• TABLE OF CONTENTS
HIDE
 Copyright
 Front Cover
 Title Page
 Table of Contents
 Introduction
 Analysis of weather data
 Computation of freeze probability...
 Sample computation
 Discussion of results
 Using the probability tables
 Limitations
 Conclusions
 Literature cited
 Acknowledgements
 Back Cover






Group Title: Florida Agricultural Experiment Station bulletin no. 777
Title: Freeze probabilities in Florida
CITATION PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00047742/00001
 Material Information
Title: Freeze probabilities in Florida
Series Title: Bulletin University of Florida. Agricultural Experiment Station
Physical Description: 22 p. : maps ; 23 cm.
Language: English
Creator: Bradley, James T
Publisher: Agricultural Experiment Stations, Institute of Food and Agricultural Sciences, University of Florida
Place of Publication: Gainesville Fla
Publication Date: 1975
 Subjects
Subject: Frost -- Florida   ( lcsh )
Atmospheric temperature -- Florida   ( lcsh )
Frost -- Forecasting -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Includes bibliographical references (p. 22).
Statement of Responsibility: James T. Bradley.
General Note: Cover title.
Funding: Florida Historical Agriculture and Rural Life
 Record Information
Bibliographic ID: UF00047742
Volume ID: VID00001
Source Institution: Marston Science Library, George A. Smathers Libraries, University of Florida
Holding Location: Florida Agricultural Experiment Station, Florida Cooperative Extension Service, Florida Department of Agriculture and Consumer Services, and the Engineering and Industrial Experiment Station; Institute for Food and Agricultural Services (IFAS), University of Florida
Rights Management: All rights reserved, Board of Trustees of the University of Florida
Resource Identifier: oclc - 02289230

Table of Contents
    Copyright
        Copyright
    Front Cover
        Front Cover
    Title Page
        Title Page
    Table of Contents
        Table of Contents
    Introduction
        Page 1
    Analysis of weather data
        Page 1
    Computation of freeze probability dates
        Page 2
    Sample computation
        Page 3
    Discussion of results
        Page 3
    Using the probability tables
        Page 4
        Page 5
        Page 6
        Page 7
    Limitations
        Page 8
    Conclusions
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
    Literature cited
        Page 22
    Acknowledgements
        Page 22
    Back Cover
        Back Cover
Full Text





HISTORIC NOTE


The publications in this collection do
not reflect current scientific knowledge
or recommendations. These texts
represent the historic publishing
record of the Institute for Food and
Agricultural Sciences and should be
used only to trace the historic work of
the Institute and its staff. Current IFAS
research may be found on the
Electronic Data Information Source
(EDIS)

site maintained by the Florida
Cooperative Extension Service.






Copyright 2005, Board of Trustees, University
of Florida





Bulletin 777 cal)
Bulletin 777 (tjec-hical)


Agricultural Experiment Stations
Institute of Food and Agricultural Sciences
University" of Florida Gainesville


June 1981


ILITIE!


.




















FREEZE PROBABILITIES IN FLORIDA

James T. Bradley
Former State Climatologist, National Weather Service
and University of Florida Agricultural Research Center,
Lakeland, Florida.





First Printing November 1975
Second Printing June 1981



















CONTENTS

Page
Introduction --------------------------------------- 1

Analysis of Weather Data _........... ------......--.-....... ..------.......... 1

Computation of Freeze Probability Dates -----... --------------.. 2

Sample Computation .... -----------....------.....---.--- 3

Discussion of Results ....... ----------------- 3

Using the Probability Tables ------___---------------------__ 4

Limitations ...........................--------............... -------------- 8

Conclusions ...........................------.....................----------.............----------------------- 8

