INNOVATIVE RESTORATION SOLUTIONS FOR A MODERN EVERGLADES BASED on HISTORIC LANDSCAPE FORM AND PROCESS

The Everglades ridge and slough landscape is an ecologically significant patterned floodplain that evolved under unconstrained sheetflow and low nutrient inputs. Degradation of landscape patterning has coincided with twentieth century reductions in water level and flow, as well as increases in phosphorus inputs. Restoring landscape patterning is now a focus of restoration efforts. However, it is not feasible to reset all of these variables to reference conditions, nor has it been clear how changes in individual variables would impact the complex feedbacks governing Everglades landscape morphology.

The reference system or historical range of variability (HRV) approach to ecosystem restoration is often criticized for its shortcomings in forecasting nonlinear behavior, its use in a nonstationary environment, and its inability to recognize valid restoration solutions outside the range of historic conditions. We developed a process-based modification of the HRV approach in which we combined best available knowledge of historic landscape conditions with mechanistic modeling to distinguish between competing mechanisms of landscape evolution. Modeling revealed that a sediment redistribution feedback process interacting with a peat accretion feedback process was sufficient to produce a landscape consistent with historical conditions. A global sensitivity analysis yielded key insights for restoration that would not be possible in a more traditional HRV approach: 1) Hysteresis in landscape evolution necessitates the removal of emergent vegetation from choked sloughs if a return to historic flows and water levels is to produce a return to historic landscape patterning, and 2) Long-term patterned landscape stability requires pulsed flows in which velocities transiently exceed the critical shear stress for sediment entrainment. Despite the present-day constraint that time-averaged flows are inevitably going to be much slower than in the original system, we found episodic engineered pulsed flow events may be sufficient to accomplish restoration objectives. Thus, by inferring processes from the historical range of variability and then using simple modeling to understand system behavior over an expanded range of conditions, the shortcomings of a traditional HRV approach to restoration can be avoided.