Table 2. Results of linear mixed-effects models of the infiltration capacity for June, September and October.

	June					September				October
Source	df	0 m		−0.02 m		0 m		−0.02 m		0 m		−0.02 m
		L-ratio	p	L-ratio	p	L-ratio	p	L-ratio	p	L-ratio	p		L-ratio	p
BL	3	4.05	0.256	4.02	0.259	10.77	0.013	11.76	0.008	2.89	0.409		2.92	0.403
SR (log-linear)	1	0.28	0.596	0.12	0.724	0.04	0.834	0.06	0.800	0.76	0.385		0.50	0.480
FG	1	0.70	0.403	2.43	0.119	0.03	0.858	<0.01	0.991	0.05	0.825		0.03	0.870
GR	1	0.31	0.994	0.65	1.00	4.27	0.116↓	4.62	0.096↓	6.39	0.045↓		7.59	0.018↓
LEG	1	0.72	0.639	0.76	1.00	7.60	0.023↑	8.96	0.012↑	10.95	<0.001↑		10.53	0.004↑
SH	1	1.98	0.934	0.70	1.00	1.46	0.454	1.36	0.488	2.01	0.314		1.70	0.386
TH	1	1.02	0.934	0.72	1.00	0.24	0.627	0.10	0.758	0.33	0.563		0.58	0.446
E	1	4.14	0.042↑	1.89	0.169	0.23	0.630	0.46	0.498	0.88	0.349		0.34	0.557
E×SR(log-linear)	1	1.33	0.248	0.19	0.664	0.21	0.645	0.14	0.706	4.69	0.030		5.50	0.019
E×FG	1	0.60	0.437	1.80	0.406	0.26	0.614	0.49	0.484	1.24	0.264		12.95	0.002

Infiltration capacity as affected by block (BL), plant species richness (SR), plant functional group richness (FG), grasses (GR), legumes (LEG), small herbs (SH), tall herbs (TH), and earthworm treatment (E) in June, September and October separately for the matric potentials ψ _M = 0 m and ψ _M = −0.02 m.

Models were fitted by stepwise inclusion of fixed effects. Likelihood ratio tests were applied to assess model improvement (L-ratio) and the statistical significance of the explanatory terms (p values). Models always included 84 observations (n) on 42 plots, df = degrees of freedom (df) for each of the predictor variables. For GR, LEG, SH and TH the adjusted p-values according to the Holm procedure are given in the table. Significant effects are marked in bold. Arrows indicate increase (↑) or decrease (↓).