Literature Cited __----.----------------------------- 22

Acknowledgments .----...._-----..---------. ..------..-......---- 22










INTRODUCTION


Freeze probabilities are useful to Florida agricultural indus-
tries which produce cold sensitive crops as well as to homeowners
and engineers. The time interval between the 10% risk and the
50% risk of the first freezing temperature is 35 to 40 days, which
is much longer than in many states. This time interval is of im-
portance to farmers who often make decisions and long range
plans based on freeze probabilities.
The usual methods used in determining climatological probabili-
ties of the first and last occurrence of low temperatures are not
particularly applicable to Florida because there is not a well-
developed fall and spring season. For example, temperatures be-
low 32'F or 280F do not occur each year. Also, in seasons
where freezes do occur, freezing temperatures may not occur be-
fore January 1, while in other seasons the last freeze may occur
before December 31. Finally, there are some cold seasons in which
there is only one cold day which is both the first and last freezing
temperature. All three of these climatic anomalies must be prop-
erly treated to estimate freeze probabilities for Florida.
For those stations with freezing temperatures every year, the
dates of first and last freezing temperatures have been shown to
be random variables with a normal frequency distribution (6).
For stations where freezes do not occur each year, Thom (5) de-
veloped techniques to combine the series of freeze and freezeless
years. Vestal (7) developed a method for analyzing data for the
"cold season," i.e., from July 1 to June 30, and has shown that
data from such a season fit a normal distribution.


ANALYSIS OF WEATHER DATA

The Florida "cold season" was defined as the period from July
1 to June 30. The dates of the first and last occurrences of each
selected temperature were abstracted from the station records of
23 climatological and first order National Weather Service sta-
tions, mostly for the 30-year period 1941-1970. Most of the data
retrieval and computation was done by the National Oceanic and
Atmospheric Administration, Environmental Data Service, Na-
tional Climatic Center at Asheville, North Carolina. Separate com-
putations were made for five stations to compare the methods
used and verify the methods.
Eleven stations tested for randomness with a runs test were
all random at the 0.10 probability level. Selected data from spring
1










INTRODUCTION


Freeze probabilities are useful to Florida agricultural indus-
tries which produce cold sensitive crops as well as to homeowners
and engineers. The time interval between the 10% risk and the
50% risk of the first freezing temperature is 35 to 40 days, which
is much longer than in many states. This time interval is of im-
portance to farmers who often make decisions and long range
plans based on freeze probabilities.
The usual methods used in determining climatological probabili-
ties of the first and last occurrence of low temperatures are not
particularly applicable to Florida because there is not a well-
developed fall and spring season. For example, temperatures be-
low 32'F or 280F do not occur each year. Also, in seasons
where freezes do occur, freezing temperatures may not occur be-
fore January 1, while in other seasons the last freeze may occur
before December 31. Finally, there are some cold seasons in which
there is only one cold day which is both the first and last freezing
temperature. All three of these climatic anomalies must be prop-
erly treated to estimate freeze probabilities for Florida.
For those stations with freezing temperatures every year, the
dates of first and last freezing temperatures have been shown to
be random variables with a normal frequency distribution (6).
For stations where freezes do not occur each year, Thom (5) de-
veloped techniques to combine the series of freeze and freezeless
years. Vestal (7) developed a method for analyzing data for the
"cold season," i.e., from July 1 to June 30, and has shown that
data from such a season fit a normal distribution.


ANALYSIS OF WEATHER DATA

The Florida "cold season" was defined as the period from July
1 to June 30. The dates of the first and last occurrences of each
selected temperature were abstracted from the station records of
23 climatological and first order National Weather Service sta-
tions, mostly for the 30-year period 1941-1970. Most of the data
retrieval and computation was done by the National Oceanic and
Atmospheric Administration, Environmental Data Service, Na-
tional Climatic Center at Asheville, North Carolina. Separate com-
putations were made for five stations to compare the methods
used and verify the methods.
Eleven stations tested for randomness with a runs test were
all random at the 0.10 probability level. Selected data from spring
1










and fall freezes, (Lakeland WSO, Tampa WSO, St. Leo, Quincy,
and Bushnell) were tested at five different temperature levels for
goodness of fit to the Gaussian probability distribution. Twenty-
six of 32 data sets examined met the criteria for normal distri-
bution given in Lilliefors' (4) table of critical values for the
Kolmogorov-Smirnov goodness-of-fit test at the 0.1 level. With
these tests and in light of previous studies (1, 5, 7), it was con-
cluded that the data for each station could be treated as if they
were random and normally distributed.


