17-003684RP Basf Corporation vs. Department Of Environmental Protection Status: Closed Settled and/or Dismissed prior to entry of RO/FO on Monday, June 11, 2018.

Summary: DEP failed to prove that the proposed rule, which would establish a Total Maximum Daily Load for nutrients in Lake Talquin, was not an invalid exercise of delegated legislative authority.

STATE OF FLORIDA

DIVISION OF ADMINISTRATIVE HEARINGS

BASF CORPORATION,

Petitioner,

vs. Case No. 17 - 3684RP

DEPARTMENT OF ENVIRONMENTAL

PROTECTION,

Respondent,

and

LEON COUNTY, FLORIDA, AND THE

CITY OF TALLAHASSEE,

Intervenors.

________________________ _______/

FINAL ORDER

The final hearing in this case was held on November 28

through 30, 2017, in Tallahassee, Florida, before Bram D.E.

Canter, Administrative Law Judge of the Division of

Administrative Hearings (ÐDOAHÑ).

APPEAR ANCES

For Petitioner BASF Inc.:

Matthew Z. Leopold, Esquire

Martha Harrell Chumbler, Esquire

James Parker - Flynn, Esquire

Carlton Fields Jorden Burt, P.A.

Post Office Box 190

Tallahassee, Florida 3230 2

For Respondent Florida Department of Environmental Protection:

Lorraine M arie Novak, Esquire

Kenneth B. Hayman, Esquire

Carson Zimmer, Esquire

Department o f Environmental Protection

3900 Com monwealth Boulevard , Mail Station 35

Tallahassee, Florida 32399 - 3000

For Intervenor Leon County Florida:

Herbert W.A. Thiele, Esquire

Leon County AttorneyÓs O ffice

301 South Monroe Street, Room 202

Tallahassee, Florida 32301

William B. Graham, Esquire

Kevin P. Blodgett, Esquire

Carr Allison

305 South Gadsden Street

Tallahassee, Florida 32301

Edwin A. Steinmeyer, Esquire

Steinmeyer Fiveash LLP

310 West College Avenue

Tallahassee, Florida 32301

For Intervenor City of Tallahassee:

Gary V. Perko, Esquire

David C. Childs, Esquire

Malcolm N. Means, Esquire

Hopping Green & Sams, P.A.

Post Office Box 6526

Tallahassee, Florida 32314

STATEMENT OF THE ISSUE

The iss ue to be determined in this case is whether proposed

Florida Administrative Code Rule 62 - 304.305(5) is an invalid

exercise of delegated legislative authority.

PRELIMINARY STATEMENT

On December 16, 2016, the Department of Environmental

Protection (ÐDEPÑ) published a Notice of Proposed Rulemaking in

the Florida Administrative Register that would establish in

Florida Administrative Code Rule 62 - 304.305(5) a new Total

Maximum Daily Load (ÐTMDLÑ) for nutrients in Lake Talquin, and

would allocate these loads am ong specific and general categories

of nutrient sources.

The notice also included a summary of the Statement of

Estimated Regulatory Costs (ÐSERCÑ) for the proposed TMDL. On

January 27, 2017, BASF Corporation (ÐBASFÑ) filed a lower cost

regulatory alterna tive (ÐLCRAÑ) with DEP. On March 6, DEP

rejected the LCRA on substantive grounds and because it was

untimely filed.

On June 19, DEP published a Notice of Change in the Florida

Administrative Register. On June 26, BASF filed a petition

challenging the pro posed rule. BASF was subsequently granted

leave to amend its petition.

Leon County and the City of Tallahassee were granted

intervention into the proceeding in support of the validity of

the proposed rule.

DEP moved to dismiss BASFÓs petition on the basi s that BASF

lacked standing. The motion was denied.

BASF filed a Motion for Summary Final Order on the issue of

whether DEPÓs SERC was inadequate. The motion was denied. BASF

filed a Second Motion for Summary Final Order, oral argument was

heard, and t he issues raised in the motion are addressed in this

Final Order.

At the final hearing, Joint E xhibits 1 - 8 were accepted into

evidence. Official recognition was taken of chapter 2013 - 156,

Laws of Florida.

BASF presented the testimony of Adrienne Lee; Mic hael

Erickson, who was accepted as an expert in water quality

modeling and hydrodynamic modeling; and Thomas Frick, the

Director of the DepartmentÓs Division of Environmental

Assessment and Restoration. BASF E xhibits 1, 9, 10, 14A (only

for the comments o f Mr. Erickson), 15 (only for the comments of

Mr. Erickson), 26, 41, 43, 44, 64A, 64B, 64C, 77, 81, and 103

were accepted into evidence. BASF Exhibit 46 was accepted as a

proffer only.

DEP presented the testimony of Julie Espy, Program

Administrator of DE PÓs Water Quality Assessment program; Thomas

Frick, who was accepted as an expert in waterbody/watershed

assessment and restoration; and Tim Wool, the United States

Environmental Protection AgencyÓs national expert in water

quality modeling, who was accept ed as an expert in TMDL

modeling. DEP Exhibits 1 - 14, 17, 18, 20, 21, 33, 41 - 43, 45, 47,

51, 53, 60 - 62, 64 - 69, and 73 were accepted into evidence.

Intervenor Leon County did not present any witnesses. It

offered two exhibits, but they were not accepted i nto evidence.

Intervenor City of Tallahassee did not present any witnesses and

did not offer any exhibits into evidence.

The Transcript of the hearing was filed with DOAH. The

parties filed proposed final orders that were considered in the

preparation of the Final Order.

FINDING S OF FACT

I. The Parties

1. Petitioner BASF is a Delaware corporation that operates

a cracking clay facility near Attapulgus, Georgia, about three

miles north of the Florida/Georgia border.

2. Respondent DEP is the state agency authorized under

section 403.067, Florida Statutes, to assess waters for the

attainment of water quality standards and to develop TMDLs for

water s that do not meet the standards.

