"1 June 1998 Bulletin (Tech.) 906

Calculating Nutrient Loads

Forrest T. Izuno and Ronald W. Rice

UNIVERSITY OF
)FLORIDA
Agricultural Experiment Station
)OC Institute of Food and Agricultural Sciences
00
'636b
106

Calculating Nutrient Loads

Forrest T. Izuno and Ronald W. Rice

Introduction

A chemical "load" is simply defined as a mass of a chemical element or chemical

compound, being moved from one location to another. Plant nutrients are chemicals

that support and sustain plant growth and development. Although nutrients are

beneficial to agricultural crops, they can be deleterious to ecologically sensitive areas if

present in quantities that would adversely affect the growth of native plant or animal

species. Nutrients can be translocated via aeolian (wind-bome), water, or human

activity. For example, in the case of the Everglades Agricultural Area (EAA), any

particular farm incurs chemical loads to the farm from aeolian deposits of particulate

matter, particulate and dissolved fractions of a chemical in rainfall, fertilization of fields,

particulate and dissolved species of chemicals introduced in irrigation water, and

seepage. Chemical loads leave the farm via aeolian transport, harvested crop matter,

drainage water, and seepage. A nutrient load can be represented as either a mass of

the chemical being translocated to or from a particular piece of land, or as a mass of

chemical per unit area relative to either its destination or its origin.

Chapter 40E-63 (SFWMD, 1992) of the Everglades Forever Act (1994) requires

1

UNIVERSITY Q*r WLLVfAILJ^

that the phosphorus (P) load leaving the EAA must be reduced by 25% relative to the

baseline years of 1979 through 1988. This basin-level regulatory target is to be

achieved by EAA growers through their collective implementation of "best management

practices" (BMPs) designed specifically to reduce P loads discharged from their farms.

Yet, the Everglades Forever Act specifically states that a threshold P concentration

level will be set for waters entering the Water Conservation Areas from the EAA.

Hence, the focus on loads in 40E-63 appears to be somewhat puzzling. However, the

hydrology and hydraulics of the region are such that drainage water cannot leave the

EAA (except via seepage which is traditionally small due to the higher water levels

maintained around the EAA borders) unless enabled by man. Additionally, the EAA

receives more rainfall in a year than can leave through evapotranspiration (ET). Hence,

if growers and water managers ceased pumping activities, the EAA would remain

flooded for much of the year. Therefore, there are naturally imposed limits on how

much water can and cannot be pumped during a year, enabling water managers to

make reasonable estimates of concentrations given loads, and vice versa.

Reducing loads can be accomplished simply by reducing the amount of water

pumped. Hence, if there was not the propensity for the EAA basin to flood (geologic

bedrock structure and high water table), one could conceivably stop pumping and let

water percolate through the root zone, effectively eliminating surface water P loading.

Since this is not the case, there is a natural limit as to how much pumping can be

reduced relative to historical patterns. Now, if one takes the maximum ET rate for the

area and compares it to rainfall inputs, the amount of "excess" water can be

2

determined. Then, comparing this volume with what was historically pumped for

drainage, one could actually determine what the maximum reduction in drainage could

be in the EAA without reversion to a swamp. For example, if rainfall is assumed to be

about 53 inches per year and the maximum ET is assumed to be around 45 inches per

year, 8 inches of water would have to be removed to avoid accruing water in the basin.

If historical drainage from the area was 11 inches per year, then drainage under BMPs

can be reduced by no more than 3 inches (about 27%). Using this information, a load

reduction can be translated into a concentration reduction (mass of P divided by the

estimated volume of water pumped under maximum retention). In fact, although 40E-

63 requires the achievement of a 25% P load reduction effected by BMP

.implementation, parallel calculations were occurring during Rule development to

determine what that would mean in terms of concentrations assuming different levels of

off-farm pumping reduction. An additional clause in the Rule states that water retention

practices used to alleviate P loading could be revisited if water supply were to become

a problem.

As illustrated above, it is necessary to understand the concepts of nutrient

concentrations, nutrient loads, and the relationship between the two, in order to

understand how to effectively achieve a reduction in the amount of P leaving the EAA.

Additionally, there are many ways of expressing concentration and load values in terms

of units. Many of these methods are being used, often interchangeably, in research,

monitoring, and regulatory presentations and documents. These units of reporting shall

be explained herein.

