Phytoplankton taxonomy was undertaken using DNA metabarcoding (along with traditional taxonomic approaches via microscopy). The filter samples were extracted for DNA using E.Z.N.A. water DNA kits using the default protocol (Omega Bio-Tek, USA). After extraction, the samples were PCR amplified using the 18s rRNA V9 region with unique dual-tags using the forward primer 1380F (5′-CCCTGCCHTTTGTACACAC-3′ ) and reverse primer 1510R (5′-CCTTCYGCAGGTTCACCTAC-3′) (Amaral-Zettler et al., 2009; Xie et al., 2018). PCR products were checked with agarose gel electrophoresis and purified using the QIAquick PCR Purification kit (Qiagen, Germany). The sequencing libraries were constructed as described elsewhere (DeBofsky et al., 2020) and sequenced using a Illumina 150PE chemistry on NextSeq® 500 sequencer, including blanks of field sampling, extraction, and PCR amplification. Bioinformatics was conducted under QIIME2 environment (version 2023.5). Raw reads were demultiplexed based on dual-tags of forward and reverse primers for each sample. Sequences with low quality, chimeras, and lengths shorter than 100 bp were discarded. Zero-radius operational taxonomic units (ZOTUs) were generated using unoise3, with a minimum frequency of 5 (Edgar, 2016). Taxonomy of ZOTUs were annotated by use of classify-sklearn approach against the Protist Ribosomal Reference (PR2) database (Bolyen et al., 2019; Quast et al., 2013). ZOTUs assigned to superphyla of Hacrobia, Stameonopiles, Alveolata, Rhizaria and Archaeplastida were kept as phytoplankton assemblage for further statistics. Samples were rarefied to equal depths of 146,218, after merging the three technical replicates for each enclosure at each sampling time point.The files included here are the R file (Phyto_Forest_Publication.rmd, along with the datasets) used to process the collapsed ZOTU file (Zotus_enclosure_clean.qza), import the annotated features using the PR2 database (Zotus-PR2.qza), and process the rarefied data output (rtable-rarefied-collapsed.qza). The metadata used for data analysis is additionally included (metadata-clean.tsv)