Table 1.

Summary statistics of GAMM models relating DNA concentration (ng/μl) and DNA integrity given by Tapestation DNA Integrity Number (DIN) of DNA extracted from 96% ethanol used to preserve five unprocessed freshwater macroinvertebrate bulk samples, in relation to ethanol subsampling day and DNA extraction method

Parametric coefficients	DNA concentration (ng/μl)				DIN value
	Estimate	SE	t‐value	p	Estimate	SE	t‐value	p
Intersect	3.854	0.617	6.249	4.5 × 10−09 ***	4.820	0.446	10.803	<2 × 10−16 ***
TISSUE	−1.385	0.814	−1.702	0.091ns	−1.136	0.472	−2.404	0.018*
SOIL	−2.230	0.828	−2.695	0.008**	−3.063	0.508	−6.031	1.3 × 10−08 ***
Smooth terms	edf		F‐value	p	edf		F‐value	p
s(day): BEAD	1.695		2.464	0.200ns	1.000		1.493	0.224ns
s(day): TISSUE	1.000		0.004	0.947ns	1.720		14.385	1.6 × 10−04 ***
s(day): SOIL	1.000		0.325	0.569ns	1.246		0.137	0.829ns

Subsampling was conducted at 1, 2, 3, 5, 7 and 14 days after field sampling, and DNA extractions were performed using three methods: BEAD, TISSUE and SOIL as described in Table S1. For each model, we provide the parameter estimates, standard errors (SE) and statistical significance of parametric terms, and the effective degrees of freedom (edf) and approximate significance of smooth terms. The shape of the smooth terms is provided in Figure 1.

^*** p < 0.001.

^** p < 0.01.

^* p < 0.05.

^ns p > 0.05.