Floyds Fork Nutrient TMDL Technical Advisory Committee (TAC) Meeting

MINUTES
/
March 27, 2013
/
9:00 AM
/
USGS, Louisville KY
Meeting called by / KY Division of Water / Type of meeting / TACMeeting
General Information: ,
Attendees / Facilitator: Adam Scott
EPA: Amy Newbold; Tim Wool
TetraTech: Brian Watson
Note Taker (KDOW): Kaye Brothers, Jenny Owen, Lisa Hicks
Peter Goodmann, Paulette Akers, Brooke Shireman, Jenny Howard Owen, Clark Dorman, Lisa Hicks, Hui Chen, Amy Siewert, Ann Fredenburg, Abigail Rains(KDOW); Kevin Gibson (OCEA), Scott Fleming (City of Shepherdsville), Tara Brinkmoeller (Home Bldg Assoc.), Joshua Edwards (City of Shepherdsville), Laura Knoth (KyCGA), Adam Andrews (KyCGA), Vicki Meredith (KSA), Richard Preston (KyCGA), Wayne Long (UK), Roger Recktenwald (KACo); Joe Cain (KyFB), Todd LaFollette (Oldham Co. Health Dept.), Michael Ballard (Lou/Metro Health), Dawn Riley (Plant Food Co.), Joan Lindop (Sierra Club), Brent Burchett, Ryan Bivens (Ky Soybean), Jay McCants (Agor Business Assoc.), Patrick Dominik (SWL, Inc.), Patrick Fitzgerald (MSD), Teena Halbig (FFEA), Tommy Clark (Lou. Mayor Dept.), Matt Milliano (Jeff. Co. Farm Bureau), Tim Joice (Ky Waterways Alliance), Mark Ferguson (NRCS), Junfeng Zhu (KGS), Karen Schaffer (URS), Beth Stuber (Oldham Co.), Tibor Horvath (KACo), Kori Andrews (SMG/BCSD), Scott Smith (SMG/Ky Home Bldg), ShriVani Seripada (SMG)
Peter goodmann / Introduction, handouts, agenda and website information. There is a sign-in sheet please be sure to sign it. We have a lot to cover this morning: EPA morning Tetra Tech afternoon. I will be leaving shortly to attend another meeting in Frankfort but I have a couple of things of concern. KDOW continues to colelct data for stream listings – some of them offer confirmation, some show improvement. Based on new data, we can make changes to listings – this info is dynamic, but the TMDL is a certainty. The only enforcement comes from Clean Water Act on permitted entities. What do we need to do? Wasteload allocations are the “meat” of the issue. Example, where are loads coming from. What we know and what we need to know and make sure this issue is dealt with. Information is sent and we are committed to fix. Calibrating out of Point Source.
SCOTT SMITH / We have questions. On the ag side, we feel there’s a tremendous disconnect. We’d like to address these issues before EPA and TetraTech leave.
PETER GOODMANn / KDOW will still be able to make adjustments. We will have access to Tim Wool.
Richard Preston / Get model at adjusted level do we have to accept the model?
Tim Wool / We will be able to adjust the model.
PETER GOODMANn / We can make adjustments and can control the model and its parameters. We get keys to the car and learn to drive; we are on a learning curve at present.
SCOTT SMITH / Last time when we found inconsistencies, we had phone time with Brian; will we still be able to do that?
peter goodmann / KDOW will have that duty now – and EPA is available.
Richard PRESTON / Can recalibration be done by KDOW?
BRIAN WATSON / Our contract is to develop the model and then to train KDOW – to hand it over. So, yes.
PETER GOODMANn / Model calibration is important. We’ll be looking at whether the inputs reality based and how well calibrated can we get the model.
joe cain / Errors found. Are we able to adjust? Is that part of the calibration?
adam andrews / Is that changing the numbers or formulas?
brian watson / Equations in the model are fixed; equations based on data are not, as inputs are flexible.
