EA Strategic Monitoring Review Some thoughts about ... · • A case study and questions ... The...

Preview:

Citation preview

the umbrella body of the rivers trust movement

where there’s water, there’s life

Rich Walmsley (EA), Dave Johnson & Michelle Walker (RT)

Monitoring & Evaluation• EA Strategic Monitoring Review• Some thoughts about collaborative monitoring• A case study and questions• Discussion

There was once an astronomer, and engineer and a mathematician on a train.....

There was once an astronomer, and engineer and a mathematician on a train..... All cows in the

universe are black and white

There was once an astronomer, and engineer and a mathematician on a train..... There is a black

and white cow in that field

There was once an astronomer, and engineer and a mathematician on a train.....

There was once an astronomer, and engineer and a mathematician on a train..... Who is right?

Monitoring UncertaintyWhich one is best:• ‘Real’ data vs.

‘Modelled’ data? • Citizen science vs.

‘Professional’ monitoring?

It depends on what we want to find out!Highly accurate data targeted on the wrong thing doesn’t tell us what we want to know!

Monitoring UncertaintyWhat does ‘best’ mean?

Monitoring Uncertainty

Sources of uncertainty:• Measurement

instruments and protocols (precision/reliability)

• What you are measuringmeasuring the right thing

(relevance/validity)• Analysis & interpretation

Solution to uncertainty:• Weight of evidence (constrains

uncertainty..)

e.g. targeting diffuse pollution work in a catchment:

Monitoring Uncertainty

?

? ?

Very precise data targeted on the wrong thing doesn’t tell us what we want to know.

e.g. targeting diffuse pollution work in a catchment:

Monitoring Uncertainty

Imprecise data targeted on the right thing can tell us a lot more.

Monitoring UncertaintyWeight of evidence= collaboration = shared ownership of the evidence base= shared, multi-benefit action

Monitoring Planner

The planner helps us to ask the right questions:

1) Why is the monitoring being carried out?2) What information will be collected?3) How will this be done, by who and at what

cost? 4) Who will the acquired data be used by &

how will they analyse it?

Identified in workshops last year as the most difficult bit of catchment planning

Discussion: Is anyone using the planner?Any feedback for us? What support do you need?

bit.ly/CaBAMonPlanner

A spot sampling case study.....

The issue under investigation: High ammonia at downstream monitoring point

Upstream STW

Downstream monitoring point

11 km

What can spot sampling tell us?

Poor

Moderate

Good

High

Q: If you only had the spot sampling below, what could you say about the cause of the elevated ammonia concentrations at ‘Downstream Monitoring Point’? (chose one statement)

1. There is no evidence that the source of the ammonia problem is ‘Upstream STW’.

2. Elevated ammonia may be associated with pollution 'events' upstream. monitoring not detailed enough to determine whether STW is source

3. Elevated ammonia is due to pollution 'events' upstream. Timing of peak ammonia suggests STW is source.

Q: How could we rapidly reduce the uncertainty as to the source of the pollution events?

Which one of these would you pick?

1. EA or other ecological monitoring (fish, inverts, diatoms)

2. Dry weather walkover survey to identify potential point sources.

3. Plot WQ against flow and or rainfall 4. Riverfly monitoring above and

below U/SSTW discharge.5. Work out the time of travel from

U/SSTW discharge to D/S monitoring.

A continuous monitoring signal give more information.....

Upstream STW Downstream monitoring point

Q: Is it possible that U/S STW is NOT the source of the high ammonia?1. The monitoring below the STW is representative of the outfall not the

mixed river water quality. (Uncertainty)2. The calibration of the kit below the STW is out. (Uncertainty)3. It only happened on a couple of days, one off issue?4. Another? Unknown unknowns........

Downstream monitoring pointUpstream STW

What monitoring would your partnership commission?Ø Why: To find out if the STW is the cause of the eco problemsØ What: Diurnal fluctuations in ammoniaØ How: Continuous sampling (every hour)Ø Who: Matt Lowenthal and his group

Ø Why: To find out if the STW is the cause of the eco problems

Ø What: Ecology above and below STW discharge

Ø How: RiverFly + Ø Who: A very clever aquatic

ecologist

Ø Why: To find out if the STW is the cause of the eco problems

Ø What: Is there any other source of pollution?Ø How: Dry weather walkover Ø Who: A volunteer who is trained to look for

CSO/Septic tanks/Slurry etc

Over to you….

• What do you think are the opportunities?• What barriers are there to developing

collaborative evidence approaches?• What about regulatory monitoring?• How can we support you?

What opportunities could this open up?

• Multiple partners needed to develop and interpret this evidence base

• Develop shared understanding• Shared ownership of the

problem• Identify multiple benefits• Engage community and

stakeholders

What tools and approaches can help?• Data and evidence tools, templates

and guidance• Training collection, interpretation,

synthesis and presentation of evidence

• Tried and tested processes:• Bring partners together using

evidence to develop shared narrative

• Share what we know – evidence inventory – who holds what data?

• Turn evidence in to action –independent broker, workshop to sift through evidence and turn in to business case / shovel-ready projects

What barriers are there?

1. Acceptance of third party data by the EA (not any more?)

2. Licensing of non OpenData (use the EA standard license)

3. Quality of third party data. (think quality vs. relevance)

4. Use of third sector data for regulation and enforcement

5. Training (this is part of a water quality course that we ran...)

6. Other…?

Q: Does the diagram below help us have a common picture of where we are heading? (Over lunch) (prototypes?)

Recommended