3. Intervenor Leon County is the local government in which

part of Lake Talq uin is located. The County owns and operates

stormwater collection systems that discharge nutrients into the

Lake Talquin watershed.

4. Intervenor City of Tallahassee owns and operates the

A.B. Hopkins Power Plant that discharges nutrients into the Lake

Talquin watershed.

II. The TMDL Program

5. In compliance with section 303(d) of the federal Clean

Water Act, section 403.067, Florida Statutes, was enacted to

require DEP to assess and list waterbodies for which water

quality standards are not being atta ined and for which

technology - based effluent limitations and other pollution

control programs are not sufficient to attain water quality

standards.

6. Such Ðimpaired watersÑ are placed on a ÐVerified ListÑ

by DEP and TMDLs are developed for them.

7. Afte r a TMDL has been established, a plan for restoring

the waterbody is developed and implemented through a basin

management action plan (ÐBMAPÑ).

III. Lake Talquin

8. Lake Talquin is located on the border between Gadsden

and Leon Counties in Florida, appro ximately 20 miles from the

Georgia border. It was formed when the Jackson Bluff Dam was

constructed on the Ochlocknee River.

9 . The parties disputed the size of the watershed for Lake

Talquin. DEP believes the watershed encompasses 1,569 square

miles. BASF contends the watershed is about 1,700 square miles,

because that is the figure in some Environmental Protection

Agency (Ð EPA Ñ) reports. DEP used the smaller watershed size

because four small sub - basins included in the EPA reports are

not hydraulicall y connected to Lake Talquin.

1 0 . Based on the figure of 1,569 square miles, 73 percent

of the Lake Talquin watershed is in Georgia and 27 percent is in

Florida.

1 1 . About 57 percent of the watershed is natural,

consisting of upland forest, open water, an d wetlands. Urban

and residential land uses make up 7 percent of the watershed.

Agricultural lands, including rangelands, occupy 36 percent of

the watershed.

1 2 . In 2009, DEP determined that Lake Talquin was impaired

for nutrients, using the Trophic Sta te Index. The Trophic State

Index for high color lakes like Lake Talquin, is an annual

geometric mean of chlorophyll a of 20 micrograms per liter

(µg/L) , total phosphorus (ÐTPÑ) between 0.05 and 0.16 ´ g/L, and

total nitrogen (ÐTNÑ) between 1.27 ´g/L and 2 .23 µg/L, not to be

exceeded more than once in any consecutive three - year period.

1 3 . Chlorophyll a is the target indicator because it

indicates the presence of algae.

1 4 . For water quality assessments to determine whether

waters are impaired, DEP has de signated waterbody identification

units (ÐWBIDsÑ). Waterbodies can be divided into more than one

WBID, based on hydrologic distinctions. Lake Talquin was

divided into two WBIDs, designated 1297C and WBID 1297D. Both

WBIDs were determined to be impaired for nutrients.

1 5 . In 2013, DEP re - evaluated Lake Talquin using the

Trophic State Index and found both WBIDs were still impaired for

nutrients. Therefore, the Lake Talquin WBIDs remained on the

Verified List.

IV. The TMDL Model

1 6 . DEP collaborated wit h EPA in developing the Lake

Talquin TMDL. Tim Wool, an EPA employee and an expert in the

application of water quality models, conducted the modeling used

in developing the TMDL.

1 7 . Mr. Wool used a suite of three models: the Loading

Simulation Program C++ (ÐLSPCÑ); the Water Quality Analysis

Simulation Program (ÐWASPÑ); and the Environmental Fluid

Dynamics Code (ÐEFDCÑ) (referred to collectively as the ÐTMDL

ModelÑ).

18 . LSPC is a watershed model that was utilized to

simulate all of the water flows and associated nutrient loads

that come off the land surfaces in the upstream sections of the

watershed, as well as the surrounding watershed that drains

directly in to Lake Talquin, including Little River and the

Ochlockonee River.

1 9 . The WASP model was de veloped by Mr. Wool and is an

advanced water quality model that simulates pollutant Ðfate and

transport.Ñ Three separate WASP models were used; two riverine

models, one for the Little River and one for the Ochlock o nee

River, and a reservoir model for Lake Talquin. The riverine

models take the predicted flows and nutrient concentrations from

the LSPC model and integrate flows and loads from point sources

within the respective watersheds. The WASP output parameters

for both Little River and Ochlockonee Riv er tributaries are then

loaded into the Lake Talquin WASP model.

2 0 . The EFDC model is a hydrodynamic model used to

simulate aquatic systems. For this application, a three -

dimensional EFDC grid was developed for Lake Talquin to provide

hydrodynamic input s to the Lake Talquin WASP water quality

model. The three - dimensional model accounts for lateral and

surface - to - bottom mixing within Lake Talquin.

2 1 . The EFDC model integrates the flows predicted from the

two riverine WASP models with the flows predicte d for the

watershed that directly drains into Lake Talquin. The WASP

models then take the water transport information predicted by

the EFDC model, integrate the loadings from the watershed

directly around L ake Talquin , as well as the two main

tributaries, and predict the nutrient effect in the reservoir

which, in this case, is algal response.

2 2 . These models were applied to simulate nutrient loading

into various areas of Lake Talquin and predict average annual

geometric mean concentrations of chlorophyll a for each year of

the period 2006 through 2012. Those results were then used to

estimate the amount of TN and TP the Lake could assimilate and

still meet the target for chlorophyll a in each year of the

2006 - 2012 simulation period.