COMPUTATION OF FREEZE PROBABILITY DATES

Each data series was analyzed to determine the climatological
probability that a preselected low temperature would occur before
a certain date and to determine the date, for a specified probabili-
ty level, when the low temperature would occur. The specified
probability level is the risk the user wants to take.
P(B) =P(B/A) x P(A)
Where: A=the threshold temperature.
B =the date upon which the first (last) event A oc-
curs.
P(B) =the probability that the first (last) event A will
occur before (after) the date B in the cold season.
P(A) =the probability that event A will occur sometime
during the cold season.
P(B/A) =the probability that the first (last) event A will
occur before (after) date B, given that event A
actually does occur.
P(A) can be estimated by letting P(A)=m/n where m is the
number of cold seasons during which the preselected temperature
was reached and n is the total number of seasons in the data
sample. The estimate of P(B/A) applies in general only for those
cases in which event B is contained in event A.
Since it was assumed that the preselected temperature thres-
holds were normally distributed, the cold season day number can
be defined by:
x =ts+Y.
Where: x =the cold season day number and is derived by as-
signing to each calendar date a unique number be-
ginning with July 1 as one, July 31 as 31, and so
on to June 30 as 365 (366 in leap years).
2









K= the mean day number of the first (last) occurrence
of the event A, calculated using the data for only
those years in which the event A actually occurs.
s=the standard deviation of the date of the first
(last) occurrence of the event A, calculated by the
formula =_m x2- (x)2
m(m-l)
t=the number of standard deviations from the mean
of the normally distributed dates of the event A.
The value of t can be obtained from a table of the
inverse normal probability distribution.

SAMPLE COMPUTATION

The record used for Tampa were the cold seasons from 1939-
1940 to 1969-1970. During this 31-year period, there were 23 cold
seasons with temperatures of 320F or lower, yielding a probability
estimate of cold temperatures, P(A), of 23/31 or 0.742. x was
calculated to be 182.696 (December 30th).
s= m x2- (yx)2 1/2
m(m-l)
s=27.08 days.
Assuming one wishes to know the latest date before there is a
25% chance of having a temperature as low as 320F, then

P(B/A) P(B) 0.25 =0.337
P(A) 0.742
From a table of the inverse normal distribution one finds that a
value of t=-0.42 is associated with the probability of 0.337.
These values are substituted to yield
x=ts+x, giving
x=(-0.42) (27.080)+182.696
x= -11.374+ 182.696=171.3 or December 18.
Thus there is a 25% chance of 320F or lower on or before Decem-
ber 18 at Tampa.

DISCUSSION OF RESULTS

The mean (50% probability) date of the first 320F in the cold
season ranges from November 20th in the Panhandle to Decem-
3









K= the mean day number of the first (last) occurrence
of the event A, calculated using the data for only
those years in which the event A actually occurs.
s=the standard deviation of the date of the first
(last) occurrence of the event A, calculated by the
formula =_m x2- (x)2
m(m-l)
t=the number of standard deviations from the mean
of the normally distributed dates of the event A.
The value of t can be obtained from a table of the
inverse normal probability distribution.

SAMPLE COMPUTATION

The record used for Tampa were the cold seasons from 1939-
1940 to 1969-1970. During this 31-year period, there were 23 cold
seasons with temperatures of 320F or lower, yielding a probability
estimate of cold temperatures, P(A), of 23/31 or 0.742. x was
calculated to be 182.696 (December 30th).
s= m x2- (yx)2 1/2
m(m-l)
s=27.08 days.
Assuming one wishes to know the latest date before there is a
25% chance of having a temperature as low as 320F, then

P(B/A) P(B) 0.25 =0.337
P(A) 0.742
From a table of the inverse normal distribution one finds that a
value of t=-0.42 is associated with the probability of 0.337.
These values are substituted to yield
x=ts+x, giving
x=(-0.42) (27.080)+182.696
x= -11.374+ 182.696=171.3 or December 18.
Thus there is a 25% chance of 320F or lower on or before Decem-
ber 18 at Tampa.

DISCUSSION OF RESULTS

The mean (50% probability) date of the first 320F in the cold
season ranges from November 20th in the Panhandle to Decem-
3









ber 30th in Central Florida and January 20th in South Florida.
Coastal sections south of Fort Pierce on the East Coast and
Punta Gorda on the West Coast have less than a 50% chance of
32F or lower in any year (Figures 1 and 2).
Since the standard deviation of the date of first (last) occur-
rence of some freezing temperature is 20 to 25 days, the interval
between the 10% risk and the 50% risk is 35 to 40 days. At those
stations where the temperature threshold is not reached every
year, this interval becomes even larger. The standard deviation
of the first 320F temperature averages 18 days in the Panhandle,
20 days in North Florida, and 25 days in South and Central
Florida.

