ILLINOIS POLLUTION CONTROL BOARD
September 29,
1975
IN THE MATTER OF
)
R71—15
PLANT NUTRIENTS
)
OPINION AND ORDER OF THE BOARD
(by Mr. Dumelle):
Certain nutrients are, essential for the growth of all
plants,
including those that are used indirectly through animals
or directly to produce most of man’s food supply.
These plant
nutrients are obtained from various sources, including reserves
in soils, plant and animal manures,
fertilizers,
etc.
If there
are excess supplies of plant nutrients,
especially of nitrogen
and to a lesser extent phosphorus, they may adversely affect
water quality.
BOARD ACTION
AND
HEARINGS IN 1971 AND
1972
In May,
1971,
the Illinois Pollution Control Board
(Board)
approved proposalE for public hearings on the application of
fertilizers, animal wastes, and sewage sludge and on controlled
access of livestock to streams.
The purposes of the hearings
were to determine
In
The Matter of Plant Nutrients,,
R71—l5,
4
PCB 127
(March 28,
1972):
“1.
Whether plant nutrients are creating pollution
problems by reaching surface or underground
waters
in excessive amounts.
2.
The present state of knowledge on the extent to
which fertilizers and animal manure contribute
to pollution problems.
3.
Whether steps can be taken to correct pollution
problems in the event that problems are identi-
fied and agricultural practices are found to be
important contributors.
Two
approaches were to be
considered:
(a) adoption of rules and regulations,
and
(b)
reliance upon needed research, education,
and voluntary changes in practices.
The Board expresses
its appreciation to Dr. Russell T. Odell,
a
former Board Member, and Professor of Agronomy,
Exreritus,
University
of Illinois, who prepared this Opinion in draft and discussed it
at length with the Board.
18
—
—2—
4.
The impact of alternative courses of action both on
the environment and on food supply and costs.
5.
The administrative feasibility of rules and regula-
tions.
Information presented at ten public hearings indicated
the status and trends of nitrates and phosphates in streams,
the role of different crop production practices in determining
the plant nutrient content of surface waters, and the environ-
mental and economic consequences of alternative practices.
How-
ever,
the hearinc~salso revealed major deficiencies
in available
information upon which the Board could make a decision concerning
plant nutrients.
The Board Opinion of March
28,
1972
(In The Matter of
Plant Nutrients,
R71—l5,
4 PCB 128-130), and especially Dr. S.R.
Aldrich’s Supplemental Statement of the same date, discuss in
detail many aspects ef nitrogen in relation to water quality.
The
nitrate content of surface waters is increasing in some streams
and during short periods may exceed the Water Quality Standard
of
10 mg/l nitrate nitrogen
(45 mg/l nitrate)
in certain streams
in east-central Illinois.
“Sources of excess nitrates
in streams
are well known bu~the magnitude of the contribution of each is
unknown.
Increasr~sin fertilizer nitrogen and in the acreage of
row crops are the most likely contributors.
“The record clearly supports the conclusion that within a
given set of supporting cultural practices, the greater the
amount of nitrogen apDlied the greater the potential for loss to
ground and surface waters
(Nov.
3,
p.
52).
But farmers who pro-
duce the highest yields and apply the highest rates of nitrogen
may produce a unit of crop with less potentially leachable nitrate
than farmers who produce average yields.
Surveys reveal that
recent increases
ir-. nitrogen consumption in Illinois, and for the
entire Cornbelt, are due to higher rates being applied on
previously underf~rtilizedfields.
The proportion of fields that
receive more than 150 pounds per acre has decreased in recent
years”
(Plant Nutrients,
4 PCB 128).
The increase in the cost of
nitrogen from approxi~nately5 cents per pound in 1971 to 17 cents
in 1975 has been greater than the change in the corn price per
bushel, which increased from approximately $1.25 to $2.75 during
the same. period.
This large and disproportionate increase in
cost of nitrogen during recent years has caused farmers to monitor
their nitrogen applications more carefully than when fertilizer
was cheaper.
The Board concluded that “the water quality standards for
nitrate nitrogen are presently being violated in certain streams
18
—
649
—3—
of the state and that the potential nitrate problem will grow as
the demand for food increases.
The record has not demonstrated
that health effects have resulted.
Deficiencies in available
knowledge on the credibility of the nitrate standard, on the con-
tributions of various nitrogen sources to nitrates in water, on
the effectiveness of possible control measures and on undesirable
side effects on the environment from alternative practices con-
vince
us that at this time we should make provision for more in-
formation on which to decide the issue rather than to promulgate
regulations of unpradictable effectiveness and side effects.
Accordingly we shall ask the Institute for Environmental Quality
to give high priority to obtaining information on nitrates which
will provide
a basis for early reconsideration of the matter.”
