Plant water stress can affect selectivity by insect herbivores. Numerous studies have shown greater insect preference for water-stressed plants, but others have reported the opposite response. We evaluated leaf consumption by adults of Nyctelia circumundata (a chewing insect) in leaves of Larrea divaricata and Prosopis alpataco. Three bioassays (two-way choice tests) were performed: two intra-specific comparisons between well-watered (+W) and water-stressed (−W) leaves of each species and one inter-specific comparison between leaves of the two species. Leaf biomass was reduced by water stress in both species. Nitrogen concentration in leaves (N) was reduced by drought in P. alpataco. In contrast, total phenolics and specific leaf area (SLA) did not differ between treatments within species. Nyctelia circumundata did not show preference by any water supply regimes in intra-specific comparisons. In contrast, in inter-specific choice tests, it showed a marked preference for P. alpataco, which is the species with the highest nitrogen concentration and lowest total phenolics concentration. In intra-specific comparisons, maximum leaf consumption was inversely related to SLA in both species. Furthermore, in P. alpataco, N concentration was positively related to maximum leaf consumption and negatively related to leaf water content (LWC). In contrast, in inter-specific comparisons, total phenolics was negatively related to maximum leaf consumption, while N concentration exhibited the opposite trend. These results suggest that food selection is a hierarchical process where chemical attributes (i.e., total phenolics and N) are taken into account for species selection, and physical attributes (i.e., SLA and LWC) for choosing individuals inside species.

Raingardens capture and filter urban stormwater using sandy soils and drought-tolerant plants. An emerging question is whether raingardens can also be used as vegetable gardens, potentially increasing their popularity and implementation. A successful vegetable raingarden will need to both retain stormwater and produce vegetables, despite potential water deficits between rainfall events. To determine whether raingardens can provide this dual functionality, we undertook a greenhouse pot experiment using two different substrates (loamy sand raingarden substrate and potting mix typical of containerised vegetable growing) and two methods of stormwater application (‘sub-surface’ and ‘surface’ watering) with the water quantity at each application determined by average Melbourne summer rainfall. Overall, potting mix produced bigger plants (biomass and leaf area) and greater yield than did the loamy sand. Yield effects were variable: tomato yield was unaffected by treatment, bean yield was greatest in potting mix, beetroot yield was greatest with sub-surface watering and parsley yield was greatest with surface watering. Bigger plants also had greater transpiration, which meant that stormwater retention was greatest for parsley and tomato plants growing in potting mix with surface watering. Although, a raingarden with potting mix and surface application of stormwater was optimal for producing food and retaining stormwater under our rainfall regime, potting mix could be problematic due to higher nutrient leaching and breakdown over time. Therefore, we recommend using a mix of loamy sand and potting mix. However, the choice of substrate and watering treatment require trade-offs between yield, stormwater retention and potential implications for water quality and long-term stability of hydraulic properties.