in a database of 15204 basin units.
Spatial assessment
The assessment of NCPs uses an indicator-based approach with two indicators: 1) nature’s potential contribution, and 2) nature’s actual contribution to people (Table 1). Nature’s potential contribution relates to the potential to provide resources, services, knowledge or inspiration. For example, nature contributes to the regulation of crop pests (NCP10) by supporting a diverse community of birds (Mayne et al., Reference Mayne, King, Andersen and Elkinton2023). This contribution depends on whether a given basin unit supports crops that require this pest regulation or whether there are people who can benefit from this pest regulation. Although the potential contribution may be large, the actual contribution may be zero, due to an absence of people or crops, for example, the inability of people to use products derived from nature. Because actual human requirements from nature generally increase as population size increases, we set the population as static so that we could observe changes in NCPs driven by nature changes alone, i.e., in the absence of population increase. Put simply, increases in human requirements could lead to an increased NCP assessment, which could mask any potential threats of natural ecosystem loss, and lead to perverse landscape management outcomes (Chaplin-Kramer et al., Reference Chaplin-Kramer2019).
Description of nature’s actual and potential contribution to people (adapted from Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023)
The data sources indicate the source of information used to assess both the potential and actual contribution to people, as well as the parameter (weighted parameter) used in the calculation of actual contribution.
The indicator framework was used to calculate a globally rapid assessment of all NCPs to identify spatiotemporal heterogeneity of the distribution rather than simulating a defined value for biophysical units. In order to develop a rapid assessment framework, no more than three global parameters were used for each NCP, except for hazard regulation (NCP9), which required four parameters to adequately parameterize. Details of the procedures and datasets used to calculate each NCP are given in Supplementary Text S1 and Figure 1. The lowest values of the parameters were assigned a value of 0, and the threshold value of 1 was set as the 90th percentile value of each originally assessed NCP value in 1992. All the values exceeding the threshold should be assigned as 1. By min-max normalization, the normalized value of every NCP was in the range of 0–1. Note that we did not change people’s needs between 1992 and 2018 (see Supplementary Text S1).
Linear models (Bates et al., Reference Bates, Mächler, Bolker and Walker2015) were used to examine differences in mean NCP values between drylands and non-drylands in relation to 1) six continents, 2) individual contributions, and 3) between 1992 and 2018. We tested for the correlation between the value of each NCP and population size using Pearson’s r. Analytical tests were performed in the R statistical software (R Core Team, 2021) prior to linear modelling to ensure that the data met the necessary assumptions implicit in linear modelling.
Results
Nature generally contributes less to people in drylands
The global average value of NCP was about 30% lower in drylands than in non-drylands (χ2 = 47.3, df = 1, 114, P < 0.001, Figure S1), consistent with our prediction. Nature’s contributions to drylands were significantly lower for Africa, Asia, Oceania and South America, but there were no differences for Europe or North America (dryland/non-dryland by continent interaction: χ2 = 15.2, df = 5, 114, P = 0.009; Figure 2). There was a small (albeit non-significant; P > 0.21) decline in NCP with increasing continent size for drylands, but not for non-drylands. We also found evidence of an increase in the magnitude of NCP with increasing population size, particularly for air quality regulation (NCP3), food (NCP12), medicine (NCP14) and learning/inspiration (NCP15; Table S1). Identity (NCP17) declined strongly in both drylands and non-drylands with increasing population size (Table S1).
Nature’s contribution to people (mean ± SE) for drylands and non-drylands by continent. Asterisks indicate a significant difference between dryland and non-dryland at P < 0.05. The number of dryland and non-dryland basins covered by each continent is as follows: Asia has 2,241 dryland basins and 2,774 non-dryland basins; Africa has 2,456 dryland basins and 888 non-dryland basins; Europe has 285 dryland basins and 924 non-dryland basins; North America has 677 dryland basins and 2,274 non-dryland basins; South America has 577 dryland basins and 1,204 non-dryland basins; and Oceania has 848 dryland basins and 56 non-dryland basins.