V. The Proposed Rule

2 3 . Proposed rule 62 - 304.305(5) would establish a nutrient

TMDL for Lake Talquin:

Lake Talquin. The nutrient TMDL for Lake

Talquin is a seven - year average of annual

loads of 915,783 kg/year TN and 76,585

kg/year TP which are intended to achieve the

appl icable annual geometric mean

chlorophyll a criterion for high color

lakes, and is allocated as follows:

(a) The WLA for wastewater point sources is

divided between the City of Quincy domestic

wastewater facility (NPDES permit FL0029033)

and the City of T allahassee A. B. Hopkins

power plant (NPDES permit FL0025518). The

allocation to the Quincy wastewater facility

for TP is 1,271 kg/year and 4,922 kg/year

for TN. The allocation to the Hopkins power

plant for TP is 2,187 kg/year and 1,020

kg/year for TN,

(b) The WLA for discharges subject to the

departmentÓs NPDES MS4 Permitting Program is

a 27% reduction of TN and a 33% reduction of

TP based on average nutrient loads from the

2006 - 2012 period,

(c) The LA for nonpoint sources is a 27%

reduction of TN and a 33% reduction of TP

based on average loads from the 2006 - 2012

period, and

(d) The Margin of Safety is implicit.

(e) While the LA and MS4 WLA for TN and TP

has been expressed as the percent reduction

needed to attain the applicable Class III

nutri ent criteria, it is the combined

reductions from both anthropogenic point and

nonpoint sources that will result in the

restoration of nutrient conditions in the

impaired waterbody. However, it is not the

intent of this TMDL to abate natural

background con ditions.

2 4 . Although the parties addressed at length the

differences between the original version of the proposed rule

and the final version, only the validity of the final version is

at issue in this proceeding. Differences in the original

version of the rule are only relevant to BASFÓs claim that the

proposed rule is invalid because DEP did not revise the SERC

when the proposed rule was changed. That issue is discussed

later.

VI. Facts Related to BASFÓs Standing

2 5 . The BASF facility discharges was tewater effluent

containing large amounts of TN into Attapulgus Creek, which

flows into the Little River, which flows into Lake Talquin. The

facility is subject to a National Pollut ant Discharge

Elimination System (ÐNPDESÑ) permit issued by the Georgia

En vironmental Protection Division, which includes limits on

BASFÓs discharge of nutrients.

2 6 . Under the proposed rule, the nutrient loading to Lake

Talquin contributed by BASF is addressed in paragraph (c), as

part of Ðnonpoint sources.Ñ The load allocati on (ÐLAÑ) for

nonpoint sources in the proposed rule is a 27 percent reduction

of TN and a 33 percent reduction of TP.

2 7 . The record contains numerous statements acknowledging

that the proposed TMDL will affect Georgia facilities that

discharge nutrients that reach Lake Talquin. For example:

ÐDEP has never denied that to reach this

TMDL goal, that Georgia sources will have to

in some way, form or fashion reduce their

emissions.Ñ Lorraine Novak

Thomas Frick agreed that ÐthereÓs going to

have to be load r eductions from Georgia.Ñ

Ð[T]his TMDL was going to affect how we deal

with permits and issue permits in the State

of Georgia.Ñ Dr. Elizabeth Booth, with the

Georgia Environmental Protection Division.

VII. Whether the Proposed Rule Contravenes the Law I mplemented

A. Reassessment Using the NNC

28 . BASF contends the proposed rule is invalid because

Lake Talquin was not re assessed to determine attainment of the

Numeric Nutrient Criterion (ÐNNCÑ), rule 62 - 302.531, that went

into effect in 2014. BASF asse rts that this action is required

by section 403.067(3)(c) before a TMDL can be established for

Lake Talquin.

29 . DEP divided the State into five basin groups and

focuses its assessment efforts on one basin group each year with

the intent to cover the enti re State over a five - year cycle.

Lake Talquin is in the Group 1 Basin. The waterbodies in Basin

Group 1 are not yet due to be reassessed.

3 0 . If a TMDL is established for Lake Talquin, it will be

used in future assessments to determine whether Lake Talq uin can

be removed from the Verified List of impaired waters, not the

NNC.

B. Annual Loads

3 1 . BASF contends the proposed rule contravenes

section 403.067 because it establishes average annual loads for

TN and TP, not a daily load. BASFÓs argument is ba sed on the

name of the standard -- Total Maximum Daily Load.

3 2 . DEP has adopted over 100 TMDLs for nutrients that

establish maximum annual loads, rather than maximum daily loads.

3 3 . The preponderance of the evidence shows that the

science regarding nutri ents and their effects on flora and fauna

make annual load limitations appropriate.

C. Detailed Waste Load Allocations

3 4 . BASF contends that the proposed rule contravenes

section 403.067(6) by making detailed waste load allocations to

the City of Qui ncy domestic wastewater facility and the City of

Tallahassee A.B. Hopkins P ower P lant. As explained in the

Conclusions of Law, section 403.067(6) allows for detailed waste

load allocations as part of a TMDL .

D. Incomplete Allocation

3 5 . BASF contends th at the proposed rule, by excluding

Georgia nutrient sources, contravenes section 403.067(6)(b),

which requires a TMDL to account for all pollutant sources.

However, t he proposed rule does not state that the maximum loads

of TN and TP exclude nutrient cont ributions from sources in

Georgia. DEPÓs interpretation of the proposed rule as including

all nutrient sources, no matter their origin , is t he most

reasonable interpretation of the words used in the rule .

E. Approved Procedures and Methods

3 6 . BASF cont ends that the proposed rule contravenes

section 403.067(6)(a), which refers to the use of water quality

models Ðusing approved procedures and methods.Ñ Findings

applicable to this issue are set forth below in the section

entitled ÐModel Performance.Ñ Bec ause of the modeling problems

identified in that section, it is found that the Lake Talquin

TMDL modeling was not conducted in accordance with approved

methods and procedures.

VIII . Whether the Proposed Rule is Vague

A. Georgia Sources

3 7 . BASF contends that the proposed rule is vague because

it does not expressly state that the required load reductions

for nonpoint sources applies to both Florida and Georgia

sources.