Scott Smith / We need a meeting to discuss targets.
peteRgoodmann / The model is new to us, but setting targets we know more about – and that we feel good about. We may find some small changes, but they’re not going to change a lot.
scott SMITH / There is enough concern about how they were developed. There is a disconnect as to what is real or math calculations. How it is done and how to move forward.
peteR GOODMANn / Please revisit the Technical Paper provided to you. It was presented last year, by Lara Panayotoff (KDOW). At this point we are looking at new data. Talking about it and covering it thoroughly.
scott SMITH / I think that is one reason we are concerned. We need to understand how it’s done. Some #s seem totally unrealisticNeed meeting with KDOW on target development.
PETER GOODMANn / They are not unrealistic. We can talk about it but not today. If we have new people, we can catch them up, but today we have too large a group to do something we have already done before.
Scott SMITH / There’s enough discontent within the group. It needs more discussion.
PETER GOODMANn / Targets are separate from the model. These are all concerns that can be addressed.
ADAM SCOTT / We can set up a call to see if we need informational setting for explanation of technical paper. We need to move on.
joe cain / More open discussion is needed as we continue to find issues. The errors are enough for concern.
Laura Knoth / We need discussion on targets and TMDL. Targets have to make sense and should be appropriate for the urban setting.
peteR GOODMANn / There can be no difference in urbanc/rural setting. Warmwater aquatic habitat is warmwater aquatic habitat. Aim at water quality standards. Water quality standards are the same for urban, rural, wooded streams, etc. Their uses may be different. When we determine WLA, we keep in mind knowns, unknowns and how certain we are. Money not spent on unimportant areas. Feel like targets are reachable, and are within 20% of reality. Management decisions are based on amount of data available.
ADAM ANDREWS / Wouldn’t that affect the model?
PETER GOODMANn / Keep eye on how certain we are, when we establish wasteload allocations and have 50% of information we want. We take into consideration what we don’t know.
adam andrews / Are targets attainable?
Ann Fredenburg / Very reachable. Need to separate TMDL from water quality standards. The stream reaches now in some of the lower reaches.
PETER GOODMANn / TMDL is done to reach the water quality standards. Targets are derived from Reference Reach streams. The model runs scenarios of achievability. The Reference stream used in FF Watershed model is already at target; it can be reached. KY has a narrative standard for nutrients; we need # for TMDL. We went through that process a year ½ ago. The numbers are set and explained in the technical paper.
richard preston / I’m just saying that you don’t have enough proof that you have found the appropriate threshold.
peteR Goodmann / As we get more data things may change.
LAURA KNOTH / Was there a public process for targets?
PETER GOODMANn / No, it’s a regulatory duty of KDOW.
laura Knoth / We need dialog; Walk through targets. The keyword is dialog; we need to talk back and forth.
ADAM ANDREWS / We’ve had no dialogue in this process. We’ve had these meetings, and we’ve been educated, but they’re not open dialogue.
peteR GOODMANn / Technical documents are on the website; move to scenarios and how they’re used by the model. nothing is final; we want the TMDL to be beneficial. We will open a dialogue for understanding current processes, not discussion on alternate methodologies.
richard preston / Appear to be off by factor of 10 in 35%of watershed What use is it to look at scenarios?
adam scott / Continue to get data.
TIM wool / May not have all loads correct but we’re pretty sure we’re not off by orders of magnitude. Will be able to tell relative differences between rates. Can look at agriculture areas to see how model works.
peteR GOODMANn / We can fix it. Tetra Tech will do everything to make the model work. The model is going to help to make decisions.
BRIAN WATSON / This manure loading rate issue is a new one. Adam, Chad, and I will talk it through and get it resolved.
TIM WOOL / [Presentation on model: “How good is good enough?”]
Model is only as good as the parameters within it. Constraints in making assumptions – not perfect, but useful.