We then focused on the average contribution to people across global drylands for different locations, i.e., the average value across all 18 contributions and considered both potential (Figure 3a) and actual (Figure 3b) contribution. We found extensive areas of low actual NCP in North Africa (Algeria, Libya, Niger, Mauritania, Mali, Chad, Egypt, northern Sudan, northern Ethiopia), West Africa (Namibia and South Africa), the west coast of South America (northern Chile and Patagonia), much of inland central Australia, the Arabian Peninsula, western Eurasia (Afghanistan, Iran, Turkmenistan), and west-central China and Russia. Conversely, high values were more insular and occurred in southern India, north-eastern China, the Iberian Peninsula, western Turkey, south-eastern South Africa, north-western USA, and a narrow strip in north-eastern Brazil and coastal eastern Australia (Figures 3a and 3b). Although potential and actual contributions were spatially similar overall, actual contributions were greater for the Iberian Peninsula, the Indian subcontinent, and the eastern side of the Eurasian drylands (Figures 3a and 3b).
Mean (a) potential contribution and (b) actual contribution of nature to people in global drylands (average of 18 NCP categories).
There were, however, large differences between drylands and non-drylands for specific contributions. For example, drylands contributed less to six of the 18 NCP categories, i.e., regulation of air quality (NCP3), climate (NCP4), water quantity and flow (NCP6), soil protection (NCP8; mainly in Asia and Oceania, Table S2), woody material (NCP13) and options (NCP18; Figure 4).
Mean contribution of each of the 18 NCP categories for drylands and non-drylands. Asterisks indicate a significant difference in contribution value between dryland and non-dryland at P < 0.05.
Spatiotemporal changes in nature’s contributions
We found a general decline in NCP between 1992 and 2018 across all contributions and for both drylands and non-drylands (−0.47 ± 0.71%, mean ± SE) but this masked the changes in some contributions. For example, the average contribution by nature declined more in drylands than non-drylands for 10 contributions: habitat (NCP1), pollination (NCP2), oceans (NCP5), water quality (NCP7), soil protection (NCP8), hazard regulation (NCP9), pest regulation (NCP10) and bioenergy (NCP11), medicine (NCP14) and experience (NCP16), but increased for climate (NCP4), water quality/flow (NCP6), food (NCP12), woody material (NCP13) and options (NCP18; Table 2, Table S2), again consistent with our second prediction.
Percentage change in NCP for drylands and non-drylands between 1992 and 2018 and the dryland trend.
We also detected some spatial changes over the 26 years. The value of climate regulation in drylands increased in north-central and southern Africa, northern and south-western Australia, northern India, western Iran and western USA, but declined in central Australia and western China (Figures 5 and 6). For water quantity/flow regulation, we detected increases in north-central Africa, the Arabian Peninsula, northern Australia, much of mainland China, India and Iran, but declines were evident in northern and southern Africa, central, northern and eastern Australia, the Iberian Peninsula, the western USA, and the west coast of South America. Similarly, there were some spatial declines in the value of pest regulation (NCP10) in drylands across extensive areas of Africa, the western USA and the western coast of South America, western Iran, northern India, central China, and large areas of Africa and central Australia.
Spatiotemporal change in NCP1-NCP8 between 1992 and 2018 for drylands.
Spatiotemporal change in NCP9-NCP18 between 1992 and 2018 for drylands.
Discussion
We used an indicator framework to compare NCP in drylands with non-drylands. Unsurprisingly, the magnitude of this contribution was about 30% lower in drylands. These differences, however, were inconsistent across continents, with significantly lower values for drylands in Asia, Oceania, Africa and South America, but no difference in Europe and North America. Furthermore, we identified some hotspots of low contribution in North Africa, the Arabian Peninsula, central Australia, and west-central China, and high values in southern India, north-eastern China, the Iberian Peninsula, eastern Australia, and the north-west coast of the United States of America. Finally, potential and actual NCP values were similar, except for the heavily populated areas in Spain, India and China. Our results are consistent with the understanding that NCP is likely to be lower where the quality of the natural ecosystem or its capacity to produce is low (Chaplin-Kramer et al., Reference Chaplin-Kramer2019) and in sparsely populated drylands where the capacity of people to use nature’s products is low (Brauman et al., Reference Brauman, Garibaldo and Polansky2020). Our results also suggest that the magnitude of nature’s contribution globally will decline as drylands expand at the expense of non-drylands.
A spatial understanding of NCP in global drylands
Within those continents with a lower drylands contribution, we found that the reduction in contribution was due largely to a reduction in the magnitude of regulating contributions such as climate (NCP4), water quantity and flow (NCP6), soil protection (NCP8) and the production of woody material (NCP13; Figure 4), reflecting a generally stronger reliance upon primary resources by drylands in contrast to non-drylands (Brauman et al., Reference Brauman, Garibaldo and Polansky2020; Hill et al., Reference Hill, Díaz, Pascual, Stenseke, Molnár and Van Velden2021).
