Stream Restoration as a Nutrient Reduction Practice:

Size: px

Start display at page:

Download "Stream Restoration as a Nutrient Reduction Practice:"

Spencer Burns
6 years ago
Views:

1 Stream Restoration as a Nutrient Reduction Practice: A Test Case of Field-Measured Results vs. Predictive Erosion Rates Using Chesapeake Bay Program Protocols

2 Agenda 1 Introduction 2 Stream Restoration CBPO Expert Panel & 2013 White Paper 3 SR Protocol 1: Test Case #1 Estimating Stream Bank Erosion 4 SR Protocol 1: Test Case #2 Nutrient Concentrations in Soils 5 Stantec Research & Development (R&D) 6 Questions

3 Stantec in the Mid-Atlantic The Chesapeake Bay is the jewel of the Mid-Atlantic Josh Running - Senior Environmental Planner (Williamsburg, VA) Alex Foraste, P.E. Senior Water Resources Engineer (Richmond, VA)

4 Little History on Chesapeake Bay Efforts : CB Agreement leading to formation of CB Program Office and CB Executive Council & 2000: CB Landmark Agreements : E. O. declaring CB a National Treasure : CB TMDL established; 6 Bay States and DC begin WIP development to achieve 2025 goals : Regulatory changes in Virginia alter way MS4 localities & agencies plan and develop in the Bay & 2023: Incremental Numeric Reduction Target dates for VA MS4s ~1.25% & ~8.75%) Target Date: Reduction of Pollution Levels by 20-25% over 2009 levels.* *Cost estimated at $7-10 Billion. Costs are for SWM only (Total = $ B if include Ag, WW) and are attributed to Local Governments and State Agency in Virginia. Costs (source: VA Senate Finance Committee).

5 The Role of Stream Restoration Degraded and Eroding Urban streams are and can be a significant source of sediments and nutrients. Some estimates have found: almost ¾ of the sediment in streams come from channel and bank erosion with only about ¼ coming from upland soil erosion. (Osmond et al summarizing several watershed studies) Stream restoration is very cost effective solution ($/lb basis). CBPO estimates that 418 miles of Urban Stream Restoration will be implemented in VA and MD alone by 2025.* *(NOTE: estimates include historical projects and is derived from Phase 2 WIP submissions to EPA in 2012 and summarized by Jeff Sweeney of EPA CBPO.)

to quantify sediment and nutrient reductions from individual

6 CBPO Stream Restoration Expert Panel Report WEG (Stantec) invited to test drive Report ~ May 2013 Oct 2013 Developed to outline methods to quantify sediment and nutrient reductions from individual projects in an effort to credit projects to help offset reduction requirements.

7 CBPO BMP Expert Panel: Stream Restoration Methods to Quantify Reductions: - Default Removal Rate Fixed rate of TN, TP, TSS reductions per L.F. of stream restoration (ex: lbs/lf/yr x 1,000 LF = 68 lbs TP/yr) OR - Application of 4 Protocols from Expert Panel Report they are:

CBPO SR Expert Panel Report: 4 Protocols P1-

Flow P3- Credit for Floodplain Reconnection

8 CBPO SR Expert Panel Report: 4 Protocols P1- Credit for Prevented Sediment During Storm Flow P2- Credit for Instream & Riparian Nutrient Processing within the Hyporheic Zone During Base Flow P3- Credit for Floodplain Reconnection Volume P4- Dry Channel RSC as an Upland Stormwater Retrofit

9 2013: WEG (Stantec) Test Drive and White Paper Project A Project B Project C

10 2011: WEG (Stantec) White Paper on SR & Nutrient Crediting 2013: WEG (Stantec) Test Drive of CBPO SR Bay Protocols and White Paper Application of Protocols summarized & compared to Default Rate Difference between Project B and Spring Branch Project is > 9 times!!! Average ~ close, but

Default Removal Rate One Size fits all? 0.

11 Default Removal Rate One Size fits all? lbs TP/ft/yr Sediment loss from stream banks varies depending on rate of lateral erosion, bank heights, stream order, channel geometry, landscape position, sediment dynamics, historical development, conditions in upland watershed, soils, vegetation, and other factors lbs TP/ft/yr = lbs TP/ft/yr

12 Default Nutrient Concentration in Stream Bank Soils (P1 BANCS) One Size fits all? 1.05 lbs TP/1 ton Sediment Nutrient Concentration in stream bank soils can vary widely depending on soil type, geology, vegetation, historical land use, soil applications, and other factors

13 Protocol 1 Prevented Sediment Acceptable Approaches: BANCS Method (BEHI/NBS) for yearly tonnage with default concentration of 1.05 lb/ton P, 2.28 lb/ton N Site monitoring with bank pins/toe pins/cross-sections, soil samples and precipitation monitoring Alternative Modeling Approach

14 Protocol 1 Two Test Case Examples Test Case #1 Lateral Stream Bank Erosion Rate Comparison of: a.) Default Removal Rate b.) Application of BANCS Method with default nutrient concentration values c.) Site monitoring of Stream Bank retreat rates over time with bank pins Test Case #2 Nutrient Concentrations in Stream bank Soils Comparison of: a.) Default Nutrient Concentration Value (cited in P1) b.) Measured Nutrient Concentration values

15 Protocol 1 Sediment Prevention Test Case #1 Lateral Stream Bank Erosion Rate

16 Test Case #1 Example Project Example: Zombie Apocalypse 10 + Headcut Banks No Vegetation left Highly erodible soils Just nasty

Protocol 1 Comparison (Test Case #1) Approved Default Removal Rate: 419 LF @ 0.068 lbs TP/ft/yr = 28.5 lbs TP/yr 419 LF @ 0.075 lbs TN/ft/yr = 31.