“Specifically we request the Institute to develop an
implementation
plan
for achieving the standard of
10 mg/l
nitrate nitrogen in public drinking and food processing water
or,
if Institute studies show that the costs to achieve com-
pliance are not justified by the benefits, to propose a revision.
Perhaps additional research will be required in stream sampling,
in connection with applications of fertilizer, animal and
human
wastes,
in groundwater hydrology, and in cropping systems before
an implementation plan can be developed.
We suggest that the
Institute also consider the technical feasibility and economic
reasonableness of re’noving nitrates from public drinking and
food processing water
in case a choice becomes necessary at a
future date between nitrate
removal versus major adjustments in
food production practices in order to reduce the sources of
nitrates”
(Plant Nutrients,
4 PCB 130).
Concern for phosphorus is caused by its possible role as
a contributor to accelerated eutrophication of surface waters.
“Phosphorus has been shown to be a serious water contaminant in
the Fox River as well as in Lake Michigan
(for which a phosphorus
standard was established) and other still waters, but not in flow-
ing streams generally.
Nor has the evidence stressed the con-
tribution of fertilizer runoff in these Illinois waters subject to
serious phosphorus problems.
...
The behavior of phosphorus
fertilizer when aprlied to the soil is well known.
Illinois soils
have retention caI.acities for phosphorus far
in excess of the
amounts applied in fertilizer.
Hence, phosphorus in tile drainage
effluent
is little affected by fertilizer applications”
(Plant
Nutrients,
4 PCB l3fl.
Phosphorus is held closely by soil particles
and
is not mobile nor lost except by soil erosion.
On the basis of
testimony given,
the E~oarddecided that regulations on phosphorus
fertilizer should not be imposed.
18
—
650
—4-
The Board found no support for a regulation to fence live-
stock from streaTns.
It also decided not to adopt a regulation
concerning rate of application of sludge on land since no nitrogen
limit was establiscied for fertilizer.
Several aspects of these
problems and of animal manures have been considered recently in
Livestock Waste Regulations, R72—9,
13 PCB 451-469
(August
29 and
September
5, 1974).
I.I.E.Q. DOCUMENT NO.
74-38, FERTILIZER APPLICATION RATES
AND NITRATE CONCENTRATIONS IN ILLINOIS SURFACE WATERS
(December
1974)
In response to the request of the Board on March 28,
1972
(Plant Nutrients,
R71—l5,
4 PCB 130),
the Illinois Institute for
Environmental Quality
(Institute) contracted with the Center for
the Biology of.Natural Systems
(CBNS), Washington University, to
(1)
analyze “the relationship between observed concentrations of
nitrate
in Illinois rivers and the rate of application of nitrogen
fertilizer to Illinois farmlands,
and
(2)
from this analysis to
determine what influence various limitations
of the rate of
nitrogen application might be expected to have on nitrate levels”
(Institute Doc.
74—38, page
1).
Variations i~nitrate concentrations among Illinois
rivers are due
to
(1) variations among different
sampling points,
(2) variations during different seasons of the year at a speci-
fied sampling station, and
(3) variations from year to year dur-
ing specified
seasons at a specific sampling station.
Nitrate
nitrogen values range “from less than
1 ppm in
certain rivers in
southern Illinois to over 10 ppm
in east-central Illinois rivers”
(Institute Doc.
74—38, page 5).
Average nitrate plus nitrite
nitrogen and total phosphorus values for 1974 in the sub-basins
in Illinois are gi’ren in Table
1;
seasonal values will range more
widely.
Land use varies markedly among different watersheds which
drain to various sampling stations.
This
is the major relationship
studied by CBNS.
Nitrate concentrations in streams are usually
highest during the sçring months each year.
Mean nitrate nitrogen
concentration for samples in the spring
(April through June)
is the measure of water quality used in this analysis.
Table
1.
Average values of nitrate plus nitrite nitrogen
and total phosphorus for 1974 within sub-basins
In Illinois.*
Sub-basin
Major basin
Nitrates plus
Total
nitrites as N
phosphorus
mg/l
mg/l
Shore water
Lake Michigan
0.3
0.06
Lake Michigan trib.
“
n
0.1
0.16
18
—
651
—5—
Sub-basin
Major basin
Nitrates
nitrites
plus
as N
Total
phosphor
mg/l
mg/i
Des Plaines direct
Des Plaines
2.1
0.80
Du Page River
“
“
3.5
2.00
Chicago San.
& Ship Can.
