Climate change is one of the world’s most pressing real threats to the drylands, which may jeopardize food security (FS), that is, physical, social, and economic access to sufficient, safe, and nutritious food by people for an active and healthy life (Samuel et al., 2019) and other sectors of our civilization (Tachiiri et al., 2021; Ukhurebor et al., 2021). Drylands are areas where precipitation is balanced by evaporation from surfaces and evapotranspiration (Middleton and Thomas, 1997). They are generally characterized by sparse vegetation, water scarcity, and unpredictability (Berg and McColl, 2021). The distinct biophysical features of drylands make them highly susceptible (Robertson et al., 2018) and complex to unanimous climate change drivers (Berdugo et al., 2020). Upsurging temperatures, changes in precipitation and rainfall patterns, land use, nutrient availability, atmospheric CO₂ (Classen et al., 2015; Copeland et al., 2017; Schlaepfer et al., 2017; Leisner, 2020), and other greenhouse gases emissions (GHGs) are key driving factors of unprecedented dryland expansion (Maestre et al., 2012; Li et al., 2019; Lian et al., 2021). Drylands are associated with substantial land degradation and extremely vulnerable to severe environmental shocks and socioeconomic crises (Fraser et al., 2011; UNU-WIDER, 2017). Due to anthropogenic change and non-climatic stressors, in tandem with other stimuli, the mean global temperature has increased by ∼1.0°C and is expected to further increase over the next century (IPCC, 2018). As a result, many dryland habitats are faced with severe threats that lead to reduced carbon sequestration and high water scarcity (UNEP, 2007; UNEP-WCMC, 2011; Bradford et al., 2020). Moreover, by the late 21st century, it is projected that ∼78% of dryland expansion will befall under the representative concentration pathways (RCPs) 8.5 scenario in developing countries (Huang J et al., 2016; Huang et al., 2017). The impaired climate–drylands connection could impact FS in all four dimensions: availability, access, utilization, and food system stability, negatively influencing the efforts toward sustainability and ecosystem resilience in Africa (Connolly-Boutin and Smit, 2016; Niles and Brown, 2017; Mbow et al., 2019). Multi-disciplinary investigations are in need to identify effective techniques and practices, including coupled earth-anthropogenic processes in conjunction with careful management and adaptation measures of potential ecological risks, to enable mitigating the repercussions.

Meanwhile, the two dimensions of the nexus approach are interdisciplinary and transdisciplinary (Pahl-Wostl, 2019). By highlighting the trade-offs and synergies between the components, the primary dimension assesses the complexity of linkages among climate, dryland, and food systems. The second dimension strengthens driving forces such as population growth, socio-economic progress, and climate change, as well as innovation, technology, and policies (Endo et al., 2020). Nonetheless, a three-node nexus of climate change–dryland variation-FS leads to complexity. It also apprehends a “wide portrayal” and facilitates bringing in the socio-economic and ecological dimensions. This approach is considered a flexible and open option (Bleischwitz and Miedzinski, 2018). Tools and methodologies are varied and context-specific, but the linkages from climate change to social and environmental impacts are difficult to model, given the unpredictable anthropogenic activities affecting the outcomes (Devereux and Edwards, 2004). Conversely, new techniques are compelled to understand the complexities that lead to abrupt non-linear/correlation between Earth’s systems (Randall et al., 2007; Stephens et al., 2020) and thresholds due to bulky and/or irretrievable effects (Devereux and Edwards, 2004). In addition to their implicitly multi-scale structure, linkage processes are difficult to simulate and/or emulate because they are rarely at the required spatial and temporal scale to establish specific reference as to the underlying changing aspects. To fully comprehend the CDF linkages, key factors (e.g., population growth, agricultural transformations and industrial development, technology and innovations, livelihood shifts, and governance and policy implementation) that drive those nexus complexities must be assessed and described for the entire system through the lens of climate change.

Correspondingly, a wide range of multi-spatiotemporal scale integrated frameworks focused on dryland changes, climate–land–energy–water (CLEW) nexus (Vinca et al., 2021), water–energy–food (WEF) nexus (Kogan et al., 2017; He et al., 2019; Kogan, 2019), water–energy–food–environment nexus (WEFE) (Malagó et al., 2021; Mirzaei et al., 2021), water–energy–food–biodiversity–health (WEFBH) nexus (Hirwa et al., 2021), and others have been set up. However, they are not sufficient anymore (Fernández-Ríos et al., 2021). Instead, current advances in climate change, dryland ecosystem management, and FS are hindered by the limitations of inadequate data on dryland environments and the methodologies commonly used for scientific data analysis, some of which are ill-equipped for capturing complex relationships present in the huge volumes of available data. Coupling large-scale field spatial observations with model simulations is now considered the most viable opportunity and accurate technique to identify dryland ecosystem shifts and evaluate dryland ecosystem stability. But, resistance and recovery after extreme events such as droughts, as a high priority needs urgent attention (Ruppert et al., 2015; Burrell et al., 2017; Wei et al., 2022). Development of using geospatial tools by multiscale frameworks continues to present key fundamental gaps (Fritz et al., 2019). In addition, various methods have been used to assess the influence of extreme events on dryland degradation (Wang et al., 2012; Dubovyk, 2017; He et al., 2019). Global FS requires transdisciplinary responses and interventions at different types of scale (Drimie and McLachlan, 2013), that is, globally (Schmidhuber and Tubiello, 2007; Yadav and Congalton, 2018), regionally (Ingram, 2011), and locally (Moore et al., 2012). With global climate change, dryland variation, and FS, there is the additional challenge of uncertainties, which is unlikely to decrease in the next coming decades (Campbell et al., 2016). There are gaps between research and technology transfer, research and implementation, research and practice, and science and policy. It is, therefore, urgent to seek alternative resources, efforts, and procedures that combine local with emerging scientific knowledge through more effective dissemination of information and technology, appropriate participatory learning, and partnerships.