38 . The term Ðnonpoint sourcesÑ has a contextual meaning

of any nonpoint sources of nu trients that flow to Lake Talquin.

The interpretation of the term by DEP to include Georgia

nonpoint sources is the most reasonable interpretation.

B. Unspecified Loads

39 . The proposed rule requires nonpoint sources to reduce

TN by 27 percent and TP by 33 percent Ðbased on average loads

from the 2006 - 2012 period.Ñ BASF contends this requirement is

vague because the rule does not identify the loads that must be

reduced (27 percent of what?).

4 0 . By subtracting the specific waste load allocations to

the City of Quincy domestic wastewater facility and the City of

Tallahassee A.B. Hopkins P ower P lant, the combin ed load s for

nonpoint sources and the NPDES MS4 Permitting Program can be

determined. However, the proposed rule does not identify the i r

individua l allocations.

4 1 . The proposed rule states that the 27 percent and

33 percent reductions for MS4 sources and nonpoint sources are

to be applied to Ðaverage nutrient loads from the 2006 - 2012

period,Ñ but the rule does not tell us what these average loads

are. It cannot be determined from the proposed rule what

initial nutrient loads DEP has allocated to nonpoint sources and

which DEP will use in the BMAP process.

C. Seven - Year Average of Annual Loads

4 2 . The proposed rule states that the TMDL is a seven - year

average of the annual loads. BASF contends this is vague

because it does not inform persons subject to the rule of what

seven - year period will be used to determine if TMDL is being

achieved.

4 3 . Mr. Gilbert, who had primary responsibility for

devel opment of the proposed rule testified at deposition that it

was his ÐguessÑ that the seven - year average would be a Ðrolling

average, moving forwardÑ but that the decision would be made

after the proposed rule was implemented.

I X. Whether the Proposed Rule is Arbitrary or Capricious

A. The Rulemaking Process

4 4 . BASF contends that the proposed rule is arbitrary and

capricious because DEP did not give reasonable consideration to

BASFÓs comments and objections. The proposed rule was not

capriciously develo ped. The rule development process began in

2013 and included several public workshops and hearings. Many

meetings with interested persons were held, which included BASF

representatives.

4 5 . As explained in the Conclusions of Law, DEPÓs refusal

to adopt BASFÓs recommendations is not evidence of

arbitrariness. Whether the proposed rule is arbitrary depends

on whether it is supported by sound science.

B. The Period 2006 - 2012

4 6 . BASF objects to the 2006 - 2012 time period of data used

for the modeling beca use it includes two unusually dry years,

2007 and 2011. DEP asserts that using 2006 - 2012 data was

appropriate because the data included high, low, and average

rainfall years. One year had above average rainfall (2008),

three years had average rainfall (2 009, 2010, and 2012), and

three years had below - average rainfall (2006, 2007, and 2011).

4 7 . It was an objective of DEP to protect Lake Talquin

during low water flow conditions when nutrients can have a

greater adverse impact on flora and fauna. Therefor e, including

dry years in the modeled period was reasonable.

48 . However, there were problems with the available data

for 2007 and the model performance for 2007, which are discussed

under the heading ÐModel Performance.Ñ

C. Never to be Exceeded

49 . I n a not altogether clear argument, BASF contends the

proposed rule is arbitrary because the TMDL establishes

pollutant levels that are not to be exceeded, but the NNC allows

for one exceedance in three years. The TMDL would be the

nutrient water quality s tandard for Lake Talquin. The NNC would

not apply. The two standards are not intended to be the same.

D. Flow Correction

5 0 . BASFÓs primary criticism of the Lake Talquin modeling

effort centers on a Ðflow correctionÑ used with the EFDC model

to balance inflows and outflows from Lake Talquin. As explained

in EPAÓs Modeling Report for Lake Talquin:

One of the calibration steps in applying

EFDC to reservoirs is to adjust inflows and

outflows to insure that the predict[ed]

water surface elevation of the hy drodynamic

model matches the measured water surface

elevation. This insures that water mass

balance is maintained throughout the

simulation period. Because it is virtually

impossible to define all inflows (tributary,

groundwater, sheet flow) and all outf lows

(dam release, dam seepage, groundwater loss)

EFDC includes a flow balancing utility that

can adjust inflow/outflow to maintain the

measured water surface elevation. The EFDC

flow balancing utility has a limitation of

only adding and removing flows at two

points. Once the adjusted inflow and

outflow time series has been estimated, the

user must go back and apply the adjusted

flows to all of the inflows and outflows.

5 1 . Initially, the EFDC model was not generating surface

water levels for Lake Talqu in that compared well with available

surface elevation gage data. In an attempt to create a better

fit between the model simulation and the surface elevation data,

Mr. Wool used a Ðflow balancing utilityÑ that increased inflows

to and outflows from Lake T alquin.

5 2 . The flow correction affected approximately a quarter

of the modeling period.

5 3 . Gage data is not always accurate. Some gages and some

types of gage measurements are more reliable than others. The

United States Geological Survey (ÐUSGSÑ) ra tes the reliability

of its gages and gives them margins of error that can range from

- 10 percent for gages deemed by USGS to be ÐgoodÑ to

- 20 percent for gages deemed to be Ðpoor.Ñ

5 4 . The preponderance of the evidence supports Mr. WoolÓs

determinat ion that the water surface elevation gage data was

more reliable than the flow gage data.

5 5 . BASF described the flow correction as Ðimaginary

water,Ñ but the purpose of the flow balancing utility is to

simulate real water inflows and outflows. Mr. Wool was seeking

to make the model replicate real inflows and outflows, real

nutrient loadings, real water volumes, and real nutrient

concentrations.

5 6 . BASF claims that ÐignoringÑ some data in order to make

adjustments for a better fit with other data is not

scientifically acceptable. However, such adjustments to a model

are routinely done and are an accepted practice in water quality

modeling. There is nothing scientifically unacceptable about

giving preference to data that is more reliable.