“
“
1.7
1.78
Fox River
Illinois
2.4
0.31
Kankakee River
Illinois
4.7
0.22
Sangamon River
“
5.5
0.44
Illinois direct
“
4.8
0.54
Rock River
Mississippi
3.8
0.45
Mississippi
R. North
“
3.2
0.43
Mississippi
R. Central
“
1.3
0.3~
Kaskaskia River
Mississippi
2.0
0.41
Bid Muddy River
“
0.8
0.46
Mississippi
R. South
“
1.1
0.38
Wabash River
Ohio
3.4
0.41
Saline River
“
1.2
0.20
Ohio direct
“
1.2
0.17
*Illinois Water Qu~1ityInventory Report
1975.
Illinois Environmental Protection Agency.
The Illinois Environmental Protection Agency has identified
eight Illinois towns
with seasonally elevated nitrate levels
in surface water supplies.
All of these towns are
in East—
Central Illinois
(see page
9
).
The eight towns are Decatur,
in the Sangamon River sub-basin
(Table
1); Bloomington,
Eureka,
Pontiac, and Streator in the Illinois direct sub—basin;
and
Charleston, Danville, and Georgetown in the Wabash River
sub—basin.
The general approach in this study was to determine the
mathematical relat4.onship between nitrate nitrogen concentrations
in streams and factors, especially nitrogen fertilizer, which are
expected
to influence nitrate concentrations.
Calculations were
then made to estimate changes needed in agricultural practices
to reduce nitrate concentrations to acceptable levels.
Finally,
estimates were made of the impact of such changes in agricultural
practices on crop production and income.
18
—
652
—6—
Relationships Between Nitrate Nitrogen
in Surface Waters and Selected Variables
In regression analysis, the choice of “independent”
variables that are expected to influence the dependent variable
(nitrate concentration in streams)
is very important.
Sources
of nitrates in streams include effluent from sewage treatment
plants and diffuse sources,
such as drainage from land.
Under
natural conditions, nitrate is produced in the soil by bacterial
action on organic matter.
The rate at which nitrate is released
from soil organic matter is affected by temperature, moisture,
soil type,
and agricultural practices such as percent of land in
row crops, which stimulates nitrate release.
CBNS used the per-
cent of
land in row crops
(corn and soybeans)
as
a proxy measure
of nitrate from soil organic matter.
A more direct measure of
nitrogen released from soil organic matter would have been
the estimated amount of soil organic matter per watershed acre,
which can be estimated from data in the University of Illinois
Agronomy Department.
Row crops are grown most extensively on
soils that are high
in organic matter.
Except
for livestock manure on some fields,
the other major
source of nitrates in soil is fertilizer nitrogen.
Since fertilizer
statistics are available only on a county—or state-wide basis,
CBNS developed methods to convert these
to the rate of fertilizer
nitrogen applied per watershed acre.
Multiple linear regression
methods were used to determine the degree to which variations
in stream nitrate levels are associated with differences in soil
organic nitrogen
(measured by proxy percent of land in row crops)
and fertilizer nitrogen..*
Because soil organic nitrogen was
considered to exert little influence
,
CBNS dropped this variable
and proceeded to develop an equation to predict stream nitrate
concentrations witn nitrogen fertilizer
as a major independent
variable.
The dropping of soil organic nitrogen has important
implications because it is closely correlated (Institute Dcc.
74-38, page 20) with fertilizer nitrogen applied per watershed
acre and the latt?r will then measure its own effects plus
intercorrelated effects of soil organic nitrogen.
Therefore,
the net effect of fertilizer nitrogen on stream nitrogen
concentrations will be overestimated and subsequently the
effects of fertilizer control regulations will be overestimated.
This
is one of the basic f4ults in the CBNS study, which is
recognized on pages 51—54 of Institute Document 74—38.
In this study water quality data in the spring
(April
through June) were paired with fertilizer nitrogen data for
the previous year and row crop data for the previous year.
After
*According to Table
2
(page 17) of Institute Document 74-38,
the
cube
of row—crop acreage
(F) explains more of the WQ
(stream nitrate)
va~:iahilitythan does nitrogen fertilizer cubed
(E) from 1964 through
1968, whereas the opposite relationship exists from 1969 through
1972.
Therefor~, the choice of years included in this analysis
(1969-1971)
markedly affects the results obtained.
18— 653
—7—
preliminary analysis, data for the three years
1969,
1970,
and
1971 were studied in detail.
1970 was atypical in that a fungus,
southern corn leaf blight, seriously reduced corn yields.
Since
the amount of corn produced and nitrogen taken up in 1970 was
small
in relation to fertilizer nitrogen applied, some of the
latter was more 1ike~.yto be free for leaching in subsequent years.
Nitrate concentrations in 22 Illinois streams were
studied during the three years
1969 through 1971, making a
total of
66 obser”ations.
Various linear and curvilinear
mathematical expressions of the relationship between nitrogen
concentrations in streams and fertilizer nitrogen used per water-
shed acre were studied.