Three dryland areas characterised by low levels of regulating contributions and sparse population densities are North Africa (e.g., Algeria, Tunisia, Libya, and Egypt), the Arabian Peninsula, and central Australia (Figure 3). Low levels of climate regulation (NCP4) across these three areas result from the sparse forest and limited mid- and groundstorey cover (< 5% Maestre et al., Reference Maestre, Benito, Berdugo, Concostrina-Zubiri, Delgado-Baquerizo, Eldridge, Guirado, Gross, Kéfi, Le Bagousse-Pinguet, Ochoa-Hueso and Soliveres2021) dominated by short stature woody perennials and low stature herbaceous biomass (Fischer and Turner, Reference Fischer and Turner1978; Stafford Smith and Morton, Reference Stafford Smith and Morton1999; Le Houerou, Reference Le Houerou2000; Brinkmann et al., Reference Brinkmann, Dickhoefer, Schlecht and Buerkert2011), but they often support a high plant species diversity (Maestre et al., Reference Maestre, Benito, Berdugo, Concostrina-Zubiri, Delgado-Baquerizo, Eldridge, Guirado, Gross, Kéfi, Le Bagousse-Pinguet, Ochoa-Hueso and Soliveres2021). Intense browsing and grazing by livestock, the dominant land use in drylands, reduces plant cover (e.g., Brinkmann et al., Reference Brinkmann, Patzelt, Dickhoefer, Schlecht and Buerkert2009), thus reducing the potential for capture of greenhouse gases and increasing climate-driven consequences for humans (Brauman et al., Reference Brauman, Garibaldo and Polansky2020). Vegetation cover and biomass are also critical parameters that influence the generation of aerosols, which are high over the Arabian Peninsula (Tandule et al., Reference Tandule, Gogoi, Kotalo and Suresh Babu2022) and North Africa (Gherboudj et al., Reference Gherboudj, NaseemaBeegum and Ghedira2017). It is unsurprising, therefore, that these three regions have a relatively lower capacity to support stable soils (NCP8) or extensive wood production (NCP13). The potential to produce wood suitable for sawmilling (NCP13) is also low due to the predominance of lower stature vegetation (shrublands at the expense of forests), highly variable precipitation, and high evapotranspiration (Stafford Smith and Morton, Reference Stafford Smith and Morton1999). The only substantial difference in Europe was the lower value for woody material (NCP13) in drylands than non-drylands (Table S2), reflecting the dominance of short stature xerophytic shrubs with low potential for forestry in the drylands of southern Spain, southern Italy and west-central Poland. Importantly, yields of woody material are likely to decline due to the increased risk of droughts and wildfires in Europe exacerbated by changing climates (Górriz-Mifsud et al., Reference Górriz-Mifsud, Ameztegui, González, Trasobares, Hetemäki, Kangas and Peltola2022).
Large areas of North Africa remote from coastal influences are mapped as having low actual values of water quantity and flow regulation (NCP6, Figure 3b). Many North African countries face severe environmental challenges due to water scarcity (Hamed et al., Reference Hamed, Hadji and Redhaounia2018), which compromises agricultural industries that rely heavily on water supply (Radhouane, Reference Radhouane2013). Surface and groundwater sources are sparsely distributed in North Africa and the Arabian Peninsula (Siebert et al., Reference Siebert, Kummu, Porkka, Döll, Ramankutty and Scanlon2015), and surface water is scarce in central Australia, where it is held for only short periods in isolated depressions and ephemeral waterways (Brim Box et al., Reference Brim Box, Leiper, Nano, Stokeld, Jobson, Tomlinson, Cobban, Bond, Randall and Box2022). Consequently, most perennial vegetation is dependent entirely on groundwater (Eamus et al., Reference Eamus, Froend, Loomes, Hose and Murray2006). Large areas of the Arabian Peninsula also lack surface water but have the capacity to access aquifers recharged from sporadic river flooding (UNDP/RBAS, 2013). Overall, these three examples of drylands are more sensitive to increasing dryness associated with climate change than non-drylands.