3

Concentrations, Volumes, and Loads: Units of Measure

Concentrations, volumes, and loads are all expressed in a myriad of ways.

These include traditional, single dimensioned, and dimensionless units. To further

confuse matters, English and metric units are, at times, equivalent and are often seen

mixed in a single equation, figure, or table.

A nutrienrload is a mass, or weight, of a chemical entering or leaving an area,

and is the product of the volume of water that the chemical is using as its transport

medium and the concentration of the chemical in the water. Hence, it is necessary to

understand how both concentrations and volumes are expressed in order to perform

the appropriate calculations to obtain loads. The actual procedures used to measure

flow volumes and chemical concentrations in water are beyond the scope of this

publication.

Volume

Water volume is expressed in many ways. Some of the more common terms are

listed below along with their definitions and necessary explanations:

1) Acre-foot (ac-ft) The volume of water represented by one foot (ft) of water
uniformly covering an acre (ac) of land. Since an ac is
43,560 square feet (ft2), an ac-ft is 43,560 cubic feet (ft3)
of water. (325,872 gallons (gal)/ac-ft or
1,233,426 liters (L)/ac-ft).

2) Acre-inch (ac-in) The volume of water represented by an inch (in) of water
uniformly covering an ac of land. Since an ac is
43,560 ft2, an ac-in is 3,630 ft3 (43,560 ft2 x 1 in/12 ft)
of water. (27,156 gal/ac-in or 102,785 L/ac-in).

4

3) Inch or Foot (in or ft) At times, a water volume is expressed as a single
dimensioned quantity. Implicit to the use of these
dimensions is a common or unit area. For example,
"the average rainfall in the EAA is around 52 in/yr"
implies that 52 in of water falls uniformly across the
EAA. Hence the actual volume of rainfall in ft3 would
be (52 in)*(1 ft/12 in)*(the EAA area in ft2). Likewise,
expressing drainage as a single dimensioned value
(i.e. "farms in the EAA drain about 11 in of water
annually") is actually a volume of water (in3) divided by
the area (in2) being referred to.

4) Cubic foot (ft3) Arguably the most commonly used English unit dimension for
expressing large volumes of water, equivalent to the amount
of water that would fit in.a cube with 1 ft long sides.
(7.481 gal/ft3 or 28.31559 L/ft3).

5) Gallon (gal) Arguably the most common everyday representation of a liquid
volume. There are 7.481 gal of water in 1 ft3. (3.785 L/gal).

6) Liter (L) Arguably the most common everyday metric unit representation of
a liquid volume. There are 1000 milliliters (mL) of water in a L.
(0.2642 gal/L or 0.03532 ft3/L).

7) Cubic meter (m3) Arguably the most common metric unit representation of
large liquid volumes. A cubic meter (m3) of water is the
volume of water that will fit in a cube with sides 1 m in
length. There are 1000 L in a m3.

8) Cubic centimeters (cm3 or cc) A metric unit representation of small liquid
volumes, equivalent to the amount of water
that will fit in a cube with sides 1 cm in length.
There are 1000 cm3 in a L.

9) Milliliter (mL) A metric unit volume equal to 1/1000 of a liter. A mL is
a volume equivalent to a cm3.

10) Centimeter or Meter (cm or m) The metric units of length used to express
a volume of water as is done with the
English units in and ft.

5

Mass or Weight

1) Pound (Ib) The most common English unit for mass or weight.
(0.4536 kg/lb). There are 2000 Ib in a ton. A gal of water is
generally estimated to weigh 8.33 lb.

2) Kilogram (kg) Arguably the most common metric unit representation of
mass or weight. There are 1000 kg in a metric ton (MT) and
1000 grams in a kg. A kg is equivalent to approximately
2.2046 Ib.

3) Gram (g) A metric unit of mass equal to 1/1000 of a kg. There are
approximately 453.6 g in a lb.

4) Milligram (mg) A metric unit of mass equal to 1/1000 of a gram or
1/1,000,000 of a kg.

5) Microgram (pLg) A metric unit of mass equal to 1/1,000 of mg,
1/1,000,000 of a g, or 1/1,000,000,000 of a kg.

