Results
We selected and analyzed 44 published papers in total, from which we extracted 100 R2 values ranging from 0.02 to 0.99 (Tables 1 and S1). All the collected studies targeted mitochondrial DNA fragments, and no study reported the positive correlation between nuclear eDNA concentration and species abundance. Most studies reported the correlation between eDNA concentration and species abundance, targeting fish species (30 of 44 studies). The filter pore size and amplicon size ranged from 0 to 10 μm and 66 to 719 bp, respectively. The majority of studies used filters with less than 3 µm pore size and amplified less than 200 bp target DNA fragments. The number of studies conducted in laboratory, lentic freshwater, lotic freshwater, and marine environments were 15, 14, 14, and 4, respectively.
The effect sizes of R2 values and their variances were estimated using Fisher’s z‐transformation and forest plots (Figure S1). Relative to fish (R2 = 0.68 [95 % CI: 0.61, 0.74]), estimated R2 values were significantly lower for crustaceans (0.22 [−0.04, 0.46]) and mussels (0.50 [0.44, 0.55]); among the target taxa, the correlation between crustacean eDNA concentration and abundance was the weakest (Figure 1). Laboratory experiments (0.78 [0.69, 0.84]) produced higher estimated R2 values than natural environments, and there was no substantial difference in the estimation accuracy among natural environment types (Figure 2). In addition, the LMM showed a significant positive effect of filter pore size on R2 values (P < 0.05; Table 2 & Figure 3). R2values tended to be higher as filter pore size increased, but some datasets also reported high R2 values with smaller filter pore sizes. PCR amplicon size did not significantly influence R2 values (P > 0.1; Table 2 & Figure S2).