Biological soil crusts (biocrusts), covering approximately 12% of the global land surface and a quarter of dryland soil surfaces, play a crucial role in dryland ecosystem functioning. They are complex communities of soil particles, photosynthetic organisms (photoautotrophs), and other organisms (heterotrophs) that stabilize the soil surface. Biocrusts enhance nutrient cycling (nitrogen and carbon), improve soil water balance, and promote plant growth. They also impact atmospheric chemistry through the emission of nitric oxide (NO) and nitrous acid (HONO), influencing ozone production and atmospheric reactivity. Their strong cohesive structure, created through the secretion of extracellular polymeric substances and the entanglement of soil particles by rhizines and hyphae, significantly increases soil resistance to erosion, thus reducing dust emissions. Atmospheric dust, a key component of aerosols, is a significant factor in climate processes. Aeolian erosion in drylands is the primary source of dust, with emission rates dependent on surface conditions and meteorology. Dust particles, transported over long distances, scatter and absorb solar and long-wave radiation, impacting the atmosphere's optical properties. They act as cloud condensation and ice nuclei and undergo chemical alterations during atmospheric transport. Dust deposition delivers nutrients to land and ocean ecosystems, influencing productivity and biogeochemical cycles. However, it can also enhance snow and glacier melting, affect freshwater supplies, and negatively impact human health through the transport of pathogens, leading to respiratory and cardiovascular diseases. While small-scale studies have demonstrated biocrusts' effectiveness in reducing dust emissions, the global impact remained unknown. This study aims to bridge this gap by combining limited existing experimental data with a comprehensive global climate model to quantify the impact of biocrusts on the global dust cycle under present and future scenarios, highlighting the crucial role of these often overlooked organisms in regulating atmospheric dust and its climate effects.

The study draws upon existing research highlighting the multifaceted roles of biocrusts in dryland ecosystems, encompassing their impact on nutrient cycling (Elbert et al., 2012; Weber et al., 2016; Maier et al., 2018; Havrilla et al., 2019), water dynamics (Eldridge et al., 2020), and atmospheric chemistry (Meusel et al., 2018; Weber et al., 2015; Andreae & Crutzen, 1997; Olsson & Benner, 1999; Rossi & De Philippis, 2015). Previous work has also demonstrated biocrusts' role in reducing soil erosion and dust emissions (Pointing & Belnap, 2014; Belnap et al., 2007; Zhang et al., 2006; Zhang et al., 2008). The impact of dust on climate is reviewed, emphasizing dust's radiative forcing (Kok et al., 2018), its effect on ice nucleation (Morris et al., 2014; DeMott et al., 2010), and its interaction with pollution (Klingmüller et al., 2019). Existing literature on dust's biogeochemical impacts (Field et al., 2010; Okin et al., 2004; Mahowald et al., 2008; Pabortsava et al., 2017) and its influence on human health (De Longueville et al., 2010) are also considered. The literature also provided the basis for the model parameterization in relation to the effect of biocrusts on threshold friction velocity (TFV).

This study integrated experimental data on biocrust effects on threshold friction velocities (TFVs) – the minimum wind speed required to dislodge soil particles – with a global climate model (ECHAM6-HAM2.1). The experimental data, including biocrust cover, TFV of biocrust-covered and bare soils, and other relevant factors (soil texture, precipitation), were compiled from literature (Supplementary Table 1). An exponential relationship, previously established (equation 1: TFV increase ratio = *e^ax*^{biocrust cover}), was used to describe the relationship between biocrust coverage and TFV increase ratio (Extended Data Fig. 1). Three models—minimum, mean, and maximum effects—were parameterized based on the experimental data to account for uncertainty (Extended Data Fig. 1b,c). Global biocrust coverage data (Extended Data Fig. 2), obtained from a previous study using a maximum entropy method, was integrated with the TFV models into the ECHAM6-HAM2.1 climate aerosol model. This created a modified model version (ECHAM6-HAM2-BIOCRUST) capable of explicitly simulating the spatial variability of biocrust effects on dust emissions (Extended Data Fig. 3). Five 30-year simulations (1990-2020) were performed: one using the standard ECHAM6-HAM2.1 (implicitly including biocrust effects), three using ECHAM6-HAM2-BIOCRUST with minimum, mean, and maximum TFV models, and one using ECHAM6-HAM2-BIOCRUST without biocrust effects (ECHAM6-HAM2-NO BIOCRUST). The standard model applied a global TFV correction factor (0.9) to fit model outputs to observed data. The modified model employed a factor of 0.80 after sensitivity testing. The effect of biocrusts on dust emission, deposition, atmospheric burden, aerosol optical depth (AOD), and radiative forcing was assessed by comparing model outputs from ECHAM6-HAM2-BIOCRUST and ECHAM6-HAM2-NO BIOCRUST, using a one-way ANOVA and Tukey's post-hoc test. Future biocrust cover loss scenarios, based on IPCC climate and land-use projections (RCPs 2.6, 4.5, and 8.5), were also incorporated into the model (Extended Data Figs. 7, 8). The effects of future climate, vegetation change, and biocrust loss on dust cycling were assessed by comparing simulations with projected biocrust coverage against simulations with current coverage. The difference represent the impact of biocrust loss, isolating it from the effects of climate change and vegetation modification.