USING THE PROBABILITY TABLES

Results of the statistical analysis of the first and last occur-
rence of preselected temperature thresholds are given in Tables
1 through 11. The probability that the indicated temperatures
will be reached sometime in the cold season is shown in Table 1.
Table 2 shows the dates by which there is a 10, 25, 50, 75, and
90% probability that 320F temperature will occur. Tables 3-6 are
similar tables for 28, 24, 20, and 160F. Tables 7-11 present the
last dates and probabilities for freezing temperatures. An index
and description of the stations used is this study is given in
Table 12.
The use of these tables can best be illustrated by examples. On
February 3rd a producer wishes to know the risk of cold damage
to sensitive crops such as tomatoes or peppers in the vicinity of
Lake Alfred and Fellsmere. By using Table 7, he can determine
there is a 50% probability of temperatures 320F or lower after
that date at Lake Alfred. Near Fellsmere, there is only a 25%
chance. If he waits until February 22, the risk of a 320F tempera-
ture will decrease to 25% at Lake Alfred and 10% chance at
Fellsmere. Rather than wait he may decide to plant hardy or
semi-hardy crops like cabbage or celeryVat Lake Alfred.
Figures 1 and 2 are maps of the mean date (50% probability)
of the first 320F in the season and the last 320F in the season.
These maps can assist users in determining the freeze risks at
locations other than those given in the tables. Similar maps can
be made for other temperature thresholds.
Figures 3 and 4 graphically portray the dates associated with
the various probabilities of the first or last 320F temperatures in
a cold season for Clermont. The probability for any date or the















12/20


FIGURE 1.


MEAN DATE (50% PROBABILITY) OF FIRST
(320F) FREEZE OCCURRENCE.


3/1


r,j .Jo'


FIGURE 2. MEAN DATE (50%7 PROBABILITY) OF LAST
(320F) FREEZE OCCURRENCE.






Jan 29 Dec 30


Jan 19 Jan 09


Jan 09- Jan 19


Dec 30 Jan 29


Dec 20 Feb 08


Dec 10- Feb I8


Nov 30- Feb 28


Nov 20 Mar 10


Nov 10 Mar 20 -


Nov O i Mar 3
0.01 0.05 02 0.5 2 5 10 20 30 40 50 60 70 80 90 aoI 0.05 02 0.5 I 2 5 10 20 30 40 50 60 70 80 90
PERCENT PROBABILITY OF (2*F) FREEZE PERCENT PROBABILITY OF (32-F) FREEZE
BEFORE GIVEN DATE IN FALL AFTER GIVEN DATE IN SPRING


FIGURE 4. STATION: CLERMONT.


FIGURE 3. STATION: CLERMONT.









date corresponding to any specific probability can be determined
from these figures.
The days (dates) in the cold season correspond to the horizontal
lines and are plotted at regular spaced intervals along the left
side of the graph. For estimating the probabilities of the first
freeze, the days are plotted so that the earlier dates are at the
bottom of the graph and the later dates towards the top. The
reverse is done for the last freeze. The vertical lines labeled with
numbers along the bottom of the graph correspond to various
probability levels (e.g. 10 indicating 10 % probability, 70 indi-
cating 70% probability).
From Table 2, the date corresponding to 10% probability at
Clermont for 320F temperature is December 1. This date is
plotted in Figure 3 at the intersection with line for December 1.
This intersection is marked with a small circle. Using another
probability value for Clermont, the procedure is repeated. For ex-
ample, the date corresponding to the 50% probability is Decem-
ber 31. This is plotted as a second circle. These points are con-
nected with a straight line extending to the bottom and top of
the graph. Any desired probability can now be estimated from
this line. From Figure 3, it is estimated that there is a 5% prob-
ability that the first 320F temperature will occur on or before
November 22nd and a 70% probability it will occur on or before
January 14th. In certain cases, when a preselected temperature
does not occur every year, a given probability may not exist.
A similar graph can be made for the probability of a spring
freeze after a given date. The graph for Clermont showing the
probability of receiving the last 320F after a given date (Figure
4) was constructed with the latest date at the bottom of the
graph. The 50% probability is January 17. A circle is placed at
the intersection of the 50% probability at January 17. The 10%
probability is February 26, which is similarly plotted on the
graph. A straight line is drawn through these points from the
top to the bottom of the graph. A vertical line is drawn at the
90% value since Table 1 shows that the probabilities beyond do
not exist.
The paper used for these graph is special probability paper.
On this paper, the so-called "normal probability curve" appears
as a straight line. If one wishes to estimate probabilities for other
locations it can be done by plotting the tabular data on probability
paper.