E. Model Perf ormance

5 7 . BASF showed that the model often performed

erratically, making predictions that deviated significantly from

even reliable data. Mr. Erickson persuasively opined that the

model exhibited Ðnumerical instability,Ñ which is a red flag

regarding t he reliability of the model outputs.

58 . Mr. Wool did not effectively explain how the model can

exhibit numerical instability, but at the same time reliably

predict chlorophyll a concentrations .

59 . Furthermore, DEP did not effectively rebut BASFÓs

evide nce that the model performed poorly in low flow conditions.

That is significant because those are the conditions when

nutrient concentrations would be highest and potential adverse

impacts to flora and fauna in Lake Talquin would be greatest.

The TMDL is aimed at protecting Lake Talquin when low flow

conditions occur.

6 0 . It was also shown that gage data for 2007, a dry year

that drove the TMDL development, was not complete.

6 1 . Mr. WoolÓs testimony that the model was Ðvery goodÑ

when looking at the tot ality of its predictions for 2006 - 2012

did not rebut the evidence that the model did not make good

predictions in low flow conditions, nor reliably characterize

conditions for the year 2007.

6 2 . DEP failed to prove that the numerical instability,

poor mod el performance in low flow conditions, and missing data

for 2007 do not materially detract from the reliability of the

model predictions and the TMDL derived from the model.

F. Sensitivity and Uncertainty Analyses

6 3 . BASF contends that the Lake Talquin modeling was

flawed because it did not include a sensitivity analysis or an

uncertainty analysis.

6 4 . Mr. Wool showed that several sensitivity analyses were

conducted during the model development. BASF did not identify

the uncertainty analysis it believe d Mr. Wool should have

conducted.

6 5 . BASFÓs argument in this regard was based in large part

on an EPA guidance document for modeling. The EPA document is

not required to be followed in Florida, and Mr. Wool explained

that much of the guidance in the doc ument is directed to model

development, not model application.

6 6 . The problems with the modelÓs performance, discussed

above, were not due to DEPÓs failure to conduct sensitivity or

uncertainty analyses.

G. Margin of Safety

6 7 . TMDLs are required to in clude a margin of safety,

which can either be implicitly accounted for by choosing

conservative assumptions in the development of the TMDL, or

explicitly accounted for in the allocation process. The

proposed rule states that the Lake Talquin TMDL uses an implicit

margin of safety.

68 . BASF contends the margin of safety in the Lake Talquin

TMDL is arbitrarily large. However, it cannot be determined

what the margin of safety is or whether it is reasonable because

the problems with the modelÓs performance, as found above, make

the TMDL calculation unreliable.

X. Whether DEP Failed to Follow Rulemaking Requirements

69 . DEP prepared a SERC for the proposed rule when it was

originally noticed in December 2016. The SERC estimated the

proposed rule would likel y have direct or indirect regulatory

costs of $5,001,849 per year after implementation.

7 0 . When the proposed rule was changed, DEP did not revise

the SERC to change its estimate of regulatory costs.

7 1 . BASF argues that section 120.541 , Florida Statutes ,

required DEP to consider costs to BASF and other Georgia

entities , and required DEP to revise the SERC when the proposed

rule was changed because the changes increased the costs to BASF

and other Georgia entities.

CONCLUSIONS OF LAW

I. Standing

7 2 . Any person substantially affected by a proposed rule

may seek an administrative determination of the invalidity of

the rule on the ground that the rule is an invalid exercise of

delegated legislative authority. § 120.56(1)(a), Fla. Stat.

(2017).

7 3 . A petit ioner has the burden of proving its standing by

preponderance of the evidence. § 120.56(2)(a), Fla. Stat.

(2017).

7 4 . Generally, to establish standing, a party must show

the challenged agency action will result in a real and immediate

injury in fact and the alleged interest is within the zone of

interest to be protected or regulated. See Jacoby v. Fla. Bd.

of Med. , 917 So. 2d 358, 360 (Fla. 1st DCA 2005).

7 5 . A less demanding test for standing is applicable in

rule challenge cases than in licensing case s. See Fla. DepÓt of

ProfÓl Reg., Bd. of Dentistry v. Fla. Dental Hygienists AssÓn ,

612 So. 2d 646, 651 - 52 (Fla. 1st DCA 1993).

7 6 . The parties stipulated to the City and CountyÓs

standing to intervene in this proceeding. The record shows they

are subst antially affected by the proposed rule. Intervenors

have standing to participate as parties.

7 7 . Ironically, DEP contend s that BASF lacks standing to

challenge the TMDL , even though DEP also contend s that BASFÓs

nutrient discharges are the primary reason Lake Talquin is an

impaired waterbody and needs a nutrient TMDL. DEP argue s that

BASF lacks standing to challenge the proposed rule because DEP

has no authority to enforce a TMDL against BASF. However, the

test for standing to challenge a rule is not wh ether the

promulgating agency can enforce the rule against the challenger,

but whether the challenger is substantially affected by the

rule.

78 . DEP cite s no case with similar facts. The

federal/state regulatory context of the proposed rule is unique.

S ection 403.607 acknowledges that FloridaÓs TMDL program is

required by the federal Clean Water Act. The Clean Water Act is

cited three times in section 403.067, showing clearly that

FloridaÓs TMDL program is intended to fit into a Clean Water Act

regulato ry structure. In this structure, Florida, Georgia, and

the Federal Government (EPA) all have formal roles in the

protection of Lake Talquin.

79 . FloridaÓs role is to establish water quality standards

for Lake Talquin. The proposed TMDL is a water qualit y

standard. GeorgiaÓs role is to issue an NPDES permit to BASF

with conditions that will assure the BASF discharges do not

cause the TMDL to be exceeded. Federal regulations prohibit a

state from issuing a NPDES permit to a facility if Ðthe

imposition of conditions cannot ensure compliance with the

applicable water quality requirements of all affected States.Ñ

40 C.F.R. § 122.4(d). By adopting the Lake Talquin TMDL, DEP is

establishing the water quality standard that governs GeorgiaÓs

regulation of BASF Ós discharge of nutrients.

