The climate warming experienced during the 20th century and that predicted for the 21st century is largely attributed to warmer winters and nighttime temperatures, yet experiments often only manipulate daytime maximum temperatures and nighttime warming is frequently ignored. We compared the effects of increasing daytime temperature (Tmax) to increasing nighttime temperature (Tmin) on top-down control of plant diversity in Connecticut old fields. In this system, plant diversity is maintained by the consumption of competitively dominant goldenrod by grasshoppers. However, grasshopper foraging time is strongly tied the risk of spider predation. Grasshoppers feed more when spiders and grasshoppers are spatially separated, but behavioral thermoregulation in response to air temperature affects the degree to which these species overlap. We examined the effects of increasing Tmax and Tmin on grasshopper feeding time and plant diversity using 24-hour behavioral observations and a two-year field experiment. Enclosures were assigned to one of three treatments: no warming, increased Tmax, or increased Tmin. This experimental design standardized the magnitude of warming (+2-5°C for 12 hrs) and changed only when enclosures were warmed (Tmax: 06:00-18:00 hrs and Tmin: 18:00-06:00 hrs), thus allowing us evaluate the importance of timing in warming experiments.
Results/Conclusions
Increasing Tmax and Tmin had opposite affects on plant diversity that arose through a series of indirect interactions mediated by spiders and grasshoppers. Behavioral experiments revealed that temperature affected spider location in the plant canopy (but not grasshopper location), thereby altering the degree in which these species overlapped spatially. Increasing Tmax required spiders to seek cooler temperatures lower in the plant canopy during the day, while grasshoppers continued to feed higher in the canopy. Thus increasing Tmax reduced predation risk and allowed grasshoppers to feed more on goldenrod. As a result, after two years the plant communities in warmed Tmax treatments had less goldenrod and increased plant species evenness. While grasshoppers did feed at night in all treatments, increasing Tmin reduced nocturnal feeding because nighttime warming increased spider activity. Consequently, plant communities exposed to elevated Tmin showed reduced plant evenness after two years. These contrasting effects of daytime and nighttime warming on species interactions and plant community evenness emphasize the need to consider the timing of warming during climate change experiments.