Skip to main content

Table 3 ANCOVA results: ANCOVA table testing the effect of the ratios between metals and DOC (independent covariate) on the log-transformed ratios between toxicity indices estimated by BLM and bioassays (dependent variable) for two species D. magna and P. subcapitata (nominal factors)

From: Predictions of Cu toxicity in three aquatic species using bioavailability tools in four Swedish soft freshwaters

 

Al/TOC

Fe/TOC

∑FeAl/TOC (mol/g)

∑Metals/TOC (mol/g)

R 2

0.739

0.945

0.980

0.985

Root mean square error

0.590

0.270

0.163

0.142

Analysis of variance

 F 3.7 ratio

3.77

23.04

65.59

87.37

 Prob > F

0.1164

0.0055

0.0007

0.0004

 Effect tests:

  Species

0.03

0.002

0.0003

0.0002

  log(Me/TOC)

0.5493

0.0144

0.0018

0.0010

  Species x log(Me/TOC)

0.93

0.79

0.80

0.72

Estimated slope of covariate ± SE

 log(Me/TOC)

−0.069 ± 0.11

−0.046 ± 0.01

−2.531 ± 0.34

−2.655 ± 0.31

  1. See Fig. 3 for further details