The spatial pattern of light gaps in mesic grasslands in central Texas with contrasting disturbance histories was assessed using patch-based landscape metrics determined from a threshold level (25% of full sunlight), as light intensities below this threshold substantially decrease survival of honey mesquite (Prosopis glandulosa var. glandulosa Torr.) seedlings. The spatial pattern of light gaps, with the exception of edge density, were significantly different between annually-disturbed and non-disturbed grasslands on all sample dates (2 April, 30 April, 29 May, and 26 June 1998). Differences in patch metrics did not occur between nondisturbed grasslands despite contrasting vegetation composition [perennial forbs and perennial bunch (tussock) grasses]. Patchbased landscape metrics of light gaps did vary temporally in both annually-disturbed and non-disturbed grasslands. The structure and spatial configuration of light gaps were distinctly different between annually-disturbed and non-disturbed grasslands: a low density of large patches characterized light gaps in annually-disturbed grassland, whereas non-disturbed grasslands had a high density of small patches. Our findings demonstrate that the current disturbance regime is the principal environmental driver influencing species dominance and composition, and indirectly vegetation structure, which collectively contribute to the observed dynamics of light gap patches in these mesic grasslands. Incorporating spatially explicit consideration of light gap structure and dynamics into experimental studies addressing invasion of weedy plant species such as honey mesquite may be an effective approach to address mechanisms and the ecological significance of disturbance operating as a driver facilitating woody plant invasions in mesic grasslands.

The spatial pattern of light gaps in mesic grasslands in central Texas with contrasting disturbance histories was assessed using patch -based landscape metrics determined from a threshold level (25% of full sunlight), as light intensities below this threshold substantially decrease survival of honey mesquite (Prosopis glandulosa var. glandulosa Torr.) seedlings. The spatial pattern of light gaps, with the exception of edge density, were significantly different between annually-disturbed and non -disturbed grasslands on all sample dates (2 April, 30 April, 29 May, and 26 June 1998). Differences in patch metrics did not occur between nondisturbed grasslands despite contrasting vegetation composition [perennial forbs and perennial bunch (tussock) grasses]. Patchbased landscape metrics of light gaps did vary temporally in both annually-disturbed and non -disturbed grasslands. The structure and spatial configuration of light gaps were distinctly different between annually-disturbed and non -disturbed grasslands: a low density of large patches characterized light gaps in annually -disturbed grassland, whereas non -disturbed grasslands had a high density of small patches. Our findings demonstrate that the current disturbance regime is the principal environmental driver influencing species dominance and composition, and indirectly vegetation structure, which collectively contribute to the observed dynamics of light gap patches in these mesic grasslands. Incorporating spatially explicit consideration of light gap structure and dynamics into experimental studies addressing invasion of weedy plant species such as honey mesquite may be an effective approach to address mechanisms and the ecological significance of disturbance operating as a driver facilitating woody plant invasions in mesic grasslands.