On the basis of these preliminary studies,
the CENS chose a curvilinear function, the cube of the rate of
fertilizer nitrogen applied to introduce this variable into the
overall predictive equation.
It should be noted that the cube
curve in Figure
3
(page
23), which was calculated from the points
in Figure
2
(page
22)
of Institute Document 74-38,
is almost
entirely dependent upon six of the
66 observations in this
analysis; otherwise the relationship would be linear.
Since the
validity of the subsequent analyses and estimates are markedly
influenced by the curvilinear relationship selected to express
this fertilizer nitrogen variable,
the findings are dependent upon
a narrow data base
in which six observations exert an inordinate
influence.
Therefore, caution should be exercised in expanding
from such a narrow data base.*
Additional variables were tested and two of them were
finally incorporated into the predictive equation:
(1) stream
discharge
in relaticn to the long term average stream flow for
the period of the year in question, and
(2) urban population
density as
a proxy variable for nitrates from urban sources.
The predicti”re equation for this study is:
“WQ
=
0.31
+
5.48 x lO~NF3
+
1.20 x 105NF3
FLO
+
0.0047 URBPOP,
where WQ is mean ritrate nitrogen in ppm in the April-June period,
NF is the nitrogen fertilizer ‘application rate per watershed acre,
FLO
is the average daily stream discharge rate for the April—June
period divided by a yearly average daily discharge over 10 to 30
past years,
and URBPOP is
a measure of urban population density
per square mile”
(Institute Doc.
74-38, page 34).
The above
equation can be summarized in words as follows:
“the average
nitrate concentration of Illinois streams
in the April—June
period
is
a function of a constant term
(O.31.),•plus a coefficient
*This and other deficiencies in the study were mentioned in review
comments by the CBNS advisory committee for this study.
See “Advisot
Committee Comments on Determination of Application Rates
of Nitrogen
Fertilizer to Achieve a Series of Nitrate Concentrations in Surface
waters and the Economic Effects Thereof,” July 1974.
18
—
654
—8—
(5.48 x l0~) times the cube of nitrogen fertilizer use per water-
shed acre, plus a coefficient
(1.20
x l0~) times, an interaction
term which
is a product of the cube of nitrogen fertilizer use
per watershed acre and the flow of the stream relative to a long
term average, plus a coefficient
(0.0047)
times the density of
urban population per square mile in the watershed.”
The co-
efficient of determination, R2
=
.76,
is
a measure of the pro-
portion of the total variation in the dependent variable
(river
nitrate concentration) that
is associated with the three inde-
pendent variables.
In this equation 76
of the variation in
nitrate concentration
in Illinois streams in the April-June period
is associated with the three independent variables included herein
or factors correlated with them.
Limitations on Nitrogen Fertilizer Estimated to Achieve
Various Levels of Nitrate Concentrations in Surface Waters
The next step was to use the above equation to predict the
stream nitrate concentrations that would be expected under specified
values of the three independent variables.
The predicted stream
nitrate concentrations are not certain, but are expressed in terms
of probability.
The predicted stream nitrate concentrations under different
specified conditions are expressed in terms of percent failure rate,
i.e., the rate of nitrogen fertilizer application
(e.g.,
150 pounds
per acre)
which would be required in order that a given stream nitrate
concentration
(e.g.,
10
ppm) be exceeded no more than a given percent
of the time.*
“This
is done by
(a) assigning typical values to
stream flow and urban population values in our predictive equation
(a relative stream flow of 1.5, which is the 1946—1971 mean for
Illinois streams, and an urban population density of 25 per square
mile, which is the median for the sample watersheds in 1970),
(b)
taking the confidence interval for predictions of nitrate nitrogen
concentration” from
the
predictive equation,
“(c) using the
confidence interval for predictions from this regression to cal-
culate the mean nitrate nitrogen concentration that would allow
10 ppm to be exceeded with a given probability which corresponds
to the failure rate,
and
(d) then calculating the nitrogen
fertilizer use per watershed acre consistent with the mean nitrate
nitrogen concentration by plugging nitrate concentration and
the given values for. stream flow and urban population into” the
predictive equation and solving for fertilizer
(Institute Doe.
74-38, page
36)’.
The calculated rate of nitrogen application
that results is expressed in the total acreage of the watershed,
not the acreage to which fertilizer
is applied.
*The current Illinois standard for public water supplies
is
10 mg/l
(or
IC
ppm)
of nitrate nitrogen; with a maximum value of
20 mg/i
allowed
for up to a total of 35 days per calendar year,
of which no singi
period shall be longer than
15 consecutive days.
18
—
655
—9—
Approximately 90
of the fertilizer nitrogen used in Illinois
is applied to corn.