8 0 . In any future regulatory context in which BASF is

directed by Georgia or EPA to take action as a consequence of

FloridaÓs nutrient TMDL for Lake Talquin, the issue will not be

whether the TMDL is valid, but whether the requi red action to

comply with the TMDL is reasonable. This rule challenge

proceeding is the only proceeding available to BASF to prevent

Florida from adopting the TMDL that Georgia and the EPA will

enforce.

8 1 . Although DEP contends it cannot enforce the TMD L

against BASF, the State of Florida can seek an order from EPA

requiring Georgia to impose such conditions on BASFÓs NPDES

permit as will prevent the TMDL from being exceeded. See

33 U.S.C. § 1319. Florida can also file a lawsuit against the

EPA to enfo rce any provision of the Clean Water Act. See

33 U.S.C. § 1365(a).

8 2 . These circumstances are unique, but FloridaÓs courts

have recognized standing when a proposed rule will have a

collateral regulatory effect, even when the adopting agency has

no direc t regulatory authority over the challenger. For

example, in Televisual Communications, Inc. v. Department of

Labor & Employment Security , 667 So. 2d 372 (Fla. 1st DCA 1995),

the court held that a publisher of educational video programs

had standing to cha llenge a proposed rule regulating physicians,

even though the agency had no regulatory authority over the

publisher, because the rule had the collateral effect of

regulating the publisher by precluding the sale of its videos.

8 3 . BASF is substantially aff ected by the proposed rule

and, therefore, has standing to challenge it.

II. General Rule Challenge Principles

8 4 . A person challenging a proposed rule must state with

particularity the reasons that the proposed rule is an invalid

exercise of delegated l egislative authority. § 120.56(2), Fla.

Stat. (2017). The petitioner has the burden of going forward

with evidence to support the allegations in the petition. Id.

If the challenger meets this burden, the burden of persuasion

shifts to the agency to pro ve by a preponderance of the evidence

that the proposed rule is not an invalid exercise of delegated

legislative authority "as to the objections raised." Id.

8 5 . A proposed rule is not presumed to be valid or

invalid. § 120.56(2)(c), Fla. Stat. (2017).

8 6 . The validity of a rule does not turn on whether it

represents the best means to accomplish the agency's purposes.

See Bd. of Trs. of Int. Impust Fund v. Levy , 656 So. 2d

1359, 1364 (Fla. 1st DCA 1995).

III. Whether the Proposed Rule Contravenes t he Law Implemented

8 7 . A proposed rule is an invalid exercise of delegated

legislative authority under section 120.52(8)(c) if it enlarges,

modifies, or contravenes the specific provisions of law

implemented. The question to be determined is whether the rule

gives effect to a specific law and whether the rule implements

or interprets the lawÓs specific powers and duties. Bd. of Trs.

of Int. Imp. Trust Fund v. Day Cruise AssÓn , 794 So. 2d 696, 704

(Fla. 1st DCA 2001).

A. Reassessment Using the NNC

88 . S ection 403.067(3)(c) , which pertains to the listing

of impaired waters , provides as follows :

If the department has adopted a rule

establishing a numerical criterion for a

particular pollutant, a narrative or

biological criterion may not be the basis

for d etermining an impairment in connection

with that pollutant unless the department

identifies specific factors as to why the

numerical criterion is not adequate to

protect water quality. If water quality

nonattainment is based on narrative or

biological cri teria, the specific factors

concerning particular pollutants shall be

identified prior to a total maximum daily

load being developed for those criteria for

that surface water or surface water segment.

89 . BASF argues that this section requires DEP to rea ssess

Lake Talquin to determine whether it is attaining the NNC as a

prerequisite for establishing a TMDL for Lake Talquin. BASFÓs

interpretation is strained. DEP complied with section

403.067(3) because, when it assessed Lake Talquin, the NNC was

not in effect.

9 0 . Nothing in the plain wording of section 403.067(6),

which pertains to TMDL development, requires DEP to reassess a

waterbody on the Verified List before developing a TMDL. The

statute allows DEP to proceed to develop a TMDL for any

waterbody on the Verified List .

B. Annual Loads

9 1 . Although the word ÐdailyÑ is a part of the term

ÐTMDL , Ñ the term is not defined as a daily standard. It is

defined in section 403.031(6) as follows:

ÐTotal maximum daily loadÑ is defined as the

sum of the indiv idual wasteload allocations

for point sources and the load allocations

for nonpoint sources and natural background.

Prior to determining individual waste load

allocations and load allocations, the

maximum amount of a pollutant that a water

body or water se gment can assimilate from

all sources without exceeding water quality

standards must first be calculated.

9 2 . No provision of section 403.067 limits a TMDL to a

daily standard.

9 3 . The Legislature has approved non - daily TMDLs for Lake

Okeechobee, the Ca loosahatchee Estuary, and the St. Lucie River

and Estuary. See § 373.4595(1)(f) - (h), (l), (2)(q), (3) - (4),

Fla. Stat. (2017). For example, section 373.4595 states Ð[t]he

Legislature finds that the Lake Okeechobee phosphorus loads set

forth in the total m aximum daily loads established in accordance

with s. 403.067 represent an appropriate basis for restoration

of the Lake Okeechobee watershed.Ñ The Lake Okeechobee TMDL for

phosphorus is an annual load standard. See Fla. Admin. C ode R .

62.304 - 700.

C. Det ailed Waste Load Allocation

9 4 . BASF contends the proposed rule contravenes section

403.067(6) by making two detailed waste load allocations.

