The URL can be used to link to this page

Home My WebLink AboutDMS_Detecting_WQ_Change_Melia_Haywood-3NCSU EcoStream Conference August 13-16, 2018 Asheville, NC 2 Determining the Likelihood of Detecting Change in Water Quality Resulting from Stream Restoration Practices over Mitigation Time Frames Greg Melia and Casey Haywood NC Department of Environmental Quality Division of Mitigation Services EcoStream Conference August 13-16, 2018 Asheville, NC 3 DMS WQ Sites Project County # Reaches Param Storm Base Heath Dairy Randolph 2 F,N,S,M Y Y Millstone Randolph 2 F,N,S,M Y Y Millstone Randolph 1 F,N,S Y Y Pen Dell Johnston 1 F Y Buckwater Orange 1 F,N,S Y Y Big Harris Cleveland 5 F,N,S Y Y Big Harris Cleveland 8 M Y F –Fecal; N –Nutrients; S –Total Suspended Res; M–Macrobenthos 4 DMS WQ Sites 5 Station Setup and Methods Drivers for Assessing Higher Level Ecological Functions Such as Water Quality and Biology in Mitigation 6 Department of Environmental Quality 2010 EEP –DMS Mitigation Instrument DMS Mitigation Plan Guidance Claim Uplift Prove Uplift The Challenges 7 The Challenges Data Requirements ❑Storm and baseflow separation. Hard to tell the story without storm data. ❑Number of storms (~15 pers. com NCSU). ❑Based on weather and errors need about 1.5 -2 years for a given treatment phase (i.e. pre -post). ❑Watershed control station. ❑Possible need for reference site(s) pre -con. The ChallengesThe Challenges 8 Optimizing Water Quality Monitoring Plans ❑Objective 4 Where pre-con data is available and/or direct measurement is required/advisable apply meaningful criteria and statistical tools to refine and optimize the post-con monitoring plan. Project County # Reaches Param Storm Base Heath Dairy Randolph 2 F,N,S,M Y Y Millstone Randolph 2 F,N,S,M Y Y Millstone Randolph 1 F,N,S Y Y Pen Dell Johnston 1 F Y Buckwater Orange 1 F,N,S Y Y Big Harris Cleveland 5 F,N,S Y Y Big Harris Cleveland 8 M Y F –Fecal; N –Nutrients; S–Total Suspended Res; M–Macrobenthos MDC =Allows you to estimate the amount of change necessary to support statistically reliable change detection. This is based on the variability observed in the parameters distribution. 9 Baseflow Base and Stormflow Optimizing Water Quality Monitoring Plans Big Harris Pre-con Water Quality Monitoring Scope Station 0 1 2 3 4 5a 6 7 8 9 10 11 12 13 14 16 17 18 19 20 Fecal Cond Solids NH3 TKN NOx TP Macro Fish 10 Optimizing Water Quality Monitoring Plans Criteria and Analyses Applied to Pre-con Data ❑Are the existing levels of concern? Reference Sites 11 Optimizing Water Quality Monitoring Plans Criteria and Analyses Applied to Pre-con Data ❑MDC values >50% were considered too high Example :Variability in data pre-construction data for TSS at station 4 produced an MDC of 81%. High MDC (low probability of reliable change detection) TSS mg/L MDC 11.86 MDC%81 12 Optimizing Water Quality Monitoring Plans Criteria and Analyses Applied to Pre-con Data ❑Proposed restoration treatment(s) for reach(s) represented by sampling have the opportunity to address the main stressors Example:Constraints or landowners will not permit stabilization of ephemeral gullies that are producing the bulk of the sediment load. Does it make sense to expect meaningful TSS reductions? 13 Optimizing Water Quality Monitoring Plans Criteria and Analyses Applied to Pre-con Data ❑Pre-con data indicates one or more other stations will adequately represent the station that was dropped. The application of these criteria and the analyses performed on the pre-con data converted the scope from this…. 