Phenological Timing & Abundance
To determine the differences in reproductive timing for each species, we totaled which treatments first reached reproductive milestones (first flowering, first peak flowering, first senescence). We found evidence that heat treatment and heat plus precipitation treatment generally experienced an earlier phenological cycle than drought treatment and control treatment (fig. 1). That is, within heated treatments, species frequently flowered, reached peak flowering, and senesced before doing so in control treatment and drought treatment. Based on statistical analysis, there was a significant trend for three species (Plectritis congesta, Cryptantha intermedia, and Clarkia purpurea). Species in drought treatment consistently reached peak flowering later than in other treatments, but senesced with or even before those in other treatments.
Drought Effects on Plant Phenology
We also expected to see a similar peak flowering time in the drought treatment and the control treatment, but a lower reproductive output in the drought treatment than in the control treatment (Hypothesis 2). We found a trend in three species (Cryptantha intermedia, Plectritis congesta, Sidalcea malviflora) that support this hypothesis (fig. S1). These three species experienced peak flowering at approximately the same time in the drought treatment and control treatment with a higher abundance of flowers in the control treatment. However, the relationship between drought treatment and control treatment are not statistically significant based on our analysis (ANOVA and Tukey’s post hoc test). We did find that the drought treatment more frequently corresponded with the control treatment while the heated treatments had a lower abundance of flowers at their peak (fig. 2). The same three species (Cryptantha intermedia, Plectritis congesta and Sidalcea malviflora) displayed this trend with statistical significance over multiple weeks. This suggests that temperature has a greater effect on reproductive timing and abundance of reproductive parts than precipitation. We suspect that this could be a result of the especially wet winter we had this growing season. A 40% reduction in precipitation might not have had such drastic effects on the drought treatment plots because they still received a sufficient amount of water. Overall, species reacted differently under each treatment, which resulted in a variation within the analysis.
Normalized Difference Vegetation Index
To test our third hypothesis, an ANOVA was performed on the week of greatest difference between the average NDVI value for each climate treatment. After finding a significant difference on Julian Day 111 (April 21), a Tukey’s HSD post hoc test was conducted to compare between treatments. We found significant difference in NDVI between the heat plus precipitation treatment and control treatment (p<0.01) and between the drought treatment and heat plus precipitation treatment (p<0.05). Further, both the heat treatment and heat plus precipitation treatment experienced an earlier peak in NDVI than the drought or control treatments, and the drought treatment and control treatment experienced a decline in NDVI during the week that the heat treatment and heat plus precipitation treatment experienced a peak (fig. 3). Therefore a significantly higher biomass production on Julian Day 111 and an earlier peak of biomass production in the heat treatment and heat plus precipitation treatment support our hypothesis. However, it is uncertain whether the heat treatment has an overall higher biomass production than the control treatment or whether the phenological cycle is just shifted to an earlier time of the year for the heat treatment.