Concentration

1) Milligram per liter (mg/L) The most common metric unit describing a mass
of chemical present in a L of water (mass/volume).
A mg/L is equivalent to the English unit "parts per
million" (ppm) when the assumption that 1 mL of
water weighs 1 g (density of water is 1 g/mL, cc, or
cm3) holds true. Therefore, 1 mg/L is
the same as saying a mg of chemical per 1,000
mg of water (mass/mass). Small variations in the
density of water occur due to atmospheric
pressures, temperatures, and water purity.
However, for general purposes the variations
cause inconsequential differences.

2) Microgram per liter (,.g/L) A metric concentration unit equivalent to
1/1,000 of a mg/L. A /g/L is equivalent to the
English unit "parts per billion" (ppb) subject to the
assumptions described above. A ug/L is a
mass/volume unit, and is numerically equivalent
to a mass/mass unit since the density of water in
metric units is one g/cm3.

6

3) Parts per million (ppm) A dimensionless English unit expression of the mass
of a chemical in a mass of water. For example,
the concentration of a chemical in water may be
expressed as 1 Ib of chemical in 1,000,000 Ib of
water. Here again, the density of water is assumed
to be a constant although it changes slightly with
changes in atmospheric pressure, temperature, and
water purity. A ppm is equal to a mg/L assuming
that the density of water is a constant 1g/cm3 A
concentration of 1 ppm phosphorus is equivalent to
1 Ib of phosphorus in approximately 120,048 gal of
water (1 gal of water = 8.33 Ib).

4) Parts per billion (ppb) A dimensionless English unit expression of the mass
of a chemical in a mass of water. For example,
the concentration of a chemical in water may be
expressed as 1 Ib of a chemical in 1,000,000,000 Ib
of water. Here again, the density of water is assumed
to be constant. A ppb is equal to a Ag/L assuming
that the density of water is a constant 1g/cm3. A
concentration of 1 ppb is 1/1,000 of a ppm.

Dispelling the Magic: 1 ppm (English units, mass)
= 1 mg/L (metric units, mass/volume)

The concentration units ppm (English units, mass/mass) and mg/L (metric units,

mass/volume) are perhaps the most widely used interchangeable units in water quality

research and monitoring. In fact, one often sees load calculated as a concentration

expressed in ppm multiplied by a volume expressed in L to attain a load in mg, g, or kg

(mass). This interchangeable use of units receives little derisive comment, in spite of

the fact that the user is apparently mixing English and metric units, as well as mass and

volume expressions. Although the scientific community views mixed units with either

disdain or amusement, the calculated values are numerically identical since 1 ppm =

7

1mg/L. To illustrate this fact, assume that there is 1 lb of P in 1,000,000 Ib of water

(mass/mass). One Ib of P = 0.4536 kg of P, applying the appropriate conversion factor

from Ib to kgs discussed above. To convert 0.4536 kg to mg, the multiplication factor of

1,000,000 mg/kg (lx106 mg/kg) is used, yielding 4.536x105 mg P in 1,000,000 Ib of

water. Now, addressing the 1,000,000 Ib of water, dividing by 2.2046 Ib/kg yields

4.536x105 kg of water. Multiplying by 1,000 yields 4.536x108 g of water. Since 1 mL of

water weighs 1 g, 4.5360x108 g of water = 4.5360x108 mL of water. Dividing 4.536x108

mL of water by 1,000 to convert from mL to L yields 4.536x105 L of water. Hence, we

now have 4.536x105 mg P in 4.536x105 L of water. Dividing both by 4.536x105 to

reduce the fraction to its lowest common denominator yields 1 mg P in 1 L of water (1

mg/L) having started the exercise with 1 Ib P in 1,000,000 Ib of water, or 1 ppm.

Typical Pump Flow Rate and Nutrient Concentration Distributions

Although P has been the nutrient of concern in much of the debate regarding the

mitigation/remediation efforts relative to potential negative impacts of agricultural

drainage water quality on the Everglades ecosystem, other nutrients, chemical

elements, and chemical compounds are also receiving research and monitoring

attention. The load calculations, and associated pump flow and concentrations to be

discussed can apply to any nutrient or chemical species. However, for the purpose of

illustrating the load calculation methods, P will be used as the example nutrient, water

will be the transport medium, and the geographic setting will be the EAA.

The basic load equation is:

8

Nutrient Load (mass) = Concentration (mass/volume or mass/mass) (1)
Flow (volume or mass).

