Effective Water Quality Monitoring Programs

Transcription

1 Effective Water Quality Monitoring Programs Nancy Mesner: USU-WATS Dept, Douglas Jackson-Smith: USU-SSWA Dept; David Stevens, Jeff Horsburgh, Darwin Sorensen: USU-CEE Dept; Ginger Paige: University of Wyoming

2 Overview Background of Little Bear Project Analysis of historic WQ Data Alternative approaches More effective monitoring

3 Little Bear River Conservation Effectiveness Assessment Project

4 From , bank stabilization, river reach restoration, off-stream watering, manure and water management, grazing management

5 Project Objectives: Did BMPs implemented in the watershed result in lower phosphorus loads? What were the strengths and weaknesses of different water quality monitoring programs? How should limited funds for BMP implementation be used most effectively to improve water quality?

6 Analysis of Historic Water Quality Trends

7 Total Phosphorus, mg/l Seasonal Kendall Trend for TP concentration at Mendon Rd (mouth of LBR) Conservation project initiation No Significant 0.6 slope before Slope mg/l yr Since Date

8 Observations Trends suggest water quality improvements Data Record Insufficient to. Tease out Exogenous Variables Link Trends to BMP Implementation Support Traditional Modeling Approaches Biological Data were variable and inconclusive

9 Alternative approaches to evaluating BMP impacts

10 Use of videography to evaluate riparian area BMPs

11 Videography Analysis Located 1992 aerial 3-band videography for the Little Bear River Re-flew the river in 2007 Classified vegetative conditions for both time periods within identical riparian zones Riparian trees Small shrubs & grasses Bare soil Water & Shadows

12 Identified paired reaches of river for comparison Upstream WITHOUT BMPs (controls) Downstream WITH BMPs

13

14

15 Percent of Riparian Zone Percent of Riparian Zone by Vegetation Type, BMP and Non-BMP Impacted Zones 80% 70% 60% 50% 40% 66% 61% 53% 39% 30% 20% 10% 20% 17% 15% 17% 17% 9% 15% 24% 18% 13% 8% 9% 0% Water/Shadow Riparian Trees Shrubs & Grasses Bare Soil BMP areas 1992 BMP areas 2007 Non-BMP areas 1992 Non-BMP areas 2007

16 Percent Change Percent Change in Riparian Vegetation by BMP Status, 1992 to % 60% 55% 40% 20% 25% 33% 0% (-20%) (-40%) (-60%) (-46%) (-47%) (-46%) (-47%) (-45%) (-48%) Riparian Trees Shrubs & Grasses Bare Soil BMP area NonBMParea Overall

17 Summary of riparian analysis Riparian conditions improved throughout watershed (more trees, less exposed soil) BMPs installed in areas with less vegetation BMPs associated with much more rapid growth in tree cover, similar rates of decline in exposed soil Fences = reduced exposed soil most In-stream work = increased trees the most

18 Implementation and maintenance of BMPS

19 Methods Gathered formal practice info from NRCS files Went through every file 90 landowners / participants Conducted field interviews Validated file information Contacted 70 of 90 participants 55 agreed to be interviewed Conducted field interviews - ~90 minutes Detailed discussion about BMP experience

20 Findings - Implementation Individual BMPs 83% of BMPs successfully implemented At a farm-level 32% farms implemented all BMPs 60% farms implemented more than ½

21 Maintenance of BMPs Overall 21% of implemented BMPs were no longer there ( = 1/3 of all originally contracted BMPs) Why were BMPs no longer in place? No longer farming or sold land 32% Still farming, no longer use 68%

22 BMP Implementation & Maintenance by "Type" Structural Planting, Clearing and Leveling Management Percent implemented Percent original BMPs still there Percent maintained

23 Implications: Maintenance Good news: Producers did not discontinue the practices because they did not like them Not so good news: The management practices had the shortest lifespan ALSO: Change in ownership or change in farmland use can also affect long term impacts

