Water‐saving strategies in rice farming entail cascading effects in prey–predator interactions across ecosystem boundaries

Abstract Water‐saving irrigation strategies have been globally promoted to mitigate the contribution of flooded rice farming to climate change. While the positive effect of those strategies in reducing greenhouse gas emissions is undeniable, their potential cascading effects across the aquatic–terrestrial interface remain completely unexplored. For instance, multiple drainages throughout the rice cycle associated with alternative irrigation practices may disrupt the emergence of semiaquatic insects from rice fields, reducing prey availability for terrestrial predators and ultimately affecting their reproductive outcomes. Here, by using a 2‐year field‐scale experiment, we addressed these issues by comparing three irrigation strategies that represent a gradient of water use intensity throughout the rice growing season: Conventional permanent flooding (i.e. no drying periods) > mid‐season drainage (i.e. one single drying period; MSD) > alternate wetting and drying (i.e. multiple drying periods; AWD). Specifically, on each experimental plot, we quantified (i) the emergence of semiaquatic insects, (ii) the breeding activity (i.e. the breeding probability) of a jumping spider species ( Bianor albobimaculatus , Salticidae) and (iii) its reproductive fitness (i.e. eggs/sac). Our results show that the emergence of semiaquatic insects and, therefore the availability of preys for spiders, were markedly reduced as water use decreased. In addition, while the breeding activity of jumping spiders did not differ among irrigation strategies, their reproductive fitness was severely compromised in the alternate wetting and drying strategy. Synthesis and applications . These results show that introducing multiple drainage periods in rice fields (i.e. AWD) indirectly hampers terrestrial spider reproduction through limiting the emergence of potential preys from the aquatic to terrestrial boundaries. MSD resulted in a more conciliatory strategy as it largely reduces methane emissions and does not affect predator–prey interactions; thus, it should be prioritized over AWD to minimize environmental trade‐offs. Our results highlight the need to account for potential trophic cascading effects when designing climate change mitigation strategies in agriculture to avoid undesirable side‐effects on agroecosystem functioning.