The study's key findings demonstrate the substantial influence of biocrusts on global dust cycling and radiative forcing. Model simulations show that biocrusts currently reduce global atmospheric dust emissions by approximately 60%, preventing the release of ~0.7 Pg of dust per year. The removal of biocrusts would result in an additional mean emission of ~700 Tg yr⁻¹ of dust (Fig. 2a). This effect is particularly pronounced in regions known for significant dust source areas, such as the west coast of North Africa, the Middle East, Central and South Asia, Australia, the western United States, and South America. The increased dust emissions would also lead to a substantial increase in global dust deposition, with continental deposition increasing by ~450 Tg yr⁻¹ and marine deposition by ~250 Tg yr⁻¹ (Fig. 2b, Extended Data Fig. 4, Supplementary Tables 4, 5). The reduction in dust emissions due to biocrusts leads to a significant decrease in the mean global atmospheric dust burden by ~8.5 Tg, corresponding to a 55% decrease (Fig. 2c, Supplementary Table 6). The global aerosol optical depth (AOD) also decreases by ~20% (Extended Data Fig. 5). This impacts radiative forcing; the overall net aerosol radiative effect at the top of the atmosphere (-3.5 W m⁻²) would decrease by up to 0.48 W m⁻² upon the removal of biocrusts (Fig. 2d, Extended Data Fig. 6, Supplementary Table 6). This is comparable to the total direct forcing of anthropogenic aerosols and three times larger than the impact of past land-use change on dust emissions. The impact of dust on nutrient cycling is substantial, particularly in regions where biocrusts are abundant. Removing biocrusts would increase phosphorus input to the Amazon rainforest by 100% of its current value. Future biocrust loss due to climate change and land-use intensification is projected to lead to a further increase in dust emissions and deposition (Fig. 3, Extended Data Figs. 7, 8). Depending on the scenario (RCPs 2.6, 4.5, and 8.5), dust emissions and deposition are predicted to increase by ~5% to 15%, while the dust burden increases by up to 16%, reducing the global radiation budget by 0.12 to 0.22 W m⁻² (Fig. 3d). The study highlights that this biocrust-induced increase in dust cycling would offset the generally expected decrease in dust production due to CO2 fertilization and sahel greening (Extended Data Fig. 9).

This study's findings highlight the critical, yet often overlooked, role of biocrusts in regulating the global dust cycle and its associated climate impacts. The significant reduction in dust emissions attributable to biocrusts (~60%) underscores their substantial influence on atmospheric dust load, radiative forcing, and biogeochemical cycles. The predicted future loss of biocrust coverage, driven by climate change and land-use intensification, will exacerbate dust production, potentially counteracting the positive effects of CO2 fertilization on dust reduction, leading to further climate alterations. The study's results have significant implications for regional and global climate modeling, requiring the integration of biocrust effects for improved accuracy and predictive capabilities. The impact on nutrient cycling is noteworthy, particularly in regions like the Amazon rainforest, where dust deposition plays a critical role in replenishing depleted phosphorus reserves. The increase in atmospheric dust burden could also affect human health. The quantifiable impact of biocrust loss on the global radiation budget emphasizes the importance of considering biocrusts in climate change mitigation and management strategies. Future research should focus on improving the accuracy of biocrust cover projections and enhancing the spatial resolution of dust emission models to better capture the localized effects of biocrusts. The study's approach could serve as a foundation for future research investigating the influence of biocrusts on other aspects of dryland ecosystem functioning and regional climate patterns.

This study demonstrates the significant and previously underestimated role of biocrusts in controlling global dust cycling and climate. Biocrusts effectively reduce dust emissions, leading to a substantial decrease in atmospheric dust load and radiative forcing. However, projected biocrust loss due to climate change and land-use change will reverse these benefits, leading to increased dust emissions and a negative impact on the global radiation budget. This study highlights the urgent need to incorporate biocrust dynamics into Earth system models and to consider their conservation in climate change mitigation and management strategies. Further research should focus on refining biocrust cover predictions and improving model resolution to accurately capture their diverse roles in dryland ecosystems and their influence on climate.

The study acknowledges limitations arising from the availability of experimental data on biocrust effects on TFVs. The data used were rather conservative, and more data are needed for a more precise parameterization of the biocrust effects. The model's spatial resolution (~210 km) might not fully capture small-scale variations in biocrust distribution and their impact on dust emissions. Furthermore, the models used rely on existing climate projections and biocrust cover predictions, which introduce uncertainties in the future scenarios. The assumed dust particle size distribution in the model may also introduce uncertainties, particularly concerning the radiative effects of dust. The study has focused on dust alone and hasn't incorporated the potential interaction between biocrust and other aerosol types.