After May 15thTetera Tech’s contract expires. Will stil be able to get answers to questions. EPA will be available for assistance.. Floyds Fork has a unique mix of urban and agriculture, which needs sophisticated tools. ; This TMDL is based on an endpoint – without it, we can’t do the simple reduction. Kentucky needs help with Floyds Fork; nutrients need numbers to come up with tools. Standard concentrations; complex system that needs time and money. New data = refined info. Stakeholders are essential.At the end we will be comfortable with the model. Magnitude of decision is enormous and we understand that. What is magnitude and what does it mean to me? Green-good; Red/Orange-where we need to do better. KDOW makes a decision; EPA approves. Not perfect and may not suit you. Based on reality model can be changed. KDOW still collecting data. We are helping them. KDOW does the TMDL; KDOW responds to public and submit to EPA for approval. Public has the right to challenge. If proof is shown-change can be allowed. End point: water quality (clean water act) and TMDL.
ADAM ANDREWS / So what is the process? Who makes the final decision?
TIM WOOL / Model can be changed as long as it’s reality. Floyds Fork is 303d-listed, which requires a TMDL. EPA reviews and approves. The TMDL is developed by KDOW, with technical assistance of EPA and stakeholders. Public process includes stakeholder comments, to which KDOW must respond. EPA reviews the TMDL as a records review. EPA’s Water Quality and TMDL staff will review and approve/disapprove.
ann FREDENBURG / KDOW has responsibility to develop the TMDL; EPA just verifies that method is accurate.
Adam scott / Let’s get started again. Are there any questions now with Tim and Tetra Tech?
richard PRESTON / I reviewed the report the last couple of days. Brian you worked with UK with loading crop land; reasonable #s for fertilizer:31 lbs for corn per acre per yr. Ref. 5 went to 260 for crops. Why the difference? Is that an error? For pastureland, 18 lbs. N but 233 in water; appears to be an error by a factor of 10. Did you review the data?
brian WATSON / No it’s manure application rates. Cropland applications are average of all numbers that went into model The loading rates for fertilizer were not changed.
ADAM ANDREWS / The #s appear to be back at first methodology, of everything in county – which we had fixed in fertilizer review. We thought we were beyond it. After the January Ag. Meeting, we were encouraged; now, we’re not. Zero manure is added to cropland.
richard preston / 18 lb nitrogen fall per year; that’s a red flag. We decided manure loading is incorrect. Did you review your data? Why did you let it happen?
brian WATSON / Values received have not changed. Information in the report is average of different crops; and crop loading has not changed. What is manure loading rate? Manure reviewed - number used has been there. Biggest change late last year Dr. Lee reported back – numbers went up but the procedure didn’t change. Concern – went through pain explaining manure. Had fixed fertilizer and added to manure.
Adam Andrews / Process did appear to change. Went back to methods of same numbers for all county numbers in Floyds Fork. When manure added, fertilizer rates did not change.
richard PRESTON / Agree that fertilizer #s were not changed; Why was manure not checked?
brian WATSON / Email went out to agriculture folks. Frustrated that you think we didn’t address the manure factor. There are a page and half of comments. It was the same process. Got numbers and looked at sites. The model is still under predicting in areas dominated by forested and pastureland land uses.
adam scott / We need to move on.
richard PRESTON / Last comment. When did we find out about this. Last night Chad and I checked numbers; ridiculous numbers. There is a discrepancy. If 92% of load is NPS. Can’t be right that 18% N with 233 load.
ann Fredenburg / The in stream data is higher than loading rates the model is showing us. We have an issue with pasture/forest. We need to work on that. Is theerror in groundwater? Need to identify discrepancy between model and in-stream data.
brain watson / Yes
SCOTT SMITH / New topic: Golf course assumptions. This may be where elevated NPS numbers are coming from?