“However,
in the predictive equation, the
relevant variable is the rate of nitrogen application per watershed
acre.
Hence,
it is necessary to convert the value computed on the
basis
of watershed acreage to one computed on the basis of corn
acreage.
The relationship between the average rates of application
of nitrogen per watershed acre to a limit set on the rate of
application to corn acreage depends on two factors:
(1)
the fraction
of the watershed area in corn and
(2)
the frequency distribution
of rates of application of fertilizer nitrogen to corn when nitrogen
use
is uncontrolledt’
(Institute Doc.
74-38, page 37).
The latter
data for 1971 were obtained from the Doane Agricultural Service
for five regions
in Illinois:
North, West—Centrals- East—Central,
Southwest,
and Southeast.
“The average application of nitrogen
per watershed acre consistent with a particular water quality
standard can be converted to the average application per corn
acre by dividing
by
the ratio of corn acreage to total acreage
in an area to which the analysis is applied.
In doing this we
assume all nitrogen fertilizer is applied to corn.”
Table
2 shows
the results of estimates, using the predictive equation,
of the
limits on the rate of fertilizer nitrogen applications to corn
that would be required in order that concentrations of
10 ppm
nitrate nitrogen
ir.. streams would be exceeded at various failure
rates.
Table
2.
Estimates of limits of nitrogen fertilizer applications to
corn
in order to meet a 10 ppm nitrate nitrogen water qualit
standard with various probabilities of exceeding the
standard for regions in Illinois.
1971
Maximum fertilizer N application rate
Regions
fraction
Fraction
(lbs N/acre corn) corresponding to
in
111.
of land
in corn
of
in
land
corn
percent time
10 ppm N03—N will be exceeded
1
5
20
50
North
.31
.50
.45
.40
.35
30
90
110
120
160
Unregulated
110
130
160
130
170
Unreg.
170
Unregulated
Unregulated
West—
.36
.50
‘
100
130
170
Unreg.
Central
.45
.40
120
150
160
Unregulated
Unregulated
.35
.30
Unregulated
Unre~u1ated
18—656
—10—
1971
Maximum fertilizer N application rate
Regions
fraction
Fraction
(lbs N/acre corn)
corresponding to
in
of land
of land
percent time
10 ppm N03—N will be exceeded
111.
in corn
in corn
_________________________________________
1
5
20
50
East—
.39
.50
80
100
110
120
Central
.45
.40
.35
.30
.25
90
110
130
150
110
130
160
260
120
170
Unregulated
170
Unregulated
Unregulated
Southwest .21
.50
.45
.40
.35
.30
90
110
120
150
100
120
160
Unregulated
120
160
Unregulated
160
Unregulated
Unregulated
Southeast
.16
.50
.45
.40
.35
100
130
170
Unregulated
120
160
Unregulated
150
Unregulated
Unregulated
lJnregulate
The estimated limits are markedly influenced by variations
in the fraction of land in corn.
For example,
in East-Central
Illinois the limit on the rate of nitrogen application which would
be required in order that 10 ppm nitrate nitrogen stream concentra-
tions be exceeded no more than 5
of the time is 110 lbs N/acre
of corn when 45
of the land
is in corn, or 170 lbs N/acre of
corn when 35
of the land is in corn.
Table
2 indicates that different regions in Illinois would need
different limits on the rate of fertilizer nitrogen application for a
given fraction of land in corn to achieve the same stream nitrate
concentration.
For example,
in the Southeast Region no limit on
nitrogen fertilizer is needed
(in order that stream nitrate nitrogen
concentrations
of 13 ppm not be exceeded more than 5
of the time)
if 40
of the land is corn,
whereas
in the East—Central Region a
limit of 130 lbs N/acre would be needed to achieve the same water
quality with the same percent of the area
in corn.
For the 1971
fractions of land in corn,
“if nitrogen fertilizer use was regulated
on a regional basis, only East—Central Illinois would require regula-
tion
for. the
10 ppm standard at the
1
failure rate”
(Institute
Doc.
74-38, page
43).
18
—
657
—11—
“If public policy is to be based on a program of limiting the
use of nitrogen fertilizer,
it must balance the hazards associated
with high nitrate concentrations against the social cost of setting
a limit on an important agricultural input.
.
.
.
The most effective
means of controlling high nitrate concentrations in problem areas
such as East-Central Illinois with a minimal effect on the overall
agricultural output of the State is to set these limits on a regional
or subregional rather than a statewide basis.”
Table
3 shows “that
a regulation designed to achieve a given water quality in East-Centra
Illinois when applied to the State rather than to the East—Central
Region alone would affect twice as many farms, double the loss
in
corn output, and double the loss in gross income”
(Institute Doc.