Implicit in the NCP concept is population size, and therefore potential contribution to people. We found generally positive relationships between NCP values and population size (Table S2), consistent with our understanding that population size and ecosystem production are positively correlated (Luch, 2007). Our three focal drylands are all relatively sparsely populated, with densities of 0.1, 1 and about 4 people km-2 for central Australia, the Arabian Peninsula and North Africa, respectively (Gapminder–Systems Globalis, 2022). Values of some NCPs (e.g., air quality, food production, medicine, pest regulation, and learning inspiration) were significantly related to population density in both drylands and non-drylands (Table S2). However, identity (NCP17) declined with increasing population size, possibly reflecting the alienation of traditional knowledge at large spatial scales or where populations are changing rapidly (Darvill and Lindo, Reference Darvill and Lindo2015).
Average contribution values for drylands in two continents, Europe and North America were similar to values in non-drylands. North American and European (southern Spain, Sicily) drylands are densely populated, have relatively large GDPs, and highly mechanised primary production (Al Shamsi et al., Reference Al Shamsi, Compagnoni, Timpanaro, Cosentino and Guarnaccia2018; Baur and Iles, Reference Baur and Iles2023; Martínez-Valderrama et al., Reference Martínez-Valderrama, Gartzia, Olcina, Guriado, Ibanez and Maestre2024). For example, the drylands in Almeria, on the Iberian Peninsula in Spain support a mixture of wooded Mediterranean forest and grassland located within a matrix of industrial agriculture such as greenhouses and irrigated agriculture (El Ghafraoui et al., Reference El Ghafraoui, Quintas-Soriano, Pacheco-Romero, Murillo-López and Castro2023) and support a moderate population density of about 80 persons per km-2. This is reflected in the high value of NCP12 (food) in drylands (Table S2). Extensive areas of farmland in Spain are located near Córdoba and Seville, the most developed locations since antiquity (Martinez-Valderrama et al., 2023), and this area is regarded as a food bowl for Europe (Ayuda and Pinilla, Reference Ayuda and Pinilla2021). Furthermore, desert regions of Almeria are highly iconic and display unique landscape features (‘badlands’ Zgłobicki et al., Reference Zgłobicki, Poesen, Joshi, Sóle-Benet, De Geeter, Singh, Wei and Anand2021) that many city dwellers will not normally experience. People prefer these natural, albeit highly eroded, landscapes more than greenhouses. This likely reflects the high value that the population places on natural landscapes and landscape diversity, which should be reflected in learning and inspiration (NCP15) and identity (NCP16). North American drylands are also highly developed, support large urban centres, and include iconic desert environments with extensive natural and mixed ecosystems (NCP1) with potential for bioenergy production (NCP11; Nabhan et al., Reference Nabhan, Riordan, Monti, Rea, Wilder, Ezcurra, Mabry, Aronson, Barron-Gafford, García, Búrquez, Crews, Mirocha and Hodgson2020), and areas that are accessible to people for experience of the natural world (NCP16, Table S2).
A greater decline in NCP in drylands
The magnitude of nature’s contribution has declined markedly over the past half century (e.g., Brauman et al., Reference Brauman, Garibaldo and Polansky2020; Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023), and results between 1992 and 2018 indicate a substantially greater decline in drylands than in non-drylands (Table 2, Figures 5 & 6). Importantly, the greatest declines were for pollination (NCP2, 65% decline), soil protection (NCP8, 56%), hazard regulation (NCP9, 42%), pest regulation (NCP10, 54%), medicine (NCP14, 59%), and experience (NCP16, 42%). Potential contributions have declined for virtually all regulating contributions, e.g., plant pollination and pest regulation (Potts et al., Reference Potts, Imperatriz-Fonseca, Ngo, Aixen, Biesmeijer, Breeze, Dicks, Garibaldi, Hill, Settele and Vanbergen2016), and most declines have been due to a loss in environmental quality (e.g. Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023). Non-material declines are also evident, for example, with increased urbanisation removing local communities and indigenous people from their connections with the land and natural environments (Soga and Gaston, Reference Soga and Gaston2016).