The calculation appears to be simple enough, but it can become troublesome given the

range of acceptable protocols for measuring flows and collecting water samples, and

the wide range of data collection frequencies. Applying standard and uniform data

management techniques to data sets is extremely important, especially when

combining hundreds of load measurements into a single database.

Figure 1 shows a typical pumped flow rate versus time curve. Figure 2a depicts

a typical P concentration versus time curve. The two curves show how each parameter

in the load equation (Equation 1) can change during a drainage event. These

distributions can be seen in the EAA for P load monitoring when water samples and

pump flow rates are collected or monitored continuously, or on short discrete time

intervals, throughout the drainage event. These distributions are perhaps the most

characteristic of flows and P concentrations leaving farms in the EAA during pumped

drainage events in response to rainfall.

The flow rate curve in Figure 1 shows a slowly declining flow rate over time as

hydraulic heads in the system increase. Over time, as the EAA main canal levels rise,

and farm canal levels drop, pumps must work harder to lift water from a falling water

level within the intake sump (on-farm) and discharge it to a sump or canal (off-farm)

where water levels are rising. As the discharge event progresses, this hydraulic head

that the pump must work against increases and pump efficiencies decline, yielding an

accelerated decrease in flow rate towards the end of the pumping event. The

9

concentration curve in Figure 2a shows an initially high P concentration, typical of a

"first-flush" type event, where large amounts of P-bearing particulate matter and

sediments near the pump intake are initially discharged. As the event continues,

concentrations decline as rain water and open canal/ditch water P species dominate

the P concentration characteristics of the water in the drainage stream.

Another typical P concentration distribution is shown in Figure 2b. In this case,

the drainage water concentration starts low, indicative of a situation where there is little

particulate matter transport or channel bottom sediment scouring when the pump starts

up. Farm canal water dissolved-P concentrations could also be relatively low. Events

that could cause this phenomenon are dilution of the farm canal P concentrations due

to initially heavy rainfall, antecedent pumping where much of the particulate matter or

sediment near the pump had already been moved during the preceding pumping event,

area main canal leakage through the pump station into the farm canal, seepage into the

farm canal through the fractured bedrock, irrigation occurring just prior to drainage

pumping, and/or a slow initial pump speed. As the event progresses, P concentrations

rise. This could be attributed to the transport of particulate matter and sediments from

the downstream reaches of the farm canal, lower water levels accompanied by higher

flow velocities and canal bottom scouring, increased pump speed, and/or the addition to

the drainage stream of rain water which has fallen on the field surfaces and passed

through, or over, the P-rich upper soil layer carrying mobile particulates and dissolved-

P.

Figure 2c shows another typical P distribution found in the EAA. In this case,

concentrations start low, peak sometime during the event, and then fall off later in the

10

event. This distribution can generally be attributed to a different time distribution of the

occurrences discussed above that cause the distributions shown in Figures 1 and 2a.

For additional information on water sampling strategies and load calculations

applicable to the EAA, the reader is referred to Izuno, Bottcher, and Davis (1992), Izuno

et al. (1996), Izuno and Rice, eds. (1997), and Rice and Izuno (1997).

Calculating Nutrient Loads For Typical Data Collection Protocols

Beginning and End of Event Sampling

Figure 3 illustrates data derived from one of the most rudimentary sampling

protocols acceptable for nutrient load determinations. In this case, water samples are

collected at the beginning and end of the drainage event. Flow rates are also

calculated only for the beginning and end of the event, usually by collecting hydraulic

head and pump rpm data when turning the pump(s) on and off and applying a pump

calibration equation.

To calculate the nutrient load using these available data, the beginning and end

concentrations and flow rates can simply be averaged. The average flow rate is then

multiplied by the length of time that the pump ran (pumping duration) to determine the

total volume of water discharged. The equation to apply to calculate load with these

data, using the variable names assigned in Figure 3, is:

Load = {(C1 + C2)/2} {(a + b)/2 Time} (2)

11

where C, and C2 are the beginning and end of event concentrations, respectively, a

and b are the beginning and end of event flow rates, respectively, and Time is the

pumping duration.

Looking back at Figures 1 and 2, one can see that this equation would yield a

fairly representative load for the event, except when the concentration distribution

depicted in Figure 2b applies. Flow rates in the EAA are generally flat over the normal

pump operating range. Hence, an event average flow rate calculated as in Equation 2

is probably sufficient unless head differences across the pump change greatly during

the event, or the pump is not installed to operate under its recommended efficient

operating range.

