Ecosystems face multiple, potentially interacting, anthropogenic pressures that can modify biodiversity and ecosystem functioning. Using a bryophyte-microarthropod microecosystem we tested the combined effects of habitat loss, episodic heat-shocks and an introduced non-native apex predator on ecosystem function (chlorophyll fluorescence as an indicator of photosystem II function) and microarthropod communities (abundance and body size). Photosynthetic function was degraded by the sequence of heat-shock episodes, but unaffected by microecosystem patch size or top-down pressure from the introduced predator. In small microecosystem patches without the non-native predator, Acari abundance decreased with heat-shock frequency, while Collembola abundance increased. These trends disappeared in larger microecosystem patches or when predators were introduced, although Acari abundance was lower in large patches that underwent heat shocks and were exposed to the predator. Mean assemblage body length (Collembola) was reduced independently in small microecosystem patches and with greater heat-shock frequency. Our experimental simulation of episodic heat waves, habitat loss and non-native predation pressure in microecosystems produced evidence of individual and potentially synergistic and antagonistic effects on ecosystem function and microarthropod communities. Such complex outcomes of interactions between multiple stressors need to be considered when assessing anthropogenic risks for biota and ecosystem functioning.

96 experimental replicates comprising a bryophyte microecosystem supporting Acari and Collembola communities [40, 41] were randomly excised (21 June 2017) using domestic steel circular cookie cutters (110 mm or 50 mm diameter giving microcosms of 95 cm² and 20 cm², respectively) from a large, continuous bryophyte sward (Mnium hornum Hedw. + rare occurrences of Polytrichastrum formosum (Hedw.) G.L. Smith; Hypnum andoi A.J.E. Smith) on a brown earth soil in a woodland (Bush Estate, Scotland, U.K. 55.861111, -3.205833). The microecosystem experiment was carried out at CEH Edinburgh (55.861111, -3.205833) over 8 weeks (21/06/2017-11/08/2017). We employed a randomised factorial blocked design with three treatments: 1) microecosystem size (95 cm², n = 48; 20 cm², n = 48); 2) frequency of heat-shock episodes (‘un-stressed’ controls: n = 32; 2 episodes: n = 32; 3 episodes: n = 32); and 3) addition of an apex predator (Dalotia coriaria -Coleptera present n= 48 or control, n = 48). We challenged the microecosystem replicates with experimental episodic heat shocks (see below) to simulate short duration heat waves. Replicates were randomly assigned to acute heat-shock episodes of two hours duration on two or three occasions during the experiment or to a control (no heat-shock applied). Experimental heat-shock treatment was applied by temporarily and carefully exposing replicate microecosystems for two hours beneath 40W light bulbs set within racks of 54 Tullgren extraction funnels (Burkard Scientific Ltd). The 96 microecosystems randomly to eight blocks (large plastic trays 100 cm x 50 cm x 16 cm) ensuring each contained a full replicate of the treatment combinations. Following the 24hr destructive harvest at the end of the experiment, we sorted and counted all invertebrates collected to the level of taxonomic subclass (Acari, Collembola) to provide an estimate of community size (total abundance and density per cm²).We also measured the body length (frons to end of abdomen) of sub-samples of Collembola individuals extracted from each large or small microecosystem following application of each heat-shock (t1/t2/t3) treatment. Moss community chlorophyll fluorescence (Fv/Fm) , as an indicator of photosynthetic capacity (photosystem II function) was measured using a Continuous Excitation Chlorophyll Fluorimeter (HandyPEA, Hansatech Instruments Ltd, UK) on randomly selected bryophyte leaves from each microecosystem.

Data relates to the open access paper in Proceedings of the Royal Society B (under revision).