LIMITATIONS


The minimum temperatures used in this study were observed
in standard-type louvered shelters at a height of 5 feet above
the ground. (Lakeland WSO was an exception.) Although the
sites for weather stations are chosen to be as representative as
possible, local conditions do affect the data. Topography is the
most significant influencing factor (1). Cold air drainage at night
from the top to the bottom of a slope can produce significantly
different minimum temperatures at adjacent sites. This factor
assumes an added dimension in Florida where many of the freezes
are of the radiation-type with little cold advection and near calm
conditions. Minimum temperature differences as much as 18F
have been noted between high and low ground locations less than
one-quarter mile apart in the central section of the peninsula (3).
Consideration must be given in relating shelter temperatures to
ground temperatures. Studies at the University of Vermont (2)
show that daily minimum temperatures at the 3-inch level aver-
age 40F below those at the 5-foot level in the spring and fall
seasons. Thus the shelter temperature will have a different mean-
ing to the citrus grower than to the truck farmer.
Data from Lakeland WSO illustrate the effect of observation
height on temperatures. The site at Lakeland is located at the
City Hall with thermometers some 56 feet above the ground level.
Table 2 shows that for this station the mean date of the first
320F occurrence is January 24. This date is almost a month later
than the mean date for surrounding stations. This station also
shows an average of two days per year on which the temperature
falls to 320F, or below; the duration of temperatures in this range
averages 11 hours per year. At a station located on the northern
edge of the city, (Lakeland No. 2), where the thermometers are
exposed at a standard height of 5 feet above ground level, similar
values are 7 days and 35 hours respectively.

CONCLUSIONS

The usefulness of these probability tables and graphs is mainly
for planning agricultural and agribusiness operations. Individual
users must make subjective corrections for differences in topog-
raphy, for nearness to urban locations or bodies of water, and for
temperatures at other than the standard level. Finally, during
any particular cold season, day-to-day operational decisions
should be based on the latest weather forecasts and outlooks.









LIMITATIONS


The minimum temperatures used in this study were observed
in standard-type louvered shelters at a height of 5 feet above
the ground. (Lakeland WSO was an exception.) Although the
sites for weather stations are chosen to be as representative as
possible, local conditions do affect the data. Topography is the
most significant influencing factor (1). Cold air drainage at night
from the top to the bottom of a slope can produce significantly
different minimum temperatures at adjacent sites. This factor
assumes an added dimension in Florida where many of the freezes
are of the radiation-type with little cold advection and near calm
conditions. Minimum temperature differences as much as 18F
have been noted between high and low ground locations less than
one-quarter mile apart in the central section of the peninsula (3).
Consideration must be given in relating shelter temperatures to
ground temperatures. Studies at the University of Vermont (2)
show that daily minimum temperatures at the 3-inch level aver-
age 40F below those at the 5-foot level in the spring and fall
seasons. Thus the shelter temperature will have a different mean-
ing to the citrus grower than to the truck farmer.
Data from Lakeland WSO illustrate the effect of observation
height on temperatures. The site at Lakeland is located at the
City Hall with thermometers some 56 feet above the ground level.
Table 2 shows that for this station the mean date of the first
320F occurrence is January 24. This date is almost a month later
than the mean date for surrounding stations. This station also
shows an average of two days per year on which the temperature
falls to 320F, or below; the duration of temperatures in this range
averages 11 hours per year. At a station located on the northern
edge of the city, (Lakeland No. 2), where the thermometers are
exposed at a standard height of 5 feet above ground level, similar
values are 7 days and 35 hours respectively.

CONCLUSIONS

The usefulness of these probability tables and graphs is mainly
for planning agricultural and agribusiness operations. Individual
users must make subjective corrections for differences in topog-
raphy, for nearness to urban locations or bodies of water, and for
temperatures at other than the standard level. Finally, during
any particular cold season, day-to-day operational decisions
should be based on the latest weather forecasts and outlooks.




