Time Discrete Water Sampling With Continuous Flow Monitoring

At times it is desirable to know the distribution characteristics of drainage water P

concentrations over time for the purpose of identifying factors which cause the

distribution to appear as it does. In this case, time discrete water sampling (collecting

water samples on predetermined, closely spaced, time intervals) is useful.

Concentration data typical of the above water sampling protocol are shown in Figure

4a. Figure 4b again shows the typical continuously monitored flow rate versus time

curve.

In Figure 4a, it is evident that drainage water sample concentrations have been

determined on two-hour intervals. Load calculations are still relatively simple.

However, one must consider several calculation options, select one, and apply the

selected option consistently. The primary determination that one must make is "what

12

volume of pumped water corresponds to each concentration value". Options are: 1)

Use the two-hour flow time period following the concentration time; 2) Use the two-hour

flow time period preceding the concentration time; 3) Average two adjacent

concentrations to obtain an average concentration for each two-hour flow period; or 4)

Average the flow rates measured an hour before and an hour after the concentration

was measured. For time discrete water sampling (Figure 4a), where a water sample

represents a point-in-time measurement, option 4 is the most desirable. The

representative flow rate is then multiplied by the time interval to determine flow volume,

and load is then calculated.

In this example, it is assumed that each concentration data point is a time

discrete measurement representing the drainage stream concentration at a point-in-

time. Hence, one must infer that the concentration distribution is a smooth curve

(Figure 4a). The load calculation option of choice is to select the flow rates at

appropriate times and use them to calculate the flow volume. Illustrating this example,

Figure 4a shows a concentration C at time=14 hours. Water samples were collected

every two hours. Hence, the sample concentration should be assumed to apply to the

volume of water pumped an hour before and an hour after the time that the water

sample was collected (half the time step before plus half the time step after the time

that the water sample was collected). In Figure 4b, the flow rates at time=13 hours and

time=15 hours are different, and should be averaged and multiplied by two hours to

calculate the flow volume for concentration C. This flow volume is represented by the

shaded area in Figure 4b. Since there are 10 discrete water sample concentrations,

the total event load would then be the summation of all the incremental loads

13

associated with each two-hour time period as written in Equation 3:

Load = Eloadi, i = 0 to 10, (3)

where loado is calculated using the flow volume that occurred during the first hour after

pump start-up, loadio is calculated using the flow volume that occurred during the last

hour prior to pump shut-down, and loads1_9 are calculated using the two-hour flow

volumes (the total volume pumped starting one hour preceding water sample collection

and ending one hour after sample collection). The calculation for the example in

Figures 4a and b is shown in Equation 4:

Load14 = C14 {(a+b)/2 (Timed Timec)}, (4)

where Timec=13 hours, Timed=15 hours, a=flow rate at 13 hours, and b=flow rate at 15

hours, C14=water sample concentration at 14 hours, and Load14=two-hour incremental

load associated with C14.

Flow Composite Discrete Sampling With Continuous Flow Monitoring

Figure 5a shows a typical flow-weighted P concentration distribution where water

samples are collected several times during an event, based on the pumping of a

predetermined volume of water. In other words, each water sample is collected after a

predetermined and equal volume of water is pumped. Hence, each concentration value

is representative of an equal volume of water. This differs from time discrete water

14

sampling (discussed above) where water samples were collected at predetermined time

intervals, regardless of actual flow conditions.

In this example, the "equal volume of water pumped" is shown as the shaded

areas a in Figure 5b. The flow volumes between water sample concentrations used in

the load equation are fixed, equal, and predetermined. Now, it becomes simply a

matter of multiplying that fixed volume by each concentration and summing the

incremental loads (Equation 5):

Load = E{(C,+Ci+1)/2 Volume}, (5)

where i=0 through the number of the last water sample collected and Volume is fixed at

a predetermined value.

Concentration Co in Figure 5a is the concentration of the initial water sample

collected immediately at pump start-up. Concentration C, is the concentration of the

water sample collected after the first predetermined volume is pumped. Hence,

between concentrations Co and C1, the predetermined volume of water is pumped.

Averaging Co and C1 yields the concentration applicable to the first volume increment.

