Dataset selection
Using Google Scholar and Web of Science (a list of keywords is available as Supplementary Materials), we collected studies from each system where an experimental disturbance was imposed and 16S rRNA gene amplicon sequencing datasets were available. Specifically, we chose studies, which : 1) were sequenced in Illumina or IonTorrent platforms; 2) sequenced the v3-v4 regions of the 16S rRNA gene; 3) were published after 2014; 4) repeatedly sampled microbial communities following a discrete disturbance or environmental change; 5) included samples from before the disturbance (i.e., controls), at least one sampling within a week after disturbance, and at least one sampling within a month after disturbance; and 6) included experimental triplicates (i.e., three samples per time point). Criteria 1-3 ensured that the sequencing techniques were comparable between studies, and avoided the biases associated with sampling different regions of the 16S rRNA gene (Yanget al. 2016). Whether shifts in community composition occur at similar rates across environments is seldom compared using empirical data. Criteria 4-6 ensured that the time scales were comparable between time series, that the effects of environmental change could be comparedwithin time series between different time points, and that the variability of the microbiomes at each time point could be measured. We defined a pulse disturbance as a “discrete, short-term event” (Shadeet al. 2012a). We excluded datasets for which raw sequencing data was not publicly available, and stopped data collection in (Hoet al. 2017) October 2020. In all, 29 datasets matched our criteria (Davidet al. 2015; Seekatz et al. 2015; Datta et al.2016; Džunková et al. 2016; Fuentes et al. 2016; Vaquer-Sunyer et al. 2016; Venkataraman et al. 2016; Wuet al. 2016; Dong et al. 2017; Jurburg et al. 2017, 2018, 2019; Qian et al. 2017; de Vries et al. 2018; Flancman et al. 2018; Frenk et al. 2018; Kennedy et al. 2018; Li et al. 2018, 2019; Mateos et al. 2018; van Kruistum et al. 2018; Lavelle et al. 2019; Lu et al. 2019; Santi et al. 2019; Ward et al. 2019; Yanet al. 2020) Table S1). We grouped these time series into three environmental categories: aquatic, mammal-associated, and soil microbiomes (including rhizosphere microbiomes).