TABLE 1. PROBABILITY THAT INDICATED TEMPERATURE
THRESHOLD IS REACHED.*
Threshold
Station 320F 280F 240F 200F 160F

Avon Park 0.833 0.400 0.067 0.033 0
Bushnell 1.000 0.933 0.733 0.200 0.067
Carrabelle 1.000 0.900 0.767 0.467 0.133
Chipley 1.000 1.000 0.900 0.733 0.233
Clermont 0.900 0.467 0.100 0.033 0
Cross City 1.000 1.000 0.900 0.600 0.167
Everglades 0.333 0.033 0 0 0
Fellsmere 0.767 0.533 0.067 0 0
Fernandina Beach 0.967 0.733 0.433 0.133 0.033
Homestead 0.700 0.133 0 0 0
Jacksonville Beach 0.967 0.833 0.367 0.067 0
Lake Alfred 0.967 0.700 0.200 0.033 0.033
Lake City 1.000 0.967 0.833 0.433 0.067
Lakeland WSO 0.679 0.250 0.033 0.033 0
Madison 1.000 0.933 0.833 0.433 0.133
Milton 1.000 1.000 0.900 0.633 0.267
Monticello 1.000 0.933 0.900 0.667 0.233
Palatka 0.967 0.700 0.333 0.033 0.033
Quincy 1.000 0.968 0.903 0.548 0.226
St. Leo 0.833 0.567 0.167 0.033 0
Tallahassee WSO 1.000 0.900 0.633 0.400 0.226
Tampa WSO 0.742 0.290 0.161 0.033 0
Titusville 0.967 0.667 0.233 0 0
* P [A]= m/n
m = number of cold seasons threshold reached.
n = total number of seasons of record.




















TABLE 2. FALL DATES BEFORE WHICH THE FIRST TEMPERA-
TURE OF 320F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities
Station 0.10 0.25 0.50 0.75 0.90
Avon Park Dec. 3 Dec. 20 Jan. 9 Feb. 4 *
Bushnell Nov. 7 Nov. 21 Dec. 6 Dec. 22 Jan. 5
Carrabelle Oct. 30 Nov. 15 Dec. 3 Dec. 20 Jan. 5
Chipley Oct. 26 Nov. 4 Nov. 15 Nov. 25 Dec. 5
Clermont Dec. 1 Dec. 15 Dec. 31 Jan. 18
Cross City Oct. 23 Nov. 7 Nov. 24 Dec. 10 Dec. 25
Everglades Dec. 24 Jan. 25 *
Fellsmere Dec. 7 Dec. 24 Jan. 14 Feb. 24
Fernandina Beach Nov. 14 Nov. 26 Dec. 10 Dec. 24 Jan. 7
Homestead Dec. 15 Jan. 1 Jan. 22 *
Jacksonville Beach Nov. 16 Nov. 30 Dec. 15 Dec. 31 Jan. 16
Lake Alfred Nov. 24 Dec. 10 Dec. 27 Jan. 14 Feb. 1
Lake City Nov. 4 Nov. 15 Nov. 28 Dec. 11 Dec. 22
Lakeland WSO Dec. 13 Dec. 31 Jan. 24 *
Madison Nov. 5 Nov. 14 Nov. 24 Dec. 4 Dec. 13
Milton Oct. 26 Nov. 6 Nov. 18 Nov. 30 Dec. 10
Monticello Oct. 31 Nov. 11 Nov. 23 Dec. 5 Dec. 16
Palatka Nov. 17 Nov. 30 Dec. 14 Dec. 29 Jan. 13
Quincy Oct. 30 Nov. 9 Nov. 20 Dec. 1 Dec. 11
St. Leo Nov. 18 Dec. 6 Dec. 27 Jan. 17
Tallahassee WSO Oct. 28 Nov. 8 Nov. 21 Dec. 3 Dec. 14
Tampa WSO Nov. 30 Dec. 18 Jan. 11 *
Titusville Nov. 29 Dec. 15 Jan. 2 Jan. 20 Feb. 8
* The chances are less than the indicated probability that a temperature of
32F or lower will occur.
** Fewer than five occurrences.




















