Hong Liu, Eric S. Menges, and Pedro F. Quintana-Ascencio. 2005. Population viability analyses of Chamaecrista keyensis: effects of fire season and frequency. Ecological Applications 15:210–221.

Appendix C. Identifying and pooling of stages of Chamaecrista keyensis.

FIG. C1. Scatterplot of probabilities of identifying seedlings (N = 311) and non-seedlings (N = 1062) of C. keyensis correctly based on plant size, assuming plants smaller than the cutoff size are seedlings, and plants larger than the cutoff size are non-seedlings.

Table C1. Summary of pooled stages. Tsf = time since last fire (years), Sv = small vegetative, Sf = small flowering plant, Lv = large vegetative, Lf = large flowering plant, Mf = medium flowering plants.

Plot	Year	Tsf	Burn season	Stage	Pre-pool N	Pooled with 1	pooled N
OC	98-99	10	Control	Sv	0 1	OC (99-00) 2	26
OS	98-99	0	Summer	Sv	0 1	OS (Seedling) 3	35
OW	98-99	0	Winter	Sv	0 1	OW (Seedling)3	76
PC	98-99	20	Control	Sv	0 1	PC (99-00) 2	17
PS	98-99	0	Summer	Sv	0 1	PS (Seedling) 3	36
PW	98-99	0	Winter	Sv	0 1	PW (Seedling) 3	34
OS	98-99	0	Summer	Sf	3	PS	12
PW	98-99	0	Winter	Sf	2	OW	9
OS	99-00	1	Summer	Lv	3	PS	7
PS	99-00	1	Summer	Lv	4	OS	7
OS	99-00	1	Summer	Lf	4	PS	17
IC	99-00	10	Control	Sv	0 1	IC (00-01) 2	20
IS	99-00	0	Summer	Sv	0 1	IS (Seedling) 3	62
IW	99-00	10	Control	Sv	0 1	IC (00-01) 2	20
DC	99-00	20	Control	Sv	0 1	DC (00-01) 2	17
DS	99-00	0	Summer	Sv	0 1	DS (Seedling) 3	63
DW	99-00	20	Control	Sv	0 1	DW (00-01) 2	21
IS	99-00	0	Summer	Sf	1	IS (Mf) 4	5
OS	00-01	2	Summer	Sf	3	PS	5
PS	00-01	2	Summer	Sf	2	OS	5
IS	00-01	1	Summer	Sf	1	DS	10
IS	00-01	1	Summer	Mf	2	DS	31
DS	00-01	1	Summer	Sv	4	IS	25
IS	00-01	1	Summer	Lf	0	DS	61
IW	00-01	0	Winter	Lf	3	Not pooled 5	3

¹ These zeros were due to the technical issue that seedling and small vegetative plants were of same size class and all newly sampled plants <15 cm were considered seedlings (see identifying seedling vs. vegetative plant).

² No matrix of the same year has small vegetative stage, so we pooled small vegetatives of census 98 at PC with small vegetatives of census 99.

³ Based on the data from IW, the only site where a reasonable number of both small vegetative plants and seedlings were present at the time of burn, small vegetative transitions are not significantly different from seedling transitions during the burn year. Therefore, we pooled small vegetative at the burned sites with seedlings of the same census of the same sites during the year of burn.

⁴ Small flowering plants from IS in the burn year could potentially be pooled with the same stage from DS, where the site was burned in the same season of the same year and the two fires were one month apart. However, the effects of the two fires on Chamaecrista keyensis were significantly different, and the pooled column of these two sites had transitions that were drastically different from other columns of the same matrix, indicating such a pooling was not reasonable. Data from OS and PS of the burn year suggested that small flowering plants behaved similarly to medium flowering plants during the year of burn transition. We therefore pooled small flowering plants of IS with medium flowering plants of the same site during the year of burn.

⁵ Large flowering plants in IW 00-01 cannot be pooled with plants of any other site as it was the only site burned in the winter of year 2000. That fire was also unique in that it was a very low intensity fire relative to the other fires.