However, section 403.067(6)(b) clearly states, Ð [a] llocations

may be made to individual basins and sources .Ñ

D. Incomplete Allocation

9 5 . BASF contends that the proposed rule, by excluding

Georgia nutrient sources, contravenes section 403.067(6)(b),

which states the TMDL is intended to establish a maximum loading

that will provide for the attainment of water quali ty standards.

However, the most reasonable interpretation of the words used in

the rule is that it applies to all nutrient sources, including

Georgia sources.

E. Approved Procedures and Methods

9 6 . Section 403.067(6)(a) authorizes DEP to develop TMDLs

with Ðmathematical modeling using approved procedures and

methods.Ñ Because of the modeling problems found above, the

TMDL modeling for Lake Talquin was not in accordance with

approved methods and procedures.

F . Summary

9 7 . The proposed rule contravenes section 403.067(6)(a)

because the TMDL modeling was not consistent with approved

methods and procedures.

IV. Whether the Proposed Rule is Vague

98 . FloridaÓs courts have described an invalidly vague

rule as one that forbids or requires the performance o f an act

Ðin terms that are so vague that persons of common intelligence

must guess at its meaning and differ as to its application.Ñ

See e.g. , Bouters v. State , 659 So. 2d 235, 238 (Fla. 1995).

99 . In ascertaining whether a proposed rule is vague, the

r ule must be examined on its face. Issues concerning the

application of the rule are beyond the scope of a rule challenge

proceeding. See Fairfield Communities v. Fla. Land & Water

Adjudicatory CommÓn , 522 So. 2d 1012, 1014 (Fla. 1st DCA 1988 ) .

A. Georg ia Sources

10 0 . BASF contends the proposed rule is vague because it

does not expressly state that the load reductions for nonpoint

sources include Georgia sources. DEPÓs interpretation of the

term to include Georgia nonpoint sources is the most reasonabl e

interpretation.

B. Unspecified Loads

10 1 . BASF contends the requirement in the proposed rule to

achieve percentage reductions in TN and TP loads for nonpoint

sources is vague because the rule does not identify the loads

that must be reduced. BASF is c orrect that it cannot be

determined from the proposed rule what initial nutrient loads

have been allocated to nonpoint sources by DEP and will used by

DEP in the BMAP process.

C. Seven - Year Average of Annual Loads

10 2 . BASF contends the reference in the proposed rule to a

Ðseven - year average of the annual loadsÑ is vague because it

does not identify the seven - year period that is to be used. DEP

did not show that the term has a specific meaning.

10 3 . DEP states that there is no confusion in this regard,

citing rule 62 - 303.720(2)(k). That rule states that a waterbody

can be removed from the Verified List when the water attains a

nutrient criterion for three consecutive years. However, this

rule does not address how attainment is determined. Because the

TMDL is expressed as a seven - year average annual load, knowing

what this means is necessary to know what this water quality

standard is.

D. Summary

10 4 . The proposed rule is vague because it does not

specify the loads which the percentage reductions appl y to, and

because it does not identify the seven - year period that

determines attainment of the TMDL.

V. Whether the Proposed Rule is Arbitrary or Capricious

10 5 . ÐA rule is arbitrary if it is not supported by logic

or the necessary facts; a rule is capri cious if it is adopted

without thought or reason or is irrational.Ñ £ 120.52(8)(f),

Fla. Stat. (2017).

10 6 . If an agency rule Ðis justifiable under any analysis

that a reasonable person would use to reach a decision of

similar importance, it would seem t hat the decision is neither

arbitrary nor capricious.Ñ Dravo Basic Materials Co., Inc. v.

State. DepÓt of Trans. , 602 So. 2d 632, 634 n.3 (Fla. 2d DCA

1992).

A. The Rulemaking Process

1 0 7 . The rulemaking process was extensive and included

general public meetings and stakeholder meetings. The proposed

rule was not capriciously developed.

1 0 8 . Neither the definition of ÐarbitraryÑ in section

120.52(8)(e), nor the case law give a meaning to ÐarbitraryÑ

that would include failing to adopt the recommendatio ns of

interested persons, even when those recommendations are good

ones. If an agency proposes a rule based on unsound science,

the rule is invalid, whether or not a better alternative was

recommended to the agency.

B. The Period 2006 - 2012

1 09 . It was r easonable for DEP to use a time period which

included dry years. However, because the data for the year 2007

were incomplete, and the data were critical to the calculation

of the TMDL, it caused the reliability of the model results to

be compromised.

C. Never to be Exceeded

11 0 . BASF argues that the proposed rule is invalid because

the TMDL establishes a never - to - exceed standard, rather than a

one - exceedance - in - three - years standard like the NNC. However,

the two standards are separate and not intended t o be the same.

D. Flow Correction

11 1 . Applying a flow correction in the TMDL Model and

giving preference to water surface elevation gage data over

inflow/outflow gage , was reasonable and consistent with accepted

modeling practices.

E. Model Performanc e

11 2 . Section 403.067(6)(a) authorizes DEP to use water

quality modeling in the development of TMDLs, but requires that

such modeling use Ðapproved procedures and methods.Ñ

11 3 . The preponderance of the record evidence shows the

model exhibited numerica l instability and poor model performance

for critical low flow conditions, making the TMDL unreliable.

Therefore, DEP should not have relied on the modelÓs outputs to

establish the TMDL.

F. Sensitivity and Uncertainty Analys e s

11 4 . BASF contends that th e Lake Talquin modeling was

flawed because it did not include a sensitivity analysis or an

uncertainty analysis, as an EPA guidance document suggested.

BASF failed to prove t his contention.

G. Margin of Safety

11 5 . Section 403.067(6)(a) requires a TMDL to Ðinclude a

margin of safety that takes into account any lack of knowledge

concerning the relationship between effluent limitations and

water quality.Ñ

1 1 6 . It is reasonable for DEP to use conservative

assumptions to derive the Lake Talquin TMDL. Howev er, because

the TMDL calculation was found to be unreliable, the margin of

safety cannot be determined.