Many of nature’s contributions, particularly material contributions, are based on vegetation-related proxies. One might expect, therefore, a generally lower contribution in drylands than non-drylands, though this was not always the case (e.g., Figure 4). Improvements in database quality and the availability of more specialised data on different contributions at finer spatial scales in drylands should lead to a more reliable assessment of the relative differences between drylands and non-drylands, particularly if new proxies are more closely aligned to particular contributions. It is clear that the benefits accruing from nature are likely to be greatest where nature is most intact (Chaplin-Kramer et al., Reference Chaplin-Kramer2019), suggesting that areas suffering from environmental degradation will contribute less. The consequences of increasing aridity are that nature’s contributions to drylands will continue to decline, particularly for dryland types that are most susceptible to changing climates. Distinct dryland sub-types are likely to respond differently to climate change (e.g., hyper-arid compared with dry subhumid) simply because nature’s contribution depends on both the potential contribution (which is dependent on vegetation and therefore rainfall) and realised contribution (lower population sizes and therefore lower demand for material, non-material and cultural contributions). Thus, a more detailed assessment of different dryland subtypes would likely reveal how increasing global dryness might alter nature’s contributions. Our results indicate that any declines in the environmental quality of drylands will have not only environmental implications but will impact human health (medicine) and the physical and psychological experiences that humans derive from nature.
Conclusions
We used relatively predictable, intuitive, yet simple proxies to calculate NCP in drylands. We acknowledge, however, that our capacity to improve these estimates is hampered by the lack of available databases at the scale commensurate with drylands and non-drylands, and/or the lack of more nuanced information that is more closely aligned with a given contribution. This is particularly relevant for non-material contributions that relate to belief systems or personal experiences. Thus, our assessments can only be based on global databases and remotely sensed, broad-scale proxies. Advances in remote sensing technologies and access to databases at finer spatial scales should allow us to refine our assessment of nature’s contribution in drylands, across large areas where data are sparsely distributed. Nonetheless, our study demonstrates that lower contributions to people in drylands can be attributed to the declining quality of environmental resources in natural systems (Liu et al., Reference Liu, Fu, Wang, Rhodes, Li, Zhao, Li, Zhou and Wang2023; Table S1). The value of these attributes declines with declining rainfall and increasing dryness, yet their value (realised and potential) also increases with increasing population pressure. Predicted large-scale increases in aridity, combined with marked population increase and therefore accelerated land degradation (Prăvălie, Reference Prăvălie2016) are likely to place increasing pressure on nature to contribute to the physical well-being and function of drylands, its biota and people.
Open peer review
For open peer review materials, please visit http://doi.org/10.1017/dry.2024.2.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/dry.2024.2.
Acknowledgments
This work was supported by the Hermon Slade Foundation, the National Natural Science Foundation of China (41991235, 41991232), and the Fundamental Research Funds for the Central Universities in China.
Author contribution
David Eldridge and Yanxu Liu designed the research and Chenxu Wang performed the analyses. David Eldridge drafted the manuscript. Yan Li, Jingyi Ding, Changjia Li, and Xutong Wu contributed to the design and editing of the manuscript.
Competing interest
The authors declare that they have no conflict of interest.
Open access
Comments
THE UNIVERSITY OF
NEW SOUTH WALES
DAVID ELDRIDGE
Professor
School of Biological,
Earth and Environmental Sciences
Dear Editors
Please find attached the manuscript ‘Nature’s contribution to people in drylands’ which we would like you to consider publishing in PRISMS Drylands.
Nature provides substantial goods, services and knowledge to people, and without nature, humans could not survive. The notion of nature’s contribution to humans is a broadening of the ecosystem services concept, but as well as considering physical services, it considers well-being, and social and spiritual aspects. Recent papers have highlighted the global contributions and a few have mapped the global extent. None has specifically compared drylands with non-drylands. Our paper aims to do so.
We found that the contribution to humans was substantially lower in drylands than non-drylands. However, drylands in Asia, Oceania, Africa and South America had lower contributions, but Europe and North America were similar. Moreover, most of the differences were due to material and regulating contributions. A better understanding of nature’s contributions is critical if we are to achieve sustainable development in drylands.
The concept of nature’s contribution to people is relatively new, and the methodology used in this paper has come under some criticism from various fields, particularly authors associated with more recent Science and Nature papers. We do not want the fate of this manuscript to rest on a debate about which database is better or whether different data sources might produce slightly different result. Catchment- or watershed-scale data are needed to discriminate between drylands and non-drylands; but these data often show high variability. We ask, therefore, that the Editor keeps this in mind when choosing appropriate reviewers. We would see the value of this manuscript in its broad conclusions rather than the specific nuances of data usage.
The enclosed work has not been published or accepted for publication and is not under consideration for publication in another journal or book. All authors have read and approved the submitted version of the manuscript, and all the persons entitled to authorship have been named.
Thank you for considering this manuscript for publication in PRISMS Drylands. We hope that the contents of the manuscript will interest the journal’s readership.
Yours sincerely,
David J. Eldridge for the authors
March 6, 2024