Fungal connections between plants and biocrusts facilitate plants but have little effect on biocrusts

Abstract

Species interactions may couple the resource dynamics of different primary producers and may enhance productivity by reducing loss from the system. In low-resource systems, this biotic control may be especially important for maintaining productivity. In drylands, the activities of vascular plants and biological soil crusts can be decoupled in space because biocrusts grow on the soil surface but plant roots are underground, and decoupled in time due to biocrusts activating with smaller precipitation events than plants. Soil fungi are hypothesized to functionally couple the plants and biocrusts by transporting nutrients. We studied whether disrupting fungi between biocrusts and plants reduces nitrogen transfer and retention and decreases primary production as predicted by the fungal loop hypothesis. Additionally, we compared varying precipitation regimes that can drive different timing and depth of biological activities. We used field mesocosms in which the potential for fungal connections between biocrusts and roots remained intact or were impeded by mesh. We imposed a precipitation regime of small, frequent or large, infrequent rain events. We used 15N to track fungal-mediated nitrogen (N) transfer. We quantified microbial carbon use efficiency and plant and biocrust production and N content. Fungal connections with biocrusts benefitted plant biomass and nutrient retention under favourable (large, infrequent) precipitation regimes but not under stressful (small, frequent) regimes, demonstrating context dependency in the fungal loop. Translocation of a 15N tracer from biocrusts to roots was marginally lower when fungal connections were impeded than intact. Under large, infrequent rains, when fungal connections were intact, the C:N of leaves converged towards the C:N of biocrusts, suggesting higher N retention in the plant, and plant above-ground biomass was greater relative to the fungal connections-impeded treatment. Carbon use efficiency in both biocrust and rooting zone soil was less C-limited when connections were intact than impeded, again only in the large, infrequent precipitation regime. Synthesis. Although we did not find evidence of a reciprocal transfer of C and N between plants and biocrusts, plant production was benefited by fungal connections with biocrusts under favourable conditions.