1. 1 Defining Least-Impacted Reference Condition for the National Wadeable Streams Assessment Alan Alan Herlihy (Oregon State University), John Stoddard (U.S. EPA, NHEERL-WED), Chuck Hawkins (Utah State University), and Susan Holdsworth (U.S. EPA, OW)

2. 2 Why do we want to select and sample reference streams? Are necessary for the interpretation of national monitoring data Are necessary for the interpretation of national monitoring data Provide benchmark for evaluation of ecological condition (e.g. biotic integrity) Provide benchmark for evaluation of ecological condition (e.g. biotic integrity)

3. 3 What is Reference Condition? Minimally Disturbed Condition – site condition in the absence of significant human disturbance (e.g., “natural,” “pristine” or “undisturbed”) Least Disturbed Condition –found in conjunction with the best available physical, chemical and biological habitat conditions given today’s state of the landscape - defined by a set of explicit criteria to which all reference sites must adhere

4. 4 How are reference sites chosen? Often chosen by best professional judgment (BPJ) Often chosen by best professional judgment (BPJ) BPJ sites have varying and unknown quality BPJ sites have varying and unknown quality Alternative: Filter survey data for physical-chemical stressors to identify best sites Alternative: Filter survey data for physical-chemical stressors to identify best sites

5. 5 Mid-Atlantic Highlands EMAP Stream Example Screen all sites and remove those with: Screen all sites and remove those with: Sulfate > 400 µeq/L (~20 mg/L) Acid Neutralizing Capacity (ANC)< 50µeq/L (~ 2.5 mg/L CaCO 3 or pH ~6) Total phosphorus > 20 µg/L Total nitrogen > 750 µg/L Chloride > 100 µeq/L (~3.5 mg/L) Mean RBP habitat score < 15

6. 6 Expected stream sulfate from deposition in this region is 100-300 µeq/L Bimodal sulfate histogram in Plateau. Mining not common in Ridge & Valley (except for Anthracite Belt) Sites with SO 4 >400 µeq/L classified as non-reference

7. 7 Contrary Creek, Virginia pH=3, SO 4 =5,000 µeq/L

8. 8 Montgomery Creek, PA pH=5.1, SO 4 =175 µeq/L

9. 9

10. 10 Filtering produced a set of Reference Sites with higher EPT Richness scores than BPJ Advantages of Filtered Sites Fewer poor biological condition sites Have a much more rigorous definition of “reference”

11. 11 Wadeable Stream Assessment Sample EMAP-West Summer 2000-2004 Summer 2000-2004 12 Western States 12 Western States 841 Probability Sites 841 Probability Sites Hand-Picked Sites sampled by EMAP protocols Hand-Picked Sites sampled by EMAP protocols State BPJ Sites State BPJ Sites Corvallis GIS/Field Screen Corvallis GIS/Field Screen STAR Grant Ref Sites STAR Grant Ref Sites WSA Summer 2004 36 Eastern & Central States 541 Probability Sites Hand-Picked Sites Sampled by EMAP Protocols State BPJ Sites

12. 12 Reference Site Data for Assessment WEMAP Filtered all sites collected with EMAP protocols Filtered all sites collected with EMAP protocols Probability, BPJ, GIS Probability, BPJ, GIS Four Approaches - Delphi Four Approaches - Delphi STAR Grant Sites STAR Grant Sites Pristine Pristine Good Good Low Low Dropped in MT ecoregions Dropped in MT ecoregions WSA Filtered all sites collected with EMAP protocols Probability, BPJ Needed more reference sites, augmented with existing, State Data NAWQA data Earlier Filtered EMAP and REMAP data

13. 13 9 Aggregate Level III Ecoregions used to set filtering criteria

14. 14 WSA Reference Exclusion Criteria Variable S. Appalachians Southern Plains Total P (µg/L) > 20 > 150 Total N (µg/L) > 750 > 4,500 Sulfate (µeq/L) > 400 --- Chloride (µeq/L) > 200 > 1,000 ANC (µeq/L) < 50 Mean RBP Habitat Score < 15 < 12.5 Turbidity (NTU) > 5 > 50 Riparian Disturbance > 2 > 2 % Fine Sediment > 25 > 90

15. 15 Filtered

16. 16 Filtered

17. 17 Filtered

18. 18 Number of Reference Sites Used in WSA EMAP/WSA Filtered External Augment Total Western States 204476680 East/Central States 130815945 Total3341,2911,625

19. 19 Number of Reference Sites by Ecoregion

20. 20 Reference Sites Used to Score Benthic IBI metrics Score Benthic IBI metrics Set Good, Fair, Poor condition categories for Set Good, Fair, Poor condition categories for Biotic Integrity Biotic Integrity Nutrients Nutrients Physical Habitat Physical Habitat Predictive (O/E) modeling Predictive (O/E) modeling Required dropping state augment data Required dropping state augment data