24 Evaluation of targeting of critical areas

25 Growing recognition of landscape variability and the need to target implementation to critical areas Critical Areas: areas where the potential contribution of pollutants to the receiving water is significantly higher than other areas

26 Combined Map of Risk Zones

27 Percent BMP coverage in farm fields with different potential impact on water quality 67% of farm fields have VERY LOW potential impact 23% 8% of farm fields have MODERATE potential impact 62% 18% of farm fields with LOW potential impact 7% of farm fields have HIGH potential impact 47% 47%

28 Implications of spatial analysis Evidence exists that higher risk zones were targeted with BMPs (not random) More than ½ of riskiest areas covered by BMPs Suggests opportunity for greater targeting & efficiency Inefficient targeting may be due to other program objectives.

29 Installation of high frequency monitoring stations to collect surrogate data

30 Since 2005, measure flow and turbidity at 30 minute intervals Stage recording devices to estimate discharge Turbidity sensors Dataloggers and telemetry equipment

31 Correlations with grab samples

32 --> high frequency TSS and TP datasets

33 Designing more effective monitoring programs

34 Common problems in many BMP monitoring programs Failure to design monitoring plan around BMP objectives Failure to identify and quantify sources of variability in dynamic systems. Failure to understand pollutant pathways and transformations choosing inappropriate monitoring approaches

35 Common problems in many BMP monitoring programs Failure to design monitoring plan around BMP objectives Failure to identify and quantify sources of variability in dynamic systems. Failure to understand pollutant pathways and transformations choosing inappropriate monitoring approaches

36 v Little Bear River Watershed, Utah

37 Total Observations at Watershed Outlet site Discharge Total phosphorus : : Number of observations each year

38 Was the original UDWQ monitoring program a failure? No.Program was intended to detect exceedences of water quality criteria. The failure was ours. In attempting to use these monitoring data for detecting change in loads

39 Failure to design monitoring plan around BMP objectives Failure to identify and quantify sources of variability in these dynamic system. A failure to understand pollutant pathways and transformations choosing inappropriate monitoring approaches

40 Flows can be highly variable and exhibit multi-year trends

41 Upper Site Flow (cfs) Turbidity (NTU) Seasonal and annual variation Variation between sites Lower Site Flow (cfs) Turbidity (NTU) January December 2006

42 Impact of rare events TSS Load Upper Site Lower Site Annual (kg) 8.9 X X 10 7 Runoff (% of total) 89% 54% Baseflow (% of total) 11% 46% Storms (% of baseflow) <1% 16%

43 Effect of sample timing

44 Variability in correlations between surrogate (turbidity) and target pollutant Upper Site

45 Coefficient of variation of estimates The relative importance of two sources of variability in estimates of annual phosphorus load Sampling frequency Regressions of TP and turbidity Grab samples -- sampling frequency (d) Continuous monitoring -- R 2 between TP and turbidity

46 Summary of site specific sources of variability in sampling data Seasonal and annual variability in flow and in water quality parameters Timing of sampling Relationship of surrogate to target pollutant may change between sites and at different times of the year Rare events need to be sampled

47 Failure to design monitoring plan around BMP objectives Failure to identify and quantify sources of variability in these dynamic system. A failure to understand pollutant pathways and transformations choosing inappropriate monitoring approaches

48 Problems with one-size-fits-all monitoring design kg / day Rees Creek TSS load Problem: excess sediment Average flow = 20 cfs 5000 BMP = series of in-stream sediment basins 0 Above Below weeks

49 load (kg/day) Bear River phosphorus load Problem: excess phosphorus 50 Average flow = 1000 cfs 0 BMP = fence cattle OUT of riparian area and revegetate weeks

50 Rethinking Monitoring there is no one size fits all monitoring program

51 Designing monitoring programs to evaluate BMP effectiveness Monitoring plans require careful thought before anything is implemented. Consider how the data will be used to demonstrate change. Use your understanding of your watershed and how the pollutants of concern behave to target monitoring most effectively. Use different approaches for different BMPs.

