PIs and collaborators: Shreeram Inamdar, Joe Galella, Sujay Kaushal

Abstract: Stream and floodplain restoration has grown into a multi-million dollar environmental industry in the United States (US). Stream and floodplain restorations are being implemented all across the US to mitigate sediment and nutrient (nitrogen (N) and phosphorus (P)) pollution and to enhance aquatic habitat.  Many of the nutrient and water quality benefits (N and P reductions) from these restorations are also being credited towards regulatory requirements (e.g., total maximum daily load (TMDL)) that municipalities and state agencies have to meet to address the impairment status of the water body. One of the key goals of restoration practices is to enhance the hydrologic connectivity between the stream and floodplains to promote N loss via the process of denitrification.  Denitrification converts reactive nitrate-N in runoff waters to inert N2 gas and thus provides a valuable ecosystem service. Very few of the restorations however, check to see if nitrogen losses are occurring due to denitrification and if the restoration is effective. Monitoring runoff nitrogen exports is time consuming, expensive, and subject to variations, and measuring soil denitrification is difficult. Microbial denitrification process however alters the stable nitrogen isotope ratios in soil due to the preferential removal of the lighter nitrogen isotopes resulting in an enriched isotopic signal in the soil. This change is time integrative and could provide a robust metric for long-term assessments of floodplain restorations. We hypothesize that where restorations that are effective, i.e., where denitrification N removal in floodplain soils is elevated, we will have enriched (high) soil nitrogen isotope values and these values will be higher than the adjacent unrestored reaches. We further expect that the isotope values will be highly variable immediately after restoration (0-2 years), but increase to high values for 5-20 years following restoration, and then level off beyond that (> 20 years) because of N limitations. We will test this hypothesis for 12 selected restored sites across the mid-Atlantic (Maryland, Delaware and Pennsylvania). Five different time periods post restoration will be sampled: 0-2, 2-5, 5-10, and > 20 with three sites within each category. Denitrification assays will also be performed to compare against soil 15N isotope values. Rigorous statistical analysis will be performed to compare unrestored and restored sites and how the differences between the two vary with restoration age.             

SAMPLED SITES 

Final Project Report (to be added)

Data sets

Refereed publication - Memories of the soils: Decadal changes in soil properties, denitrification rates, and 15N in floodplains following restoration (in preparation). 

FINAL RECOMMENDATIONS: (to be added)