Forests of the Apalachicola River floodplain had shorter flood durations, were drier in composition, and had 17 percent fewer trees in 2004 than in 1976. The change to drier forest composition is expected to continue for at least 80 more years. Floodplain drying was caused by large declines in river levels resulting from erosion of the river channel after 1954 and from decreased flows in spring and summer months since the 1970s. Water-level declines have been greatest at low and medium flows, which are the most common flows (occurring about 80 percent of the time). Water levels have remained relatively unchanged during large floods which continue to occur about three times per decade.
A study conducted by the U.S. Geological Survey compared temporal changes in hydrologic conditions, forest composition, forest characteristics, and individual species of trees, as well as estimated the potential for change in composition of floodplain forests in the nontidal reach of the Apalachicola River. The study was conducted with the cooperation of the Florida Department of Environmental Protection and the Northwest Florida Water Management District. Forest composition and field observations from studies conducted in 1976-1984 (termed '1976 data') were used as baseline data for comparison with data from plots sampled in 2004-2006 ('2004 data').
Flood durations were shorter in all periods subsequent to 1923-1976. The periods of record used to calculate flood durations for forest data were subsets of the complete record available (1923-2004). At sampled plots in all forest types and reaches combined, flood durations changed an average of more than 70 percent toward the baseline flood duration of the next drier forest type. For all forest types, changes in flood durations toward the next drier type were greatest in the upper reach (95.9 percent) and least in the lower reach (42.0 percent).
All forests are expected to be 38.2 percent drier in species composition by 2085, the year when the median age of surviving 2004 subcanopy trees will reach the median age (99 years) of the 2004 large canopy trees. The change will be greatest for forests in the upper reach (45.0 percent). Forest composition changes from pre-1954 to 2085 were calculated using Floodplain Indices from 1976 and 2004 tree-size classes and replicate plots.
Species composition in high bottomland hardwood forests is expected to continue to change, and some low bottomland hardwood forests are expected to become high bottomland hardwood forests. Organisms associated with floodplain forests will be affected by the changes in tree species, which will alter the timing of leaf-out, fruiting, and leaf-drop, the types of fruit and debris produced, and soil chemistry. Swamps will contain more bottomland hardwood species, but will also have an overall loss of tree density.
The density of trees in swamps significantly decreased by 37 percent from 1976 to 2004. Of the estimated 4.3 million (17 percent) fewer trees that existed in the nontidal floodplain in 2004 than in 1976, 3.3 million trees belonged to four swamp species: popash, Ogeechee tupelo, water tupelo, and bald cypress. Water tupelo, the most important tree in the nontidal floodplain in terms of basal area and density, has declined in number of trees by nearly 20 percent since 1976. Ogeechee tupelo, the species valuable to the tupelo honey industry, has declined in number of trees by at least 44 percent.
Greater hydrologic variability in recent years may be the reason swamps have had a large decrease in tree density. Drier conditions are detrimental for the growth of swamp species, and periodic large floods kill invading bottomland hardwood trees. The loss of canopy density in swamps may result in the swamp floor being exposed to more light with an increase in the amount of ground cover present, which in turn, would reduce tree replacement. The microclimate of the swamp floor would become wa