Errata for Tenhumberg, B., E. E. Crone, S. Ramula, and A. J Tyre. 2018. Time-lagged effects of weather on plant demography: drought and Astragalus scaphoides. Ecology 99:915–925. https://doi.org/10.1002/ecy.2163

In Tenhumberg et al. the authors made a mistake when calculating the drought index SPEI from weather data using the software R. The authors applied the function spei () to stacked weather data from four different sites, but erroneously omitted to group the data by site before making the spei-calculations. As a result of the mistake there were the following small changes to the results (see corrected Fig. 3): The relationship between SPEI and flowering probability of flowering A. scaphoides plants was no longer significant, but mean temperature was. Additionally, the flowering probability of small plants was no longer influenced by weather variables, but in this data set small plants rarely flower (on average <10%, Appendix S1). Also, in the corrected analysis the signal of SPEI on recruitment and survival of small plants is longer (time lag of two years instead of one year). However, the integrated effect of all vital rates on life history of A. scaphoides assessed by the asymptotic growth rate, log(λ_t) was not affected by the mistake. Hence, the mistake did not affect the core finding that log(λ_t) was highest if dry conditions in year t – 1 were followed by wet conditions in the year t, and that the positive effects of wet years exceeded their negative effects, suggesting that increasing frequency of drought conditions would reduce population viability of A. scaphoides.

The mistake in the R-code on Dryad was discovered by Dr. Eric Scott (University of Florida).

Corrected Fig. 3: Effect of past (a, c–f, h) SPEI or (b, g) mean temperature (M. Temp) on vital rates and population growth rate, log(λ_t), of A. scaphoides during the demographic transitions between July in year t – 1 and t. The x-axes show time in the past at a resolution of months. The gray and white stripes indicate winter/spring and summer/fall periods, respectively. The horizontal line indicates no effect, and the solid line indicates the average effect. There is a significant effect of SPEI or mean temperature during those time periods where the 95% confidence intervals (dashed lines) do not overlap with the horizontal line. V-sm (small vegetative plants with one to five leaves), V-lg (large vegetative plants with six or more leaves), and F (flowering plants).

Tenhumberg, B.、EE Crone、S. Ramula 和 A. J Tyre 的勘误表。2018. 天气对植物人口学的滞后影响：干旱和黄芪。生态学99：915-925。https://doi.org/10.1002/ecy.2163

在 Tenhumberg 等人。作者在使用 R 软件根据天气数据计算干旱指数 SPEI 时犯了一个错误。作者将函数 spei() 应用到来自四个不同站点的堆叠天气数据，但在进行 spei 之前错误地省略了按站点对数据进行分组的情况。计算。由于错误，结果有以下小的变化（参见更正的图 3）： SPEI 与开花A 的开花概率之间的关系。舟状体植物不再重要，但平均温度是。此外，小植物的开花概率不再受天气变量的影响，但在这个数据集中，小植物很少开花（平均 <10%，附录 S1）。此外，在校正分析中，SPEI 对小植物补充和存活的信号更长（时间滞后两年而不是一年）。然而，所有生命率对A生活史的综合影响。由渐近增长率评估的舟状体，log（λ _t）不受错误影响。因此，如果t – 1年的干燥条件随后是该年的潮湿条件，则该错误不会影响 log(λ _t ) 最高的核心发现t，并且潮湿年份的积极影响超过了它们的消极影响，这表明干旱条件频率的增加会降低A 的种群生存能力。舟状舟。

Dryad 上的 R 代码中的错误是由 Eric Sc​​ott 博士（佛罗里达大学）发现的。

校正图3：过去的效力（A，C-F，H）或SPEI（B，G）平均温度生命率和人口增长速率（M.温度），日志（λ_吨），的甲。在t – 1年 7 月和t之间的人口转变期间，scaphoides。该X-axes 以月的分辨率显示过去的时间。灰色和白色条纹分别表示冬季/春季和夏季/秋季时期。横线表示没有效果，实线表示效果一般。在 95% 置信区间（虚线）与水平线不重叠的那些时间段内，SPEI 或平均温度有显着影响。V-sm（具有一到五片叶子的小型营养植物）、V-lg（具有六片或更多叶子的大型营养植物）和 F（开花植物）。