74—38,
page 44).
Table
3.
Percentage of farms affected, percentage change in cor
output, and change in gross farm income due to a state
wide limit of 130 pounds of nitrogen per acre and
differentiated by regions in Illinois.
Item
Regions in Iilinois
N
W-C
E-C
SW
SE
Percent of land in corn,
1971
31
36
39
21
16
Limit on N/acre corn
Differentiated limits
standard with a
5
to meet a 10
ppir
failure rate
none
none
130
none
none
r~
Percent of farms &ffected
0
0
37
0
0
Percent reduction in
corn output
0
0
1.1
0
0
Decrease in gross income
per affected farm
(S)*
.
—
539
—
Total decrease in gross
farm income
(mu
$)~
0
0
6.6
0
0
Percent of farms affected
Statewide limit of 130 lbs N/acre corr
23
24
37
23
13
Percent reduction in corn
output
0.6
0.7
1.1
1.6
0.7
Decrease
in gross income
per affected farm ~$)*
417
429
539
573
386
Total decrease in gross
farm income
(mu
$~
1.5
1.4
6.6
2.4
0.6
*The price used per bushel of corn
is $1.25.
18
—
658
—12—
Economic Implications of Limitations on Fertilizer Use
The establishment of public policy relative to the stream nitrate
problem involves
ar.. evaluation of the cost,
to individuals and society,
of any measures that might be introduced which reduce agricultural out-
put.
“In evaluating this cost it is
important to determine how many
farmers
would be affected by the imposition of a given limit on the
rate of fertilizer nitrogen application and how these limits would
affect the consumption of nitrogen fertilizer
(and hence the
enterprises that depend on selling it),
the output of corn,
and the
gross and net farm incomes” in the different agricultural regions
in Illinois
(Institute Doc.
74-38, page 55).
The estimates in this
subsection are based upon the relationships found in the previous
two subsections.
In this study,
1971 prices were used for corn
($1.25 per bushel)
arid nitrogen
(S cents per pound).
Administrative
or enforcement costs of possible controls were not considered.
Table
4 shows the estimated percentage of farms affected by
regional limitations on the rate of nitrogen application to corn which
would be required tn order that the 10 ppm nitrate nitrogen water
quality standard not be exceeded by specified percentages of time.
East-Central Illinois would be much more severely affected than
other regions, as would be expected from the estimates
in
Table
2.
In fact, for the 1971 fractions of land in corn,
only
East—Central Illinois would be affected by a fertilizer nitrogen
limitation needed to avoid exceeding the 10 ppm nitrate concentration
in streams.
Table
4.
Percentage of farms affected by the nitrogen fertilizer limits
for the 10 ppm nitrate nitrogen standard by regions in I1linoi~
Regions
In
Illinois
197
of
in
1 fraction
lana
corn
Fra
of
in
ctio
land
corn
n
Percent failure rates
1
5
~0
50
North
.31
.50
.45
.40
.35
57
51
37
8
51
23
5
23
5
8
West—Central
.36
.50
.45
.40
38
29
14
24
6
6
East—Central
.39
.50
.45
.40
.35
.30
76
71
‘59
52
12
66
59
37
12
59
37
18
52
22
1
Southwest
.21
. 50
.45
.40
.35
47
42
30
9
37
30
9
30
9
14
18
—
659
—
13—
Regions in
Illinois
197
of
In
1 fraction
land
corn
Fra
of
in
ctlon
land
Percent failure rates
corn
1
5
20
50
Southeast
.16
.50
.45
.40
36
20
9
13
5
3
Table
5 presents estimates
of the percentage decrease in nitroge
fertilizer use by regions
if controls were established on nitrogen
fertilizer.
East-Central Illinois would be affected most for any giv
failure rate, which reflects the greater proportion of land in corn
with associated nitrogen fertilization in that region.
A 15%
decrea~
in nitrogen use
(3.5 tons per affected farm)
is estimated for contro~
corresponding to a 5% failure rate in East—Central Illinois
if corn
~.
planted on 40% of the land.
Table
5.
Percentage decrease in nitrogen fertilizer use due to the
nitrogen fertilizer limits for the
10 ppm nitrate nitrogen
standard by regions in Illinois.
Regions in
Illinois
197
of
in
1
fraction
land
corn
Fra
of
in
ction
land
Percent failure rates
corn
1
5
20
50
North
.31
.50
.45
.40
.35
33
21
15
4
21
11
3
11
3
4
West—Central
.36
.50
.45
.40
25
16
7
13
5
4
East—Central
.39
.50
.45
.40
.35
.30
43
36
25
19
6
30
25
15
6
25
15
8
19
10
1
Southwest
.21
.50
.45
.40
.35
32
26
16
6
21
16
6
16
6
7
Southeast
.16
.50
.45
.40
25
15
6
12
4
3
18—
660
—14—
The percentage decrease in corn output
in any given cell in
Table
6
is less than for the corresponding cell
in Table
5 because
reductions in fertilizer use which might be associated with controls
occur along the upper, nearly level segment
of the yield response
curve.