Few current research on climate change, dryland variation, and FS in Africa have been published (Wheeler and von Braun, 2013; Cervigni and Morris, 2016; Guilpart et al., 2017; Li and Zhang, 2017; Leakey, 2018; Schouten et al., 2018; Nyberg et al., 2019; Chimwamurombe and Mataranyika, 2021). Consequently, developed and developing nations started focusing on new tools and strategies for boosting agricultural production to meet future challenges, and improving or advancing techniques that would help deal with food (in) security and monitor the expansion of drylands (Peng et al., 2021). The apparent potential for developing more holistic and cost-effective tactics, including using existing strategies and procedures as foundations, through developing novel methods that integrate RS and local participation, necessitates a suitable synopsis of dryland dynamics and FS on distinctive spatiotemporal scales.

In a nutshell, this succinct review aims to address both the vexing and progressive threats between climate change, dryland dynamics, and FS through the lens of novel systems approach, advances, challenges, and future opportunities. The CDF nexus provides a strong foundation for scientists, environmental decision-makers, and activists and actors who are interested in achieving all targets of the 17 sustainable development goals (SDGs), particularly SDG 13 (climate action), SDG 15 (use of ecosystem services), SDG 2 (zero hunger), and the Paris Climate Agreement, thereby devising effective policy for action and planetary well-being. This study also proposes a conceptual framework clarifying the interlinkages between influencing systems (i.e., drylands, climate, ecosystems, socio-ecological, and food systems) that consistently unravel and build greater resilience to the confounding vulnerabilities, shocks, and stresses within the food networks.

This study employs various research publications, books, reports, and case studies collected from official websites. Hence, we organize our review into four major aspects and then discuss them using past and current literature. These aspects are: 1) the CDF nexus, including dryland distribution and their associated impacts factors, the relationship between compounded climate, dryland, and FS; 2) technique advances in drylands monitoring methods, including regional observation networks and innovative technologies; 3) challenges and uncertainties for climate change, dryland dynamics, and FS measurements; and 4) future directions and research opportunities to improve dryland ecosystems management and cope with ongoing risks related to FS under climate change conditions.

Climate models (CMs) are weather forecasting extensions. Moreover, these models provide information on hydro-biogeochemical cycles (Foley, 2010; Wang et al., 2015). Scientists utilize the CMs to draw past, current, and future conclusions about complex earth systems (Huang et al., 2017). The most intricate and reliable models for understanding climate systems and forecasting climate change are General circulation models (GCMs) and regional climate models (RCMs), which may need bias corrections and model output statistics (MOS) (Eden and Widmann, 2014). For instance, Keenan et al. (2016) testified that during the last decades, in the warming break of drylands, a current hiatus of crop growth rate was linked to a rise of atmospheric CO₂ in the terrestrial sink, which was attributed to the effects of atmospheric CO₂ on vegetation (Ballantyne et al., 2017). Consequently, because global carbon cycle dynamics are not included in some CMIP5 models, CMIP5 cannot duplicate this trend without significant uncertainty (Huang et al., 2017). Even for state-of-the-art models of global carbon cycling, the carbon concentration still has a lot of uncertainty. The case of the West African monsoon is an example (Klein et al., 2017). Nonetheless, various models are built based on the same modeling institutions. Thus, the ensemble of CMs is not weighted. There are great uncertainties remaining in evaluations of the global trends in dryness and wetness under climate change conditions (Trenberth et al., 2014). To handle the uncertainties in aridity projections and the aridity index (AI) calculation against the hydro-ecological variables, there is a need to consider regions where the overwhelming of models agree in sign (Greve et al., 2019). Moreover, the use of time series precipitation and evapotranspiration datasets from meteorological stations could be helpful to reduce uncertainties in AI projections and regional dryland climate modeling (Tarek et al., 2021).

Dryland climate system uncertainty over human action possesses two main sources, including uncertainty due to unknown future emission concentrations of greenhouse gases and aerosols, and uncertainty of the climate system’s response to our actions (Trenberth and Trenberth, 1992; Smith et al., 2009). This information, combined with climate models, allows decision makers at all levels of governance to determine how both natural and manmade influences have and will impact changes in our climate.