14 Station 0 1 2 3 4 5a 6 7 8 9 10 11 12 13 14 16 17 18 19 20 Fecal Cond Solids NH3 TKN NOx TP Macro Fish Baseflow Base and Stormflow Optimizing Water Quality Monitoring Plans Big Harris Pre-con Water Quality Monitoring Scope 15 Optimizing Water Quality Monitoring Plans Big Harris Post-con Water Quality Monitoring Scope Station 2 3 5a 6 8 9 10 13 14 Fecal Base and Storm Cond Baseflow Solids Stormflow NH3 TKN NOx TP Macro Fish 16 Optimizing Water Quality Monitoring Plans Criteria and Analyses Applied to Pre-con Data ❑Data driven. ❑Technically Sound ❑~50% cost-scope reduction between pre and post ❑Optimized. Questions that Need to be Addressed 17 Which parameters and under what land use histories will detection of change be possible in mitigation timeframes? What are the key variables of a reach and its watershed that govern the likelihood of detecting change? What sort of sample sizes and levels of effort (i.e. cost) will typically be required to enable statistically reliable detection of change? How do we arrive at appropriate performance standards and optimize post-construction sampling plans? 18 DMS Monitoring Plan and Objectives ❑Overarching Goal of DMS Plan. Provide information and data resources to the mitigation/restoration community that will assist practitioners in making decisions about the inclusion of water quality goals and performance standards at the reach scale and to augment models and tools with quality data. This will reduce the need for direct measurement of water quality in the long run. 19 DMS Monitoring Plan and Objectives ❑Objective 1 –Near Term. Conduct or compile rigorous water quality monitoring of restoration stream reaches to determine the reach and watershed attributes that permit documented improvement within mitigation timeframes. Provide case examples Project County # Reaches Param Storm Base Heath Dairy Randolph 2 F,N,S,M Y Y Millstone Randolph 2 F,N,S,M Y Y Millstone Randolph 1 F,N,S Y Y Pen Dell Johnston 1 F Y Buckwater Orange 1 F,N,S Y Y Big Harris Cleveland 5 F,N,S Y Y Big Harris Cleveland 8 M Y F –Fecal; N –Nutrients; S –Total Suspended Res; M–Macrobenthos Heath Dairy –NCSU (D.E. Line) larger reach showed storm load reductions ranging from 41 to 67% for nutrients and solids. Smaller reach only demonstrated reductions in NH3/4 20 DMS Monitoring Plan and Objectives ❑Objective 2 –Long Term. a.Over the long term compile enough reach outcomes with simple hypothesis tests (i.e. significance determined Yes or No between pre and post) from monitored reaches or from the literature with adequate monitoring practices to generate a quality reach data set. b.It is intended that the reach data set will span the reach and watershed variables that are likely deterministic in supporting reliable statistical detection of change. (i.e. can we find thresholds?) c.develop a model/relationship in which variables determined to be explanatory yield a probability of detection for each water quality parameter ❑Example Explanatory Variables . 1.Watershed Proportionality Variables e.g. drainage area e.g.treated footage as a proportion of the drainage network e.g. treated area as a proportion of the drainage area 2.LULC Variables –Continuous Variables e.g.Proportion forested cover and/or grassland e.g. Proportion Ag 21 DMS Monitoring Plan and Objectives ❑Objective 3 Use the same data to augment/calibrate existing models and tools to improve their predictive capability hopefully reducing the need for direct measurement given its challenges. 22 Acknowledgements and Citations ❑Casey Haywood –DMS ❑Jamie Blackwell –NCSU and DMS. ❑Dan Line -NCSU ❑Jean Spooner -NCSU ❑DMS Management. Jean Spooner, Steven A. Dressing, and Donald W. Meals. 2011. Minimum detectable change analysis. Tech Notes 7, December 2011. Developed for U.S. Environmental Protection Agency by Tetra Tech, Inc., Fairfax, VA, 21 p. Daniel E. Line 2015. Effects of Livestock Exclusion and Stream Restoration on the Water Quality of a North Carolina Stream. ASABE Vol. 58(6): 1547-1557 Terziotti, Silvia, Capel, P.D., Tesoriero, A.J., Hopple, J.A., and Kronholm, S.C., 2018, Estimates of nitrate loads and yields from groundwater to streams in the Chesapeake Bay watershed based on land use and geology: U.S. Geological Survey Scientific Investigations Report 2017–5160, 20 p., https://doi.org/10.3133/sir20175160. 23 DMS S&A Website https://deq.nc.gov/about/divisions/ mitigation-services/dms-science- data