. 2016 Nov 30;11(11):e0167265. doi: 10.1371/journal.pone.0167265

Table 2. Indicator species with associated spatial interaction type of the three types of ecosystem engineers and of bare soil, identified by indicator species analysis performed on the “relevés x species individuals with associated SIT (number)” matrix, with the observed indicator value and significance level.

Group with maximum IV	Species	SIT^a	IV	P^b
Cushion	Nototriche turritella	4	0.625	^***
	Calamagrostis sp.	4	0.600	^***
	Belloa kunthiana	4	0.558	^***
	Calamagrostis breviaristata	4	0.553	^***
	Poa aequigluma	4	0.533	^***
	*Parastrephia lucida*	4	0.532	^***
	Werneria aretioides	4	0.374	^***
	Luzula racemosa	4	0.331	^***
	Calamagrostis breviaristata	2	0.270	^**
	*Festuca orthophylla*	2	0.260	^**
	*Festuca orthophylla*	4	0.257	^***
Shrub	Belloa longifolia	4	0.343	^***
	Nototriche turritella	3	0.321	^***
	Perezia ciliosa	4	0.278	^***
	*Calamagrostis rigida*	4	0.267	^**
	*Calamagrostis rigida*	3	0.229	^***
	*Calamagrostis rigida*	2	0.228	^**
	Aetheolena campanulata	3	0.221	^**
	*Festuca orthophylla*	3	0.212	^**
Tussock	Belloa longifolia	3	0.769	^***
	Perezia ciliosa	3	0.520	^***
	Perezia sp.	3	0.367	^***
	Belloa kunthiana	3	0.363	^***
	Silene andicola	3	0.218	^**
	Gnaphalium badium	3	0.215	^***
	Hypochaeris echegarayi	3	0.212	^**
	Werneria aretioides	2	0.201	^*
Bare soil	Nototriche turritella	1	0.933	^***
	Calamagrostis breviaristata	1	0.500	^***
	Werneria aretioides	1	0.400	^***
	Nototriche pedicularifolia	1	0.367	^***
	Poa aequigluma	1	0.367	^***
	Viola granulosa	1	0.300	^***

IV, observed indicator value; SIT, spatial interaction type.

^a SIT1, individual growing outside the nurse canopy at a distance greater than 20 cm from the border of the nearest nurse patch; SIT2, individual growing less than 20 cm from the border of the nearest nurse patch but not under its canopy; SIT3, individual growing in the shadow of the nurse canopy; SIT4, individual growing inside the nurse canopy.

^b P is the probability of type I error, namely the proportion of times that the maximum IV_i from the randomized data set (4,999 iterations) equals or exceeds the maximum IV_i from the actual data set, under the null hypothesis that the maximum IV_i is no larger than would be expected by chance

(*P < 0.05

**P < 0.01

***P < 0.001).

Only significant indicator values (P < 0.05) higher than 0.20 are shown.

Potential nurse species are in bold.