Caroline A. Havrilla

Soil degradation is one of the greatest environmental issues our planet faces today, with over 33% of Earth's soils currently degraded. Drylands are especially vulnerable to soil degradation given their history of intensive land use and desertification. Active soil restoration has been identified as a leading strategy to combat soil degradation and promote ecosystem recovery. However, soil‐based dryland restoration techniques have shown varying success, potentially due to a lack of understanding of the ecological contexts in which soil‐based treatments are most beneficial. To improve our understanding of how to best use active soil restoration to restore degraded drylands, we conducted a global meta‐analysis of soil treatment effectiveness at improving soil health across varying environmental gradients. The soil health metrics we analysed were aggregate stability, bulk density, soil moisture, soil organic carbon (SOC), soil nitrogen, mycorrhizal colonization and basal respiration. For this meta‐analysis we collected 155 publications, yielding 1403 unique studies spanning six continents. We found that overall, soil restoration had a beneficial effect on all measures of soil health ranging from a +11% increase in bulk density (inverse) to a +6967% increase in mycorrhizal colonization. Aridity and soil texture greatly influenced restoration effectiveness for certain soil health metrics. Specifically, for soil carbon and nitrogen, restoration was found to be most effective in arid, fine‐textured soils and mesic, coarse‐textured soils. Additionally, we found that organic amendments were most effective at increasing SOC, while fungi inoculation was most effective at increasing mycorrhizal colonization. Synthesis and applications : Our findings indicate that active soil restoration is an effective tool for increasing soil health and provide information on optimal treatments and site conditions for improving certain aspects of soil health. This could greatly help inform decision‐making, and thus improve outcomes, in dryland restoration worldwide.

Restoration in dryland ecosystems often has poor success due to low and variable water availability, degraded soil conditions, and slow plant community recovery rates. Restoration treatments can mitigate these constraints but, because treatments and subsequent monitoring are typically limited in space and time, our understanding of their applicability across broader environmental gradients remains limited. To address this limitation, we implemented and monitored a standardized set of seeding and soil surface treatments (pits, mulch, and ConMod artificial nurse plants) designed to enhance soil moisture and seedling establishment across RestoreNet, a growing network of 21 diverse dryland restoration sites in the southwestern USA over 3 years. Generally, we found that the timing of precipitation relative to seeding and the use of soil surface treatments were more important in determining seeded species emergence, survival, and growth than site‐specific characteristics. Using soil surface treatments in tandem with seeding promoted up to 3× greater seedling emergence densities compared with seeding alone. The positive effect of soil surface treatments became more prominent with increased cumulative precipitation since seeding. The seed mix type with species currently found within or near a site and adapted to the historical climate promoted greater seedling emergence densities compared with the seed mix type with species from warmer, drier conditions expected to perform well under climate change. Seed mix and soil surface treatments had a diminishing effect as plants developed beyond the first season of establishment. However, we found strong effects of the initial period seeded and of the precipitation leading up to each monitoring date on seedling survival over time, especially for annual and perennial forbs. The presence of exotic species exerted a negative influence on seedling survival and growth, but not initial emergence. Our findings suggest that seeded species recruitment across drylands can generally be promoted, regardless of location, by (1) incorporation of soil surface treatments, (2) employment of near‐term seasonal climate forecasts, (3) suppression of exotic species, and (4) seeding at multiple times. Taken together, these results point to a multifaceted approach to ameliorate harsh environmental conditions for improved seeding success in drylands, both now and under expected aridification.