52 Best Management Practices Monitoring Guide and Web Site Focuses on the considerations and decisions necessary as a project is first being considered. NOT a how-to manual of protocols

53 Section 1: The importance of clear monitoring objectives Long term trends? UPDES compliance? Educational? Assessment for impairment? Track response from an implementation?

54 Section 2. Understanding pollutants within a natural system. How does the pollutant move from the source to the waterbody? How is the pollutant processed or transformed within a waterbody? What is the natural variability of the pollutant? Will concentrations change throughout a season? Throughout a day? What long term changes within your watershed may also affect this pollutant? What else must be monitored to help interpret your data?

55 Section 3. Consider spatial and temporal scale

56 Section 4. Monitoring and modeling: strengths and weaknesses of different approaches Pollutant Direct Monitoring Surrogate Monitoring Other important variables * Response variables Models Temperature Probes, launched monitors (e.g. hobo), and direct measurements Light / shading, ground water signal (stable isotope variables) Air temperature, flow, time of day, depth, turbidity, cloud cover Algae, macros, and fish CEQual WASP(7) SNTEMP (USGS) Dissolved Oxygen (DO) Probes and direct measurements Temperature, redox, and Flow /temperature/algal biomass Temperature will affect percent saturation, depth, flow, velocity Macros and fish Streeter Phelps Nutrients (phosphorus and nitrogen) Grab samples and integrated samples In some cases use probes, or streamside auto-analyzers to collect surrogates Turbidity or sediment ph, temperature, and DO might affect the solubility of phosphorus, flow, sediment transport Algae, macros, and fish UAFRI SWAT QUAL2K Sediment Grab samples and integrated samples Turbidity Flow Physical characteristics, embeddedness, macros, and algae PSIAC /AgNPS SWAT KINEROS2 SELOAD Salts / TDS Probes and grab samples Riparian vegetation Flow Macros and fish QUAL2K Pathogens Grab samples and integrated samples Fecal Coliform Bacteria, E.coli Turbidity, nutrients Human health, livestock health Metals Grab samples Bioaccumulation in DO might affect Bacteria in the MINTEQAQ

57 Section 5. Choosing the best monitoring design Control Treatment A Abovetreatment monitoring stations Sampling points Below-treatment monitoring stations BACI Design Above and below treatment design

58 Section 6: Site-specific considerations Legal and safe access? Is access limited during winter? Equipment limitations? - Power needs / satellite or cell phone access Holding times of your samples?

59 Section 7. Protocols Physical / habitat Biological Water Column Behavioral monitoring

60 Section 8. Quality Assurance and Quality Control Provides built in feedback on project plan so improvements can be made. QA: Project management considerations, including how data will be stored, documentation of methods. QC: Error control Look at the data results make sense?

61 Section 9. Data Management Good site and sample ID system Metadata that help explain to others the system and the protocols Data sheet design helps avoid missed observations Transcribe data and results to electronic file system. Don t trust your memory

62 Section 10. Analysis of data Classification: Group by similarity or by characteristics Correlation: Look for linkages among observations Implied causality, relationships Modeling: Cause and effect, prediction

63 Section 11. Interpreting and using the data Keep objectives in mind How well are BMPs working? What are lessons learned? What needs to be changed or revised?

64 Also in manual: Monitoring resources Modeling resources Terms and definitions Contacts

65

66 Decision Tree Identifies KEY components Shows links between components Links to information in the Guidance doc Non linear!!

67 Check List Method to help identify KEY components that need to be considered Takes one through the thought process.

68 Manual available through USU Water Quality Extension

69 Future Actions Assistance with Watershed Coordinators in developing effective monitoring plans; Application of many of the lessons learned on a Utah watershed project Evaluation of effectiveness of Utah s NPS program. Testing manual in 6 states

70 QUESTIONS? CONTACT INFO: This research is supported by NIFA CEAP Competitive Watershed Grant ; NIFA 406 Watershed grant