The GLADES Project is a long-term, spatially explicit study of tree demography, vegetation dynamics, and disturbance in an oak savanna/woodland landscape at Cedar Creek. The project was initiated in 1988 following an exceptionally hot and dry summer, when the absence of pre-existing, high-temporal-resolution tree data at the site highlighted the need for a monitoring system capable of documenting ecological responses to individual-year events such as drought and fire.
The permanent study grid was established in 1988. It is a square, 16-ha plot divided into 1,600 cells measuring 10 × 10 m. Grid intersections are marked by 1,681 tagged rebars. The grid is oriented approximately north–south and east–west and uses a Cartesian coordinate system for spatial reference.
Each cell is identified by the coordinates of its southwest corner. The grid was shifted 20 m east during establishment to avoid placing its westernmost cells within or beyond an existing firebreak. Consequently, the working grid origin is 20,0, with the remaining corners at 20,400; 420,400; and 420,0.
Individual stems are assigned both to a grid cell and to an estimated x–y position within that cell. This permits analyses at scales finer than the 10 × 10 m cell and allows tree demographic records to be integrated with topography, soils, fire history, aerial imagery, and other spatially referenced variables.
A prescribed-burn program in the area began in 1987. The original design divided the site into three burn units: a frequently burned unit, generally treated every two to three years; a less frequently burned unit, initially treated approximately every five years; and an unburned control.
After 2000, the burn frequency of the second unit increased to approximately every two to three years, and prescribed burning was also initiated in the former control unit. Beginning in 2003, all three units were generally burned in the same years. As a result, the original long-term contrast among frequent, infrequent, and unburned treatments was not maintained.
Total number of live stems (all sizes) in 1995/96 = 9362, with 5333 (57.0%) of the stems being bur oaks, 3129 (33.4%) being pin oaks, and 900 (9.6%) being mesophytes.
Of the stems alive in 1995/96, 2787 (29.8%) were still alive in 2024: with most of stems being bur oaks (2286, 82.0%), followed by pin oaks (437, 15.7%), and the mesophytes (64, 2.3%).
In 1995/96, the small stems (<10 cm dbh) represented 47.7% of the live stems; medium sized stems (dbh > 10 cm and < 25 cm) constituted 39.6% of the live stems; and 12.7% of the live stems had a dbh > 25 cm (large stems).
In 2024, the size distribution of stems was very different from what it was in 1995/96: small (7.9%), medium size (62.2%), and large (31.9%).
Davis, MA and Condit, R. 2022. Neighbours consistently influence tree growth and survival in a frequently burned open oak landscape. Journal of Ecology 110: doi: 10.1111/1365-2745.13906.
Davis, MA. 2021. Twenty-five years of tree demography in a frequently burned oak woodland: implications for savanna restoration. Ecosphere 2(12):e03844. 10.1002/ecs2.3844
Davis, MA and Miller, A. 2018. Savanna restoration using fire benefits birds using dead trees, up to a point. American Midland Naturalist 179: 94-104