In the past four decades, drastic population development has been observed in drylands (Smit, 2016; Ellis et al., 2021). Subsequently, modern dryland farming and intensive land use are necessary. Sustainable agriculture comprises multiple components, including the introduction of climate-adapted cultivars and sustainable environmental protection that integrates provision and preservation of ecosystem services by enhancing durable intensification programs based on conservation agriculture and community-based adaptation and mitigation with operational support services (e.g., biodiversity, food production, and reduction of GHG emissions) (Mbow et al., 2014; Sanz et al., 2017). Therefore, planning of the so-called food–energy–water–biodiversity–human health (WEFBH) nexus has revealed practicality in evaluating strategic policy to achieve the SDGs prior to the rising demands, dryland resource scarcity, and climate variability (Albrecht et al., 2018; Hirwa et al., 2021).

Remote sensing data have been utilized to provide information in data-scarce areas to address climate variability and FS induced by shifts in foundational dynamic ecosystems. The extension of dryland-specific modern observation models, networks, and evaluations of new RS technologies is a key to successful dryland ecosystem management. These technologies exist in some areas across Africa (e.g., AngoSat-2, NileSat-301, and NARSSCube-2) (Woldai, 2020). Owing to advances in model development from the late 1990s until now, modeling efforts have inspired more current observational investigations. In this instance, measurements are frequently provided apropos of regression models, and multidecadal aerial images are used to identify vegetation changes, for example, in the case of Niger over a forty-year interval. Nonetheless, some models (e.g., Brusselator model) can be overly mechanistic in their representation of many processes at hand, resulting in a high dimensionality that must be calculated from data. As a result, this modeling approach is frequently linked to observation and involves comparisons to field-based assessments (Figure 6). Ultimately, we recommend close collaboration between geo-information data-driven modeling approaches and terrestrial ecosystem modelers to more swiftly categorize model structural deficiencies and hence intrinsically empower more precise dryland ecosystem functioning model projections with the social and ecological system.

Commonly known technologies are categorized into two main types of hardware and software resources, including open-source and affordable tools and scanners, sensors, and platform networks. Therefore, an increasingly growing pool of comparatively low-cost innovations is spurring the transition from catchment to subnational measures (Richardson et al., 2018). The ecosystem phenology camera network can be used to estimate the carbon flux, photosynthesis, and canopy greenness in dryland vegetation (Richardson et al., 2013), and mobile devices can be redeployed to record and capture ecological data. For instance, the Land-Potential Knowledge System (LandPKS) recognizes soil and land types, monitors soil health and vegetation, and identifies management options (Herrick et al., 2016). This could be used to verify remote sensing products, assess earth surface model projections, seasonally explain ecosystem-scale data, and investigate the climate change effects on the terrestrial ecosystem (Seyednasrollah et al., 2019). Public and private institutions can reduce expenditures on design, research, and development, via surplus non-custom devices that are relatively inexpensive and widely available.

At present, dryland ecosystem degradation meets increasingly severe climate change. Increasingly, widespread, frequent, and extreme weather events substantially impact food security, especially the sufficiency and regularity of food production. In this review, the bibliometric approach was used to assess the research trends, which identified that research demand on the impacts of climate change on drylands and FS has been increasing. African drylands harbor enormous exceptional levels of biodiversity via diverse land-use systems and provide a variety of ecosystem services. However, they are ecologically fragile in a plethora of ways. There is a strong relationship between climate change, dryland change, and food systems. With regards to the digital revolution in the RS field, in addition to continuing to use conventional methods to detect the impact of climate change on arid and semi-arid regions, technical innovations (e.g., ecosystem observational/research networks) and modern practices (e.g., climate modeling tools) focusing on dryland changes and FS in Africa are very rare across all sectors. Novel methods, such as coupling different vegetation indices, are urgently needed and encouraged to support conventional dryland RS and FS assessment, from basin to regional scales. With particular efforts to the tactically explored studies, we propose an integrated conceptual framework of different systems (i.e., drylands, climate, ecosystem, socio-economic, and food system) (Figure 3). The foresight and prediction assessment of driving forces of the climate, drylands, and FS needs further research (Section 6). The framework has the potential to reveal new insights into climate change, dryland ecosystem dynamics, and FS with the availability and accuracy of data in the entire system. The nexus approach combines intradisciplinary sections involving all socioeconomic and ecological fields to better understand the regional impacts and develop adaptive strategies while mitigating the climate change impacts on drylands. Herein, we propose new research opportunities to strengthen the CDF nexus: ① promoting sustainable agricultural best management practices and innovations as a tool to enhance community resilience and cope with climate change impacts on FS, ② using modern observational data and developing idealistic models to better understand the CDF nexus approaches, and ③ strengthening dryland research and management effectiveness through emerging and affordable technologies. By combining these research directions, we may gain new insights into dryland dynamics, ecosystem services, and FS. We recommend decision makers design policy instruments that consider CDF fields as a multidisciplinary nexus.