Fig 1.
Examples of flowering individuals of Bulbostylis paradoxa (Spreng.) Lindm. in the control area (A), after clipping (B), and after a prescribed fire (C) in a campo sujo at the Área Alfa da Marinha, Brasília, DF, Brazil. All photos are from individuals of B. paradoxa observed during the experiment by the authors.
Fig 2.
Effect size (with confidence intervals) on the probability of flowering of individuals of Bulbostylis paradoxa (Spreng.) Lindm. in the control (control, n = 35), after clipping (clipped n = 70), and after a prescribed fire (fire, n = 70) in a campo sujo at the Área Alfa da Marinha, Brasília, DF, Brazil.
The effect size refers to the expected probability of flowering individuals irrespective of the time taken for flowering in each treatment based on a Generalized Linear Model (GLM) with a binomial distribution. Different letters above the bars indicate significant differences (model contrast analysis, P< 0.05).
Fig 3.
Percentage of flowering individuals of Bulbostylis paradoxa (Spreng.) Lindm. over time in the control area (A, n = 35), after leaf clipping (B, n = 70), and after a prescribed fire (C, n = 70), in a campo sujo at the Área Alfa da Marinha, Brasília, DF, Brazil. Arrows indicate the amount (mm) of the first rains registered in the area at the onset of the rainy season. Different letters at the top of the graphs indicate significant differences in the proportion of flowering individuals between the time periods (model contrast analysis, P< 0.05) and n.s. indicates no significative differences. Note that the scales of the x axes differ among treatments, although the comparison among treatments considered the same scale (weeks). The individuals from all treatments were monitored daily during the first week after the fire and weekly after that until nine weeks after the fire. The number of individuals flowering in the fire treatment stabilized in the first week after the fire. Clipping was performed in the same day of the fire.
Fig 4.
Total number of flowering individuals of Bulbostylis paradoxa (Spreng.) Lindm. with (burned) or without (unburned) fire scars through the year in different herbariums that provided high-resolution images in the SpeciesLink database for the Cerrado biome (n = 209).
Fig 5.
A) Percentage of Bulbostylis paradoxa (Spreng.) Lindm. individuals with different numbers of inflorescences per individual after a prescribed fire (Fire, n = 70), and after leaf removal (Clipped, n = 70) in a campo sujo in the Área Alfa da Marinha, Brasília, DF, Brazil. Data for the control area is not included (n = 35, however, only four with inflorescences). B) Percentage of B. pardoxa with different number of flowering stalks with (burned) or without (unburned) fire scars in the high-resolution images of the SpeciesLink database for the Cerrado biome (n = 209).
Fig 6.
Germinative parameters (± SE) of Bulbostylis paradoxa (Spreng.) Lindm. in the control area (Control), after a prescribed fire (Fire), and after leaf removal (Clipped), considering plant seed mass (A), seed germination (B), seed viability (C) and germination time (D) in a campo sujo at the Área Alfa da Marinha, Brasília, DF, Brazil. Different letters at the top of each bar indicate significant differences in mean values (model contrast analysis, P< 0.05). Seed viability = germinated + those stained red with tetrazolium after 45 days.