TIM WOOL / Sub-basins can be reviewed
Brian WATSON / Golf course assumptions non-point source equals 93 percent – 1 of 7 stations high percent of load from non-point source. Largest % of NPS load is pasture.
tim wool / After lunch let’s discuss in depth. Could be very small amount
adam andrews / Assuming manure applied, actual contribution is 0. It’s not an issue, not happening.
brian WATSON / Good thing to say manure load being added. We can address. Problem / largest loading: pasture land. Only row crops being treated.?
tim WOOL / Will work out going forward.
joe cain / How do you have the numbers when nothing is applied to it? How do you get that 15 percent of the watershed assuming 92%?
TIM WOOL / 92% is that sub-basin.
laura KNOTH / Model, in our opinion, applies wrong #s to livestock. Number of animal units; pasture/crop. Would like to walk through it. For us we thought it was wrong.
brian WATSON / Let’s revisit; not making numbers up. Numbers told to us; they have not been changed. If there is a disagreement then lets talk.
adam ANDREWS / Got numbers three weeks ago, agreed on fertilizer rates. We looked at them three weeks late–saw 92% and got discouraged. Why did #s, methodology change? Thought it would be fixed
tim WOOL / Nothing was done to be deceitful.
richard preston / We do this everyday. We are trying to help. We don’t get paid for this. We volunteer. It is our livelihood. We are fearful after 1-1/2 yrs. There is a timeline.
adam scott / Worried there is a timeline, that we’ll be told we can no longer talk about it. Same with targets, we were told “we’re working on it, working on it” then the #s were set and there was no place to talk about it.Perceived notion.
tim Wool / Understand totally, we get that. The end will come. Give it a chance. We can talk later and will talk more about this. Targets won’t change much. Even with current endpoints, the scenarios will provide options.
CLARK DORMAN / I’m wondering if perhaps the group didn’t realize the importance of the targets. We will provide another discussion, in Q&A format, if needed – how developed, why/when changed; maybe small % changed but not much. The aquatic life determines that.
laura KNOTH / Look at our numbers for forestry and numbers impervious they are similar. What was evaluated? Square foot not acres? MSD / point sources – what was evaluated. Ammonia? Why not Total Nitrogen?
TIM WOOL / Total N was considered; ammonia was not.
LAURA KNOTH / Large/small treatment plants – differences? Golf course numbers.
ADAM ANDREWS / Golf courses measure their loading on square feet, not acres
TIM WOOL / Golf courses can be changed, but won’t make a lot of difference
laura knoth / Have to get model right because of how it might be applied across state – get it right the first time.
tim wool / This model will not be globally applicable to every watershed across the state.
scott SMITH / Working through difficulties?
adam andrews / We want to make sure that the process is right.
tim wool / We have to work through these issues. These are site specific models.
roger recktenwald / Clarify what Ann said? Explain why that is the case.
ann FREDENBURG / We have in-stream measured data. If the model is underpredicting that value, we need to figure out why.
roger recktenwald / That would definitely call into question the assumptions plugged into the model.
SCOTT SMITH / We’re trying to improve model, too – golf courses, WWTPs. Model not perfect but correctable errors. We’re wanting to help fill the gaps.
tim WOOL / We can explore why the model is so much lower than the measure data. The model over predicts in some spots and under predicts in others.
Scenarios will show relative differences between them to allow us to feel comfortable with model.
scott SMITH / Curious, help find areas. We are trying to help.
adam andrews / Manure – did they have time to review before meeting?
tim wool / NRCS 590 standard for manure… new to us.
Scenarios are not the TMDL. Relative difference will be shown in color.
brian watson / Take TN - look at data down stream assumptions. Reflective of dominant source.
ann fredenburg / Collecting additional nutrient data from 23 facilities. Collecting nutrient data.
scott smith / How groundwater, run-off, and wildlife were contributed? We need a better understanding of how MSD data used. AG converted to subdivision underestimated. Factor for wildlife in model?
ann fredenburg / Included in the calibration.
brian watson / Some standards and trends shown as multiple data. Different land use contributes. We will show this after lunch.