This average concentration is then multiplied by the predetermined volume to attain

load for the first incremental volume pumped. During the time between the collection of

C, and C2, an equal volume of water was pumped. Hence, averaging C, and C2 and

multiplying the average concentration by the predetermined volume will yield the load

for the second incremental volume of water pumped. To find the total load for the

event, the process continues as shown in Equation 6:

15

Load = {(Co+C1)/2 Volume a} + {(C1+C2)/2 Volume a} (6)
"+ {(C2+C3)/2 Volume a} + {(C3+C4)/2 Volume a}
"+ {(C4+C5)/2 Volume a}.

In essence, when flow composite samples are collected, and total flows are measured

using the appropriate instruments, the flow monitoring and water sampling equipment

are performing the above calculation automatically, yielding an event average

concentration C and the total volume pumped which are then simply multiplied together

to attain the total event load.

Flow Composite Water Sampling When a Pumping Event Spans More Than One
Sampling Period

Another common scenario that occurs in the EAA when monitoring for P loads is

when the flow composite water sampling protocol is being used, the pumping duration

is long, and the water sample pick-up time occurs sometime during the pumping event.

This yields a situation where the volume drained during a single pumping event is

associated with two composite water sample concentrations as depicted in Figures 6a

and 6b. The load calculation procedure in this case is actually a subset of the scenario

described in Figures 5a and 5b, where averaging of the period concentration is done by

the autosampler yielding concentrations C1 and C2. To calculate the two load

components that make up the total event load in this example, the volume of water

pumped between hours 0 and 12 (area abde in Figure 6b) is multiplied by C, and the

volume of water pumped between hours 12 and 20 (area bcde in Figure 6b) is

multiplied by C2. The two component loads are then simply added together to attain the

total event load. This calculation is done using Equation 7:

16

Load = (C,*Volume abde) + (C2*Volume bcef). (7)

In Figure 6b, note the heavy dashed lines delineating volumes pumped. These

lines represent approximations of water volumes pumped when data are collected only

three times during the event, at hours 0, 12, and 20. In this case, approximate water

volumes pumped are calculated assuming straight lines for curve segments a to b and

b to c. The unshaded areas between the dashed line and the pump flow curve is the

error in volume calculation that occurs if one samples three times during the event

rather than continuously. If the continuous flow monitoring protocol is being used, the

volumes pumped would be represented by abde and beef, following the curved line

rather than the dashed straight line. If the monitoring equipment is set up for

continuous monitoring and flow totalizing, the volumes reported would be those under

the curved line, where the instrumentation does the calculations automatically.

Adjustments to Load Measurements

It would be less than sensible to compare nutrient loads from different farms

based on the total mass of a nutrient discharged alone. Assuming a uniform rainfall

distribution over the area, a larger farm will discharge more water to maintain adequate

crop root zone conditions than a small farm, simply because a larger volume of rain

falls over the larger farm. In fact, the larger farm may actually be achieving less

drainage on a per acre basis and lower drainage water nutrient concentrations.

However, due to the farm size, absolute nutrient loads would appear to be much higher

17

for large farms than for small ones. This makes comparing the relative load

contributions and reductions between farms less than fair to the large farms. To

account for the farm size differences, the absolute nutrient loads need to be adjusted.

To normalize the data for farm size, one can simply divide the farm total load for the

event or time period of interest by the gross farm area, yielding a loading expression

"unit area load" (UAL) whose units are mass per unit area, typically Ib/ac or kg/ha.

In addition to adjusting absolute loads for farm size, it is also useful to adjust the

loads to account for hydrologic differences between years or other time periods.

Nutrient loads leaving farms during wet years (high rainfall volume years) are higher

than loads measured for dry years, simply because more rainfall occurs, requiring more

pumping to achieve suitable root zone conditions during a wet year. Antecedent soil

conditions and climatic conditions are also different between wet and dry years. These

differences make it less than rational to simply compare absolute loads or UALs for a

particular farm across years without factoring out the hydrologic differences. Data

normalizing techniques that adjust for both farm size and hydrologic differences across

time periods depend heavily on statistics and computer models, resulting in loading

values expressed as "adjusted unit area loads" (AUALs). Unit area loads and AUALs

are discussed in depth in Rice and Izuno (1997).

Summary

Different units of measure for expressing pumped water volumes, chemical

concentrations in water, and chemical loads were discussed. Conversion factors for

18

English and metric units commonly used, sometimes interchangeably, were explained.