17 Protocol 1 Comparison (Test Case #1) Approved Default Removal Rate: lbs TP/ft/yr = 28.5 lbs TP/yr lbs TN/ft/yr = 31.4 lbs TN/yr BANCS Method (using NC Curves & actual bank heights of 15 ft 23 ft) BEHI = Extreme; NBS = Moderate Total Erosion = 2518 Tons/Yr (1322 lbs TP/yr, 2871 lbs TN/yr)* *Using default concentrations, after 50% reduction Modified BANCS (using NC Curves & reduced bank heights of 10 max) Total Erosion = 1049 Tons/Yr (551 lbs TP/yr, 1196 lbs TN/yr)* *Using default concentrations, after 50% reduction

19 Monitoring Needed (Test Case #1) Network of Bank Pins Soil Concentrations Rainfall Observation ½ year No Bankfull events (conservative)

projects that have shown through monitoring that the higher rates can be justified subject to approval by

20 CBPO Protocol 1 Page 33 : Monitoring through methods such as cross section surveys or bank pins is the preferred approach Page 36 : The Panel felt that efficiencies greater than 50% should be allowed for projects that have shown through monitoring that the higher rates can be justified subject to approval by the states. This will hopefully promote monitoring (e.g., Big Spring Run in Pennsylvania) of stream restoration projects.

Monitoring Results (Test Case #1) Extrapolated for 1 year Provided range using lower soils concentration 90% efficiency was estimated (rather then 50%) Bank Length (ft) Lateral Erosion (ft)

21 Monitoring Results (Test Case #1) Extrapolated for 1 year Provided range using lower soils concentration 90% efficiency was estimated (rather then 50%) Bank Length (ft) Lateral Erosion (ft) Bank Height (ft) Volume Lost (cf) SUM /21/14 03/10-15 (~5 months)

Monitoring Results (Test Case #1) # Method Description 1 Default Removal Rate Notes Fixed Rate 2 BANCS 15 ft 23 ft bank hts 3 BANCS 10 ft max bank hts 4 Monitoring

22 Monitoring Results (Test Case #1) # Method Description 1 Default Removal Rate Notes Fixed Rate 2 BANCS 15 ft 23 ft bank hts 3 BANCS 10 ft max bank hts 4 Monitoring (bank 5 Monitoring (bank w/ 525 ppm TP (default) w/ 128 ppm TP (measrd) TP (lbs/yr) TN (lbs/yr)

65M) Project Implementation : $700K (includes construction and soft cost)

23 Zombie Apocalypse (Test Case #1) Phosphorous : $15,000 LB in WS Phosphorous value (110 LB TP@ $15K = $1.65M) Project Implementation : $700K (includes construction and soft cost) Project Cost per pound ($700k/111lb TP= $6,364 lb TP) Cost to monitor & produce justification $5,000 Having data to support your MS4? : Priceless

24 Protocol 1 Sediment Prevention Test Case #2 Nutrient Concentration in Stream bank Soils

05 lbs TP/ton sediment (~525mg/kg) selected as CBPO default value for ALL projects.

25 Protocol 1 Sediment Prevention (Test Case #2 Nutrient Concentrations in Stream bank Soils) CBPO Default TP concentration: aaaa 2013 White Paper Sample Findings: 1.05 lbs TP/ton sediment (~525mg/kg) selected as CBPO default value for ALL projects. However, range is (10 x) Looked at 16 past Restoration Reaches w/ 124 bankline soil samples

26 Protocol 1 Sediment Prevention (Test Case #2 Nutrient Concentrations in Stream bank Soils) What WEG 2013 White Paper found:

27 Protocol 1 Sediment Prevention (Test Case #2 Nutrient Concentrations in Stream bank Soils) WEG (Stantec) 2013 White Paper Findings: Number of Projects Sample Locations by Physiographic Province 1 Test Year Total # of Samples 1 TKN Conc. Range (lbs TN/ton SED) 2 TKN Conc. Avg. (lbs TN/ton SED) TP Conc. Range (lbs TP/ton SED) 2 TP Conc. Avg. (lbs TP/ton SED) 16 Piedmont lowland & upland, Coastal Plain All projects in tidewater and northern Virginia; most projects tested 2-5 samples; three projects contained a large number of samples; 2 TKN as Total Kjeldahl Nitrogen; TP tested with USEPA SW-846 method; total samples for TKN less than TP 3 All samples tested at A&L Eastern Laboratories in Richmond, VA and reported as ppm; results coverted to lbs/ton of SED by WEG. Summary: 124 sample Average = 0.33 lbs TP/Ton Sediment w/ range of (vs lbs TP/Ton Sediment CBPO default) (High value is 100 x greater than low value)

