Few direct responses to water availability
Our results did not support the expectation that water availability plays a direct role in survival, seed production, and population growth rates of herbaceous plants (Mu et al. 2021). We found the impact of water availability depended on other abiotic conditions and was more frequently important in models of survival than seed production or population growth rate. For three species, the wet treatment appeared to buffer the negative effects of increasing light and decreasing nutrients on survival, whilst the fourth species only experienced reduced survival along PC1 in the dry treatment. This result supports existing knowledge that interactions among shade, soil nutrients, and moisture influence plant survival, particularly in arid environments (Valladares 2003). In contrast, water availability rarely modified the effect of neighbours on vital or population growth rates. In perennial grasses, Adler et al. (2009) similarly reported little evidence of interactive effects between precipitation and the presence of neighbours, however unlike our results they found strong support for direct effects of precipitation.
One reason why the watering treatment had few direct impacts on vital and population growth rates, and indirect impacts via interactions with competition, may simply be that the watering treatment did not significantly alter soil moisture compared to ambient conditions. In the year we conducted the study, conditions were drier than average at 110 mm of rainfall over the growing season and as such our watering treatments represented 30-90% of average rainfall. Based on our study and results from other watering trials in this system (Wainwrightet al. 2018; Towers et al. 2020), it seems likely that in many years water is not limiting performance during the growing season.
The timing of rainfall events is an important driver of variation in vital rates, plant-plant interactions, and population dynamics (Levineet al. 2008; Compagnoni et al. 2016; Conquet et al.2023). Even in the single year studied here, unseasonably heavy summer rainfall cued emergence outside the growing season for the two exotic species almost exclusively. In the future, it would be interesting to implement the watering treatment before the onset of winter rainfall to assess emergence responses, which can have large contributions to population growth rates (James et al. 2020). It would also be valuable to track soil moisture in each plot after rainfall events to measure the magnitude and duration of the effect of the watering treatment on soil conditions. Since annual plants can employ different ecological strategies to buffer performance over time in water-limited environments, the effect of interannual variation in water availability on vital rates is another important avenue of research (Angert et al. 2007).