TABLE 3. FALL DATES BEFORE WHICH THE FIRST TEMPER-
ATURE OF 280F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities

Station 0.10 0.25 0.50 0.75 0.90
Avon Park Dec. 26 Jan. 14 *
Bushnell Nov. 20 Dec. 6 Dec. 24 Jan. 13 Feb. 6
Carrabelle Nov. 7 Nov. 21 Dec. 7 Dec. 26
Chipley Oct. 31 Nov. 13 Nov. 29 Dec. 14 Dec. 28
Clermont Dec. 15 Jan. 7 *
Cross City Nov. 4 Nov. 22 Dec. 12 Jan. 1 Jan. 19
Everglades ** ** ** ** **
Fellsmere Dec. 19 Jan. 6 Feb. 12 *
Fernandina Beach Dec. 3 Dec. 21 Jan. 12 *
Homestead ** ** ** **
Jacksonville Beach Dec. 4 Dec. 19 Jan. 6 Jan. 30 *
Lake Alfred Dec. 10 Dec. 26 Jan. 20 *
Lake City Nov. 10 Nov. 24 Dec. 10 Dec. 27 Jan. 13
Lakeland WSO Jan. 2 *
Madison Nov. 14 Nov. 28 Dec. 14 Jan. 2 Jan. 24
Milton Oct. 31 Nov. 13 Nov. 28 Dec. 13 Dec. 26
Monticello Nov. 5 Nov. 18 Dec. 3 Dec. 19 Jan. 8
Palatka Dec. 4 Dec. 23 Jan. 19 *
Quincy Nov. 12 Nov. 24 Dec. 7 Dec. 21 Jan. 4
St. Leo Dec. 16 Jan. 6 Feb. 11 *
Tallahassee WSO Nov. 2 Nov. 22 Dec. 14 Jan. 8
Tampa WSO Jan. 1 Feb. 8 *
Titusville Dec. 16 Jan. 2 Jan. 26 *
* The chances are less than the indicated probability that a temperature of
28F or lower will occur.
** Fewer than five occurrences.




















TABLE 4. FALL DATES BEFORE WHICH THE FIRST TEMPERA-
TURE OF 240F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities
Station 0.10 0.25 0.50 0.75 0.90
Avon Park ** ** ** ** **
Bushnell Dec. 6 Dec. 24 Jan. 16 *
Carrabelle Nov. 27 Dec. 14 Jan. 3 Feb. 12
Chipley Nov. 17 Dec. 2 Dec. 20 Jan.8 *
Clermont ** ** ** ** **
Cross City Nov. 27 Dec. 13 Jan. 1 Jan. 21
Everglades ** ** ** ** **
Fellsmere ** ** ** ** **
Fernandina Beach Dec. 17 Jan.13 *
Homestead ** ** ** ** **
Jacksonville Beach Dec. 26 Jan.25 *
Lake Alfred Jan. 13 *
Lake City Nov. 29 Dec. 16 Jan. 5 Jan. 31
Lakeland WSO ** ** ** ** **
Madison Nov. 26 Dec. 14 Jan.5 Feb. 3 *
Milton Nov. 12 Nov. 28 Dec. 16 Jan.6 *
Monticello Nov. 19 Dec. 4 Dec. 22 Jan. 10 *
Palatka Dec. 23 Jan. 29 *
Quincy Nov. 21 Dec. 8 Dec. 27 Jan.18 Mar. 6
St. Leo Jan. 16 *
Tallahassee WSO Nov. 15 Dec. 9 Jan. 14 *
Tampa WSO ** ** ** ** **
Titusville Dec. 31 *
* The chances are less than the indicated probability that a temperature of
24F or lower will occur.
** Fewer than five occurrences.




















TABLE 5. FALL DATES BEFORE WHICH THE FIRST TEMPERA-
TURE OF 200F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.

Probabilities
Station 0.10 0.25 0.50 0.75 0.90
Avon Park ** ** ** **
Bushnell Jan. 10 *
Carrabelle Dec. 26 Jan. 18 *
Chipley Dec. 12 Dec. 28 Jan. 19 *
Clermont ** ** ** ** *E
Cross City Dec. 12 Dec. 31 Jan. 29 *
Everglades ** ** ** **
Fellsmere ** ** ** ** *
Fernandina Beach ** ** ** ** *
Homestead ** ** ** **
Jacksonville Beach ** ** ** **
Lake Alfred ** ** ** ** **
Lake City Dec. 15 Jan. 11 *
Lakeland WSO ** ** ** *
Madison Dec. 19 Jan. 13 *
Milton Dec. 9 Dec. 27 Jan. 23 *
Monticello Dec. 9 Dec. 28 Jan. 23 *
Palatka *' ** ** ** -
Quincy Dec. 6 Dec. 27 Feb. 2 *
St. Leo ** ** ** ** **
Tallahassee WSO Dec. 18 Jan. 7 *
Tampa WSO ** ** ** ** **
Titusville :* ** ** ** **
* The chances are less than the indicated probability that a temperature of
20 F or lower will occur.
** Fewer than five occurrences.





















