Doing ecohydrology backward: Inferring wetland flow and hydroperiod from landscape patterns

Human alterations to hydrology have globally impacted wetland ecosystems. Preventing or reversing these impacts is a principal focus of restoration efforts. However, restoration effectiveness is often hampered by limited information on historical landscape properties and hydrologic regime. To help address this gap, we developed a novel statistical approach for inferring flows and inundation frequency (i.e., hydroperiod, HP) in wetlands where changes in spatial vegetation and geomorphic patterns have occurred due to hydrologic alteration. We developed an analytical expression for HP as a transformation of the landscape-scale stage-discharge relationship. We applied this model to the Everglades "ridge-slough'' (RS) landscape, a patterned, lotic peatland in southern Florida that has been drastically degraded by compartmentalization, drainage, and flow diversions. The new method reliably estimated flow and HP for a range of RS landscape patterns. Crucially, ridge-patch anisotropy and elevation above sloughs were strong drivers of flow-HP relationships. Increasing ridge heights markedly increased flow required to achieve sufficient HP to support peat accretion. Indeed, ridge heights inferred from historical accounts would require boundary flows 3-4 times greater than today, which agrees with restoration flow estimates from more complex, spatially distributed models. While observed loss of patch anisotropy allows HP targets to be met with lower flows, such landscapes likely fail to support other ecological functions. This work helps inform restoration flows required to restore stable ridge-slough patterning and positive peat accretion in this degraded ecosystem, and, more broadly, provides tools for exploring interactions between landscape and hydrology in lotic wetlands and floodplains.