Required to show uptick in water quality value to proceed

28 Test Case #2 - Nutrient Concentrations in Stream bank Soils (Example) Potential Mitigation Bank Located in the Piedmont Required to show uptick in water quality value to proceed Spring-fed streams eroding into pasture, minimal wooded riparian corridor

29 Quick Data needed Didn t have time to monitor the site for even ½ year. Performed a BANCS (NC Curve) : 358 tons/year Collected soil samples within the channel and in the field

30 Results Soils within Woods: 314 ppm TP, 1344 ppm TN Soils within the field: 1250 ppm TP, 2000 ppm TN Need to consider the heads as a large percentage of nutrient input. Major difference in field vs woods?

31 Bio-solids!

32 Findings: Pluses and Minuses Default Removal Rate: Ok for Planning purposes, but final estimates should be based on site specific methods. Default rate does not factor in bank height, severity of channel degradation, watershed land use, or soils and may over/under estimate SIGNIFICANTLY (10 x or more). Could incentivize selection of streams w/ only minor degradation (same credit for <$). Over time, this could potentially lead to decreased reductions (lb/lf) Expert Panel SR Protocols: Does offer better approach to capture site specific conditions. Monitoring is best, both for sake of accuracy, but also may lift 50% efficiency cap, nearly doubling credit.

Findings: Pluses and Minuses P1: Prevented Sediment BANCS: More site specific stream channel conditions and can be assessed in a short period of time.

33 Findings: Pluses and Minuses P1: Prevented Sediment BANCS: More site specific stream channel conditions and can be assessed in a short period of time. However, limited availability of regional Bank Erosion Rate Curves and Hickey Run or NC? Selection can affect results by multiples (~4x). Need exists for local Bank Erosion Rate Curve(s). Nutrient Concentrations: Observed conc. in stream bank soils varied by multiple of 100 times ( ppm TP); Default value in CBPO may be high on average; actual is sometimes +/- and is site dependent. Sampling costs are very low ($25 lab fee). Monitoring (Toe/Bank Pins): Time consuming, more expensive, but Greater Reductions -Ability to measure/document much greater erosion in severely eroding streams; Search for Gross Pollutors - Encourages/rewards search for and fixes to the gross pollutors, aka the Zombie Apocalypse Increased Efficiency - Able to increase efficiency (pre to post restoration), nearly doubling credit. Headcut Migration - Able to capture reductions from repair of activity migrating headcuts. P1 (BANCS) accounts only for lateral erosion of existing centerline, not upstream migration Improved Accuracy - Better science and encouraged by CBPO.

34 Stantec s Solutions Monitor when you can! Most projects typically have years from assessment to construction. BANCS if time limited Improved local data and curves Stantec s R&D Program.

35 Stantec s Solutions CBPO Stream Restoration Expert Panel, Section 8.2 Research Management and Needs: Provide support for the development of regional stream bank erosion curves for the BANCS method using local stream bank erosion estimates throughout the watershed and a statistical analysis of their predicted results. Ideally, measured bank erosion rates within each subwatershed or County would be used to validate the BANCS Method specific to that location.

Stantec s Solutions Stream Bank Erosion Rate

36 Stantec s Solutions Stream Bank Erosion Rate (BER) Curves: Limited Availability (e.g. Yellowstone, Colorado, Hickey Run, NC) Selection of Curve can affect results significantly and NC or Hickey? Need exists for local BER curve(s) based on local conditions and field data

Stantec s R&D Program Stantec R&D Program Grant Awarded: May 2015 to develop VA regionally specific stream bank erosion rate curves Participants: contacted clients and municipalities with

37 Stantec s R&D Program Stantec R&D Program Grant Awarded: May 2015 to develop VA regionally specific stream bank erosion rate curves Participants: contacted clients and municipalities with letter detailing program; we monitor, they get results Initial Results: 26 candidate sites ID d (Fairfax, James City, Prince William, Henrico, York County, Town of Ashland, Town of Vienna)

38 Stantec s R&D Program Goal: produce urban stream bank erosion rate curves in the Piedmont and Coastal plain Results: produce & publish BER curves that can be used with Expert Panel P1 BANCS method. Everyone can benefit from better data, better science. Anticipated timeline: May 2015 Sept 2015: Grant Award, Project Setup, and Candidate Site Identification Sept late 2016: Data Collection and Evaluation

39 Questions Thoughts?

Figure 0-18: Dendrogeomorphic analysis of streambank erosion and floodplain deposition (from Noe and others, 2015a)

Figure 0-18: Dendrogeomorphic analysis of streambank erosion and floodplain deposition (from Noe and others, 2015a) Appendix 9A: Stream to River During the development of the Phase 6 Watershed Model, multiple methods for determining coefficients were often attempted. In some cases, the methods are averaged or otherwise