Table
6.
Percentage decrease in corn output due to the nitrogen
fertilizer limits for the
10 ppm nitrate nitrogen standard
by regions
in Illinois.
Regions
in
Illinois
197
of
in
1 fractio~i
land
corn
•
Fra
of
in
ction
land
corn
Percent failure rates
1
5
20
50
North
.31
.50
.45
.40
.35
3.3
1.7
1.0
0.1
1.7
0.6
~0.05
0.6
.~0.05
0.1
West—Central
.36
.50
.45
.40
2.2
1.3
0.2
0.7
0.1
.~0.05
East—Central
.39
.50
.45
.40
.35
.30
8.6
6.3
2.9
1.8
0.1
4.3
2.9
1.1
0.1
2.9
1.1
0.2
1.8
0.4
0
Southwest
.21
.50
.45
.40
.35
6.9
5.1
2.6
0.3
3.7
2.6
0.3
2.6
0.3
0.5
Southeast
.16
.50
.45
.40
3.3
1.4
0.4
0.7
0.1
..~.0.05
Table
7 showsthe estimated chanqe in net income per farm if
nitrogen fertilizer use were controlled for the various conditions
specified.
Although the estimates
indicate that nitrogen fertilizer
controls would reduce net farm income under most circumstances,
they
also indicate that the changes in net income for higher failure rates
and lower percentages of’land in corn are positive.
Even though these
estimates suggest that unneeded fertilizer
is applied where a relatively
small percentage of the land is in corn,
this
is unlikely because
most farmers take care to avoid fertilizing beyond the point of
diminishing economic returns.
In East—Central
Illinois,
if 40
of
tne
1an.-~ is
in corn and a failure rate of
5
is acceptable,
the average
annual net income p~raffected farm would be reduced $186
if controls
were placed on the rate of nitrogen fertilizer application.
18
—
661
—15—
Table
7.
Changes in net income
(in dollars) per farm for farms
affected by the nitrogen fertilizer limits for the
10 ppm
nitrate nitrogen standard by regions in Illinois.
Regions in
Illinois
1971
fraction
of lane
in corn
Fra
of
in
ction
land
corn
Percent failure rates
1
5
20
50
North
.31
.50
.45
.40
.35
—554
—278
—146
+254
—278
—100
+378
—100
+378
+254
West—Central
.36
.50
.45
.40
—479
—329
+120
—127
+331
+345
East—Central
.39
.50
.45
.40
.35
.30
—1589
—1176
—532
—319
+318
—797
—532
—186
+318
—532
—186
+180
—319
+51
+1072
Southwest
.21
.50
.45
.40
.35
—948
—753
—508
—20
—596
—508
—20
—508
—
20
—106
Southeast
.16
.50
.45
.40
—353
—163
+4
—39
+237
+273
OTHER CONSIDERATIONS
The
CBNS
stucy considered only one policy alternative to
deal with the stream nitrate problem
—
limitations on rates of
nitrogen fertilizer in combination with the percentage of land in
corn.
In addition to considering regulations
for such a problem,
other alternatives
such as education, economic incentives, and
public investment should be considered.
Dr.
E.R. Swanson discussed some of the above alternatives
and the kinds of information needed to evaluate them in “Non—Point
Sources of Water Pollution, Problems and Policy Alternatives:
Agriculture”
durinci a workshop on March
20, 1975.
Research and
18—662
—16—
education are effective in identifying and applying farm practices
(such as time and amount of fertilizer application, use of
cover crops,
etc.)
that minimize adverse effects on water quality.
Economic incentives might include excise taxes on nitrogen fertilizer,
establishment of market “rights” to purchase fertilizer,
etc.
Although the tax system and the market “rights” system would involve
less administrati’c’e cost than a regulatory system, monitoring would
be necessary under a tax system to prevent imports from other
areas.
Public investment includes those investments which are not
economically justifiabl~efor individual users of water,
but which
can be justified by governmental units
to achieve desired levels
of water quality.
For example, removal of excess nitrates from
public water supplies is now feasible and is being done at Garden
City, Long Island, New York.
Ion exchange using resins would probably
be most feasible for nitrate removal.
The costs would be about
5
cents
per 1000 gallons.
Under some circumstances an alternate public water
supply may be justified.