The most commonly used methods and equations used for calculating P loads in the

EAA were described. Finally, normalizing data to enable comparisons between farms

and across different time periods was discussed briefly. Although load calculations are

relatively simple, care must be exercised when determining which concentration should

be considered to be representative of what volume of pumped water.

The selection of appropriate water sampling and flow volume measurement

methods, and the consistent and uniform application of load calculation techniques, are

preconditions for developing useful and interpretable load databases. The reader

should be aware that it is essential to know how flow and concentration data were

collected, and how load calculations were made, prior to drawing conclusions or making

recommendations based on existing data sets. Additionally, many monitoring

instruments can actually perform many of the cumbersome calculations automatically, if

properly programmed to do so, reducing load calculations to a simple concentration

multiplied by volume exercise.

19

References

Everglades Forever Act. 1994. Florida Statute Section 373.4592. Tallahassee, FL.

Izuno, F.T., A.B. Bottcher, and W.A. Davis. 1992. Agricultural water quality sampling
strategies. Florida Cooperative Extension Service Circular 1036. Institute of Food
and Agricultural Sciences, University of Florida. Gainesville, FL.

Izuno, F.T., R.W. Rice, R.M. Garcia, and L.T. Capone. 1996. Time versus flow
composite water sampling for regulatory purposes in the Everglades Agricultural
Area. ASAE Faper #962129 presented at the 1996 ASAE International Meeting on
July 17 in Phoenix, Arizona. American Society of Agricultural Engineers. St. Joseph,
MI.

Izuno, F.T. and R.W. Rice, eds. 1997. Implementation and verification of BMPs for
reducing P loading in the EAA. Phase V Final Report submitted to the EAA
Environmental Protection District. Belle Glade, FL.

Rice, R.W. and F.T. Izuno. 1997. Assessing phosphorus load reductions under
agricultural best management practices. Florida Cooperative Extension Service
Circular Institute of Food and Agricultural Sciences, University of Florida.
Gainesville, FL. (In review).

SFWMD. 1992. EAA Regulatory Program. Part 1 of Chapter 40E-63 of the Florida
Administrative Code. South Florida Water Management District. West Palm Beach,
FL.

20

i 0
ro ------- C--------^s --------

0- o \
-- g.......

Time Time

Figure 1: Typical phosphorus flow rate curve Figure 2a: Typical phosphorus concentration
showing changes over time. curve showing changes over time,
typical of an event with a "first-flush".

------event before declinin.---------
- ---.-..---[..-- E --

C-------------------.'---- C ---A------ -
o 0

a- ,-----/ -"

o - - - - - - _____

Time Time

Figure 2b: Phosphorus concentration change Figure 2c: Phosphorus concentration change
over time typical of an event where over time typical of an event where
concentrations increase as soil water concentrations increase as soil water
and particulate matter enter the and particulate matter enter the
drainage stream late in the event. drainage stream in the middle of the
event before declining.

*o \ o ~ ~-
0o
L -I-
CU --------- CU -- ^ $
I-c

\)

Time Time

Figure 3: Typical phosphorus concentration and flow rate curves showing changes over time when data
are collected only at the beginning and end of the pumping event.

o -

0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time, hr Time, hr

Figure 4a: Phosphorus concentration Figure 4b: Flow rate distribution through
distribution through a drainage a drainage event with flows
event with water samples calculated every two hours.
collected every two hours.

C--
0 -

I 0
0 I i I i i I I I

OI I

------I-- ------------
I I

0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time, hr Time, hr

Figure 5a: Flow-weighted Figure 5b: Predetermined equal
phosphorus concentrations. flow volumes between
hours 0 and 20.

C

S11

Of

--- d
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time, hr Time, hr

Figure 6a: Flow-weighted composite Figure 6b: Estimated flow volumes
phosphorus concentrations between hours 0 through 12
and the time period that and 12 through 20.
a concentration represents.

Florida Agricultural Experiment Station, Institute of Food and Agricultural Sciences, University of Florida, Richard L Jones, Dean for Research,
publishes this information to further programs and related activities, available to all persons regardless of race, color, age, sex, handicap or
national origin. Information about alternate formats is available from Educational Media and Services, University of Florida, PO Box 110810,
Gainesville, FL 32611-0810. This information was published June 1998 as Bulletin (Tech.) 906, Florida Agricultural Experiment Station.
ISSN 0096-607X