H. Summary

11 7 . The proposed rule is arbitrary because it is based on

unreliable modeling results.

VI. Whether DEP Failed to Follow Rulemaking Requi rements

1 18 . S ection 120.541(2)(c) requires a SERC to include the

following:

A good faith estimate of the cost to the

agency, and to any other state and local

government entities, of implementing and

enforcing the proposed rule, and any

anticipated effect on state or local

revenues.

1 1 9 . BASFÓs interpretation of this wording as requiring

agencies to consider costs to persons outside of Florida when

preparing a SERC is plainly wrong. The statute is clearly

refer ring to other Florida state entities and ot her Florida

local governments. Similarly, the reference in the statute to

the effect on state or local revenues does not mean the revenues

of other states.

12 0 . Determining that BASF has standing to challenge the

rule because it is substantially affected by the proposed rule

does not require that costs to BASF be accounted for in the

SERC. Legal standing and the SERC requirement serve separate

and distinct purposes.

121. There is no reason to believe that the Florida

Legislature, in requiring agencies t o consider the regulatory

costs of a proposed rule on economic growth, job creation,

employment, and business competitiveness, was seeking to protect

not only Florida and Florida businesses, but was also seeking to

protect economic growth, job creation, em ployment, and business

competitiveness in other states.

12 2 . It is concluded, therefore, that DEP did not have to

estimate the regulatory costs to BASF or other Georgia entities

when it prepared the SERC.

CONCLUSION

Based on the foregoing Findings of Fac t and Conclusions of

Law, it is determined that DEP did not meet its burden to prove

that proposed rule 62 - 304.305(5) is not an invalid exercise of

delegated legislative authority.

DONE AND ORDERED this 2nd day of March , 2018 , in

Tallahassee, Leon County, Florida.

S

BRAM D. E. CANTER

Administrative Law Judge

Division of Administrative Hearings

The DeSoto Building

1230 Apalachee Parkway

Tallahassee, Florida 32399 - 3060

(850) 488 - 9675

Fax Filing (850) 921 - 6847

www.doah.state .fl.us

Filed with the Clerk of the

Division of Administrative Hearings

this 2nd day of March, 2018.

COPIES FURNISHED:

Lorraine Marie Novak, Esquire

Department of Environmental Protection

Mail Station 35

3900 Commonwealth Boulevard

Tallahassee, Florida 32399

(eServed)

Martha Harrell Chumbler, Esquire

Carlton Fields Jorden Burt, P.A.

Post Office Drawer 190

Tallahassee, Florida 32302

(eServed)

James E. Parker - Flynn, Esquire

Carlton Fields Jorden Burt, P.A.

Post Office Drawer 190

Tallahassee, Florida 3 2302

(eServed)

Carson Zimmer, Esquire

Department of Environmental Protection

Mail Station 35

3900 Commonwealth Boulevard

Tallahassee, Florida 32399

(eServed)

Herbert W. A. Thiele, Esquire

Leon County Attorney's Office

Room 202

301 South Monroe Street

Ta llahassee, Florida 32301

(eServed)

Edwin A. Steinmeyer, Esquire

Steinmeyer Fiveash LLP

310 West College Avenue

Tallahassee, Florida 32301

(eServed)

William B. Graham, Esquire

Kevin P. Blodgett, Esquire

Carr Allison

305 South Gadsden Street

Tallaha ssee, Florida 32301

(eServed)

David W. Childs, Esquire

Hopping Green and Sams

119 South Monroe Street, Suite 300

Tallahassee, Florida 32301

(eServed)

Gary V. Perko, Esquire

Hopping Green & Sams, P.A.

Post Office Box 6526

Tallahassee, Florida 32314

(eS erved)

Malcolm N. Means, Esquire

Hopping Green & Sams, P.A.

Post Office Box 6526

Tallahassee, Florida 32314

(eServed)

Kenneth B. Hayman, Esquire

Department of Environmental Protection

Mail Stop 35

3900 Commonwealth Boulevard

Tallahassee, Florida 32399 - 3000

(eServed)

William L. Anderson, Esquire

Crowell & Moring, LLP

1001 Pennsylvania Avenue Northwest

Washington, DC 20004

Ken Plante, Coordinator

Joint Admin istrative Proced ures Committee

Room 680, Pepper Building

111 West Madison Street

Tallahassee, F lorida 32399 - 1400

(eServed)

Ernest Reddick, Program Administrator

Department of State

R.A. Gray Building

500 South Bronough Street

Tallahassee, Florida 32399 - 0250

(eServed)

Robert A. Williams, General Counsel

Department of Environmental Protection

Legal Department, Suite 1051 - J

Douglas Building, Mail Station 35

3900 Commonwealth Boulevard

Tallahassee, Florida 32399 - 3000

(eServed)

Lea Crandall, Agency Clerk

Department of Environmental Protection

Douglas Building, Mail Station 35

3900 Commonwealth B oulevard

Tallahassee, Florida 32399 - 3000

(eServed)

Noah Valenstein, Secretary

Department of Environmental Protection

Douglas Building

3900 Commonwealth Boulevard

Tallahassee, Florida 32399 - 3000

(eServed)

NOTICE OF RIGHT TO JUDICIAL REVIEW

A party who is adversely affected by this Final Order is

entitled to judicial review pursuant to section 120.68, Florida

Statutes. Review proceedings are governed by the Florida Rules

of Appellate Procedure. Such proceedings are commenced by

filing the original not ice of administrative appeal with the

agency clerk of the Division of Administrative Hearings within

30 days of rendition of the order to be reviewed, and a copy of

the notice, accompanied by any filing fees prescribed by law,

with the clerk of the Distric t Court of Appeal in the appellate

district where the agency maintains its headquarters or where a

party resides or as otherwise provided by law.