Dr. Swanson ~iso indicated that in order to evaluate the economic
implications of different public policy alternatives, information
was needed concerning
(1)
the costs of adjustments
in farm production
related to the generation of potential pollutants,
(2)
the policy
implementation costs,
(3)
the physical relationships between levels
of on—farm
pollutant generation and the various parameters
of water quality at the locations of interest,
(4)
the functional
relationship between these parameters and damage to the environment,
including human health, and
(5)
the response of individuals
in the
system to various alternatives.
Quantitative information is
especially deficient concerning item
(3)
above.
This inadequate
physical data base makes
it difficult to determine the validity
of economic studies based upon such limited physical data.
Al-
though economic studies of several alternative policies have been
made,
the results are open to question because of the weak linkages
between on—farm practices and water quality.
This emphasizes the
need of measuring the effect of nitrogen fertilizer rates on
the nitrate content of surface and groundwater.
More permanent
water sampling stations are also needed to determine water quality
status and trends,
including nitrates.
The administrative
feasibility or enforcement costs of
possible controls on nitrogen fertilizer were not considered by
CBNS
(Institute Dcc.
74-38, page 55).
It would be difficult
to fairly administer such controls on approximately 124,000
farms
in Illinois or 38,000 farms
(1969 Census of Agriculture)
in the East-Central Region, especially since farms differ in
soil types,
slope, conservation practices, cropping patterns,
and livestock interprises, and in alternative ways of accomplishing
similar goals.
Under such circumstances,
regulations should not
he
ac~opted
lightly.
18
—
663
—17—
Our food production system is very complex, with close
multi-state and international linkages.
Corn is currently
being grown where it can be produced most economically, and
East—Central Illinois has
a prime economic advantage for
corn production.
For a given supply
of corn, nitrogen fertilizer
controls would shift production to less adapted areas, where costs
would be greater and alternate problems, such as accelerated
erosion on more sloping soils, would occur.
Such a trade-off would
not be in either the private or the public
interest.
FINDINGS OF THE BOARD
On the basis of the record in
this proceeding,
the Board makes
the following findings:
1.
The nitrate problem is not a statewide problem in Illinois
with our water quality standard of
10 ppm nitrate nitrogen
(20 ppm
is permitted for short periods).
The nitrate problem in streams
is concentrated
in East—Central Illinois, with lesser problems
in the West—Central and Northern Regions.
Streams
in southern
Illinois are low in nitrates, but the water in some shallow
wells on farms
is high in nitrates.
These high nitrates in farm
wells are ascribed to livestock wastes and septic fields rather
than to fertilizer.
2.
Differences among regions in the intensity of nitrogen
fertilizer use and the percentage of land
in corn lead to differences
between regions in
the estimates of limits on the rate of application
of fertilizer nitrogen to corn which would be required to reduce
stream nitrate concentrations to specified levels.
3.
Estimates of these limits are more sensitive to the
intensity of fertilizer use and the percentage of land in corn than
they are to variations
in the other two factors
(relative stream
flow &nd the density of urban population per square mile)
that were
included in the study.
However, the omission of soil organic
matter nitrogen from the analyses will probably overestimate the
effects of nitrogen fertilizer on water quality and the effects of
possible nitrogen fertilizer controls.
4.
If nitrogen fertilizers were regulated,
for example,
in
East-Central Illinois, corn production and net income would be
reduced on most farms
in the region.
5.
The relationship between nitrogen fertilizer practices on
farms and nitrate concentrations
in streams is not adequately enough
established to adopt regulations on nitrogen fertilizer use at this
time.
Special attention should be given to measuring these relation-
ships.
18—
664
•1
—
6.
Information is too meager to determine whether alternatives
such as economic incentives or public investments should be used to
control nitrate concentrations in water from streams.
There
is very
little information on the administrative feasibility or enforcement
costs of possible controls on nitrogen fertilizer.
On the basis of
available information on the above topics,
the alternative of
municipal water treatment, and the uncertainties as to the biases
in the predictive equation in
the CBNS study, the Board finds that
a regulation on nitrogen fertilizer is not justified at this time
because the likely improvement in water quality does not outweigh
the probable disruption of our food supply.
7.
Special attention should be given to research and
education programs to reduce the nitrate concentrations in streams
in East-Central Illinois.
ORDER
IT
IS THE ORDER OF the Illinois Pollution Control Board that:
1.
The regulation of nitrogen fertilizer to control nitrates
in Illinois streams is not justified at the present time.
2.
The proceeding in R71-l5
is hereby concluded.
I, Christan L. Moffett, Clerk of the Illinois Pollution Control
Boardb, hereby certify the above Opinion and Order were adopted on the
~gl’
day of September,
1975 by
a vote of
4/..Ø
Q~471
GJ_
Christan L. Moff~/Clerk
Illinois Pollution Control Board
18
—
665
