Statewide Assessment of the Biotic Condition of California s Perennial Streams. March 2006

Transcription

1 Statewide Assessment of the Biotic Condition of California s Perennial Streams March 2006

2 US Environmental Protection Agency, Region IX California Water Board (SWAMP, NPS) California Dept. of Fish and Game (ABL) Technical development EPA s Office of Research and Development, especially Tony Olsen who created the survey design and produced the condition assessment Chuck Hawkins, Utah State University, predictive model NPS Section 319 funds

3 Outline Condition Assessments context EMAP, CMAP Steps for Producing Condition Assessments Assessment Tools Condition Assessment Result using O/E index

4 California condition assessments: context What is the condition of the nation s waters? Is it getting better? Is it getting worse? Are we allocating $$$$ wisely? National Water Quality Inventory EPA and the States cannot make statistically valid inferences about water quality and lack data to make management decisions. ~GAO 2000

5 Section 305(b) of the Clean Water Act requires states to report on the chemical, physical and biotic condition of their streams: -reports provide an overview of stream health -help agencies optimize allocation of limited WQ resources To date, California has taken a regional approach to reporting: -corresponds to structure of 9 regional WQCBs -inconsistent criteria for targeting sites and assessing condition -large gaps of unassessed streams throughout the state

6 EMAP USEPA s Environmental Monitoring and Assessment Program (EMAP) a long-term research effort to enable status and trend assessments of aquatic ecosystems... with a known statistical confidence. translating environmental monitoring data into assessments of ecological condition... a necessary first step in the Agency's overall strategy for environmental protection and restoration

7 EMAP Strategy 1. Use probabilistic survey design to select sites each site represents a known stream length with known statistical precision provides defensible answers and permits change detection EMAP Design: Spatially Balanced Probability Survey For design details:

8 EMAP Strategy 2. Collect extensive biological, chemical and physical data from each site: Biological: fish, algae, invertebrates Physical: channel morphology and condition, sedimentation index, riparian vegetation Chemical: nutrients, fish tissues, metals 3. Analyze data to make ecological assessments: For example: 25 ±7% of streams in California have degraded fish assemblage 76 ±9% of biologically impaired northern coastal streams are also impaired by fine sediments, while only 6 ±5% of these streams are impaired by nutrients

9 Western EMAP Pilot: W-EMAP in California: -base statewide study (50 sites/year) -special study areas (northern coast, southern coast, central coast)

10 From EMAP to CMAP 1. Produce condition assessments of California (starting with baseline assessment using EMAP data) 2. Continue development of assessment tools needed by the State 3. Tailor monitoring program to meet assessment needs of the State

11 From EMAP to CMAP 1. Produce condition assessments of California (starting with baseline assessment using EMAP data) 2. Continue development of assessment tools needed by the State 3. Tailor monitoring program to meet assessment needs of the State

12 From Survey to Condition Assessment (5 steps) Step I. The population of interest is defined (all perennial and wadeable streams in CA)

13 California Stream Condition Assessment Step II. Probability design is used to generate a list of sites (each site represents a known length of stream) California condition assessments based on 4 combined surveys: Northern Coastal California Southern Coastal California Central Coastal California Statewide -field crews get a list of lat/long coordinates to evaluate More on probability-based surveys: signpages/design&analysis.htm

14 Step III. Site Evaluation

15 Step III. Field site evaluation to determine status of site Southern Coastal California Example Reasons for rejection:

16 Step III: Determining access

17 Step IV. Target sites are sampled: -bugs, fish, algae -reach scale conditions.physical habitat.chemistry.riparian condition

18 Step V. Turning site data into ecological condition assessments Currently we focus on benthic macroinvertebrates because we have tools for scoring them, but other assemblages are also good candidates. will be coming next.

19 From EMAP to CMAP 1. Produce baseline condition assessments of California using EMAP data 2. Continue development of assessment tools needed by the State 3. Tailor monitoring program to meet assessment needs of the State

20 Assessment tools for California Tools for scoring ecological condition using benthic macroinvertebrates 1. Index of Biotic Integrity Regional indices: Northern and Southern Coastal IBIs for links 2. Observed/Expected Index Newly developed for California gives us the ability to score ecological condition for sites throughout the state organism-based approach: Compare number of observed ( O ) taxa to number of expected ( E ) taxa

21 Cumulative Distribution Function: O/E scores vs. cumulative stream length assessed =the condition assessment Cumulative % Stream Length ~35% of stream km impaired ~10% very impaired Impairment Threshold (0.77) 3 sd 1.5 sd O/E Score (RIVPACS output)

22 - To avoid understating impairment, use higher cutoffs - To avoid overstating impairment, use lower cutoffs landowner denials and other barriers to access prevent assessment of some of the stream length

23 Key issues when interpreting condition assessments: 1. Bias in unsampled locations (e.g., if landowner denial sites are much better or worse than average) 2. Where impairment cutoffs are drawn Cumulative % Stream Length O/E Score

24 From EMAP to CMAP 1. Produce condition assessments of California (starting with baseline assessment using EMAP data) 2. Continue development of assessment tools needed by the State 3. Tailor monitoring program to meet assessment needs of the State

25 CMAP Five year design stratified by four landcover classes: urban, agricultural, forested, other sites/ year sampled statewide starting in 2004, same data collected as EMAP What s next? Work with FISH and ALGAE data to develop indicators Build on current RIVPACS models with new reference sites Improve our understanding of BMI response to key stressors Timber harvest plan (THP) data AG data Develop better BMI tolerance values using EMAP datasets

26 Probability surveys have much to offer California water quality community: directly through assessments and stressor associations; indirectly through tool development

27 CONTACT INFORMATION: Emilie Reyes, Surface Water Ambient Monitoring Program (SWAMP), California State Water Resources Control Board, Pete Ode, Dept. of Fish and Game-Aquatic Bioassessment Laboratory, REPORTS posted at Condition assessment of coastal streams in southern coastal California (draft) Probabilistic assessment of the biotic condition of perennial streams and rivers in California. (draft) Development of a benthic index of biotic integrity for wadeable streams in northern coastal California and its application to regional 305b assessment. (draft)

28 Selecting a Geographically Balanced Set of Sampling Points from a Stream Network Stream network is converted into a line Segments all given IDs Segment lengths are preserved Segment arrangement is geographically balanced throughout the region of interest (i.e., California)

29 Approach allows us to see how extent and risk varies with stressor concentration Extent of Stream Length

30 Concept of relative risk Analogous to medical health advisories (e.g., smoking is associated with a 12 fold increase in risk of lung cancer ) Relative Risk: increased probability of biotic impairment that is associated with high stressor levels high nitrate concentrations associated with a 12X higher chance of biotic impairment than low nitrate streams

31 Comparing stressor extent and relative risk (all examples have high risk) * technique provides an objective mechanism for determining greatest threats to stream condition

32 Dealing with disproportionate stream lengths: Strahler Order Total 1st 2nd 3rd 4 th + Rivers ,000 1,500 2,000 Length (1,000 km) Non Perennial Perennial PROBLEM: Smaller streams make up the vast majority of stream segments Simple random selection of sites would bias toward small streams SOLUTION: Increase the probability of sampling large streams by increasing the segment length