Quantifying hydrologic benefits of meadow restoration while accounting for interannual and spatial variability

Abstract:

Wet meadows in the Sierra Nevada are highly productive ecosystems that are dependent on an intact hydrologic regime to maintain ecosystem function. For that reason, wet meadow restoration projects are often focused on restoring meadow hydrology, namely surface water and groundwater regimes. We implemented a before, after, control, impact (BACI) experimental design on three meadows in the Tahoe National Forest, utilizing Upper Loney meadow as an upstream reference control meadow and Deer meadow as an adjacent degraded control meadow for comparison with Loney meadow, which was restored in 2017. Over a period of six years (2015-2020), we collected groundwater, surface water, and vegetation data to better understand how the partial fill and restoration of Texas Creek in Loney meadow would alter the surface water hydrologic regime, groundwater levels, and vegetation composition. The experimental design allowed us to account for natural annual hydrologic variability and improve our understanding of the measured benefits of restoration actions at Loney meadow. Our findings support that restoration actions shifted the hydrologic regime and delayed the recession at the outflow of Texas Creek in Loney meadow compared to both pre-restoration conditions and upstream conditions at Upper Loney. At Loney, we saw an 8% increase in groundwater levels during the summer months compared to pre-restoration conditions. For the same time period, groundwater levels decreased at our control sites (Upper Loney -4%, Deer -12%). Loney meadow plant composition shifted toward a wetter plant community, with increased percent cover of FACW and OBL plant species. Plant composition change, measured as beta-diversity, shifted significantly between pre- and post-restoration time steps (p<0.05) in Loney while changes were non-significant (p>0.05) in Upper Loney and Deer. This study highlights where changes in annual hydrologic conditions could confound the quantification of restoration benefits in the absence of control data and will be informative for integrating into future studies where controls are not available.