TABLE 6. FALL DATES BEFORE WHICH THE FIRST TEMPERA-
TURE OF 160F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities
Station 0.10 0.25 0.50 0.75 0.90
Avon Park ** ** ** **
Bushnell ** ** ** ** **
Carrabelle ** ** ** ** **
Chipley Jan.7 *
Clermont ** ** ** ** **
Cross City Dec.24 *
Everglades ** ** ** ** **
Fellsmere ** ** ** ** **
Fernandina Beach ** ** ** ** **
Homestead ** ** ** ** **
Jacksonville Beach ** ** ** ** **
Lake Alfred ** ** ** ** **
Lake City ** ** ** ** **
Lakeland WSO ** ** ** ** **
Madison ** ** ** ** **
Milton Jan. 4 Feb. 9 *
Monticello Dec. 29 *
Palatka ** ** ** ** **
Quincy Jan. 7 ** *
St. Leo ** ** ** ** **
Tallahassee WSO N/A *
Tampa WSO ** ** ** ** **
Titusville ** ** ** ** **
* The chances are less than the indicated probability that a temperature of
16F or lower will occur.
** Fewer than five occurrences.




















TABLE 7. SPRING DATES AFTER WHICH THE LAST TEMPERA-
TURE OF 320F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities
Station 0.90 0.75 0.50 0.25 0.10
Avon Park Dec. 16 Jan.14 Feb.5 Feb.24
Bushnell Jan. 18 Feb. 4 Feb. 23 Mar. 14 Mar. 31
Carrabelle Jan. 29 Feb. 13 Mar. 1 Mar. 17 Apr. 1
Chipley Feb. 5 Feb. 17 Mar. 3 Mar. 17 Mar. 30
Clermont Dec. 25 Jan. 17 Feb. 8 Feb.26
Cross City Feb. 11 Feb. 25 Mar. 12 Mar. 27 Apr. 9
Everglades Dec. 25 Jan. 26
Fellsmere Nov. 28 Jan. 12 Feb. 4 Feb. 23
Fernandina Beach Dec. 30 Jan. 21 Feb. 12 Mar. 5 Mar. 24
Homestead Jan. 2 Jan. 24 Feb. 10
Jacksonville Beach Dec. 31 Jan. 19 Feb. 8 Feb. 26 Mar. 15
Lake Alfred Dec. 28 Jan. 16 Feb. 3 Feb. 21 Mar. 8
Lake City Feb. 7 Feb. 19 Mar. 5 Mar. 18 Mar. 30
Lakeland WSO Jan. 7 Jan. 21 Feb. 15
Madison Jan. 28 Feb. 11 Feb. 26 Mar. 13 Mar. 26
Milton Feb. 9 Feb. 20 Mar. 5 Mar. 18 Mar. 30
Monticello Jan. 31 Feb. 14 Mar. 3 Mar. 20 Apr. 4
Palatka Dec. 27 Jan. 14 Feb. 1 Feb. 19 Mar. 7
Quincy Feb. 8 Feb. 21 Mar. 7 Mar. 21 Apr. 3
St. Leo Jan. 2 Jan. 23 Feb. 8 Feb. 21
Tallahassee WSO Feb. 4 Feb. 18 Mar. 7 Mar. 23 Apr. 6
Tampa WSO Jan. 12 Feb. 3 Feb. 19
Titusville Dec. 20 Jan. 10 Feb. 1 Feb. 21 Mar. 11
* The chances are less than the indicated probability that a temperature of
32 F or lower will occur.
** Fewer than five occurrences.




















TABLE 8. SPRING DATES AFTER WHICH THE LAST TEMPERA-
TURE OF 280F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities

Station 0.90 0.75 0.50 0.25 0.10
Avon Park Jan. 10 Jan. 29
Bushnell Dec. 12 Jan. 11 Feb. 3 Feb. 25 Mar. 16
Carrabelle Jan. 29 Feb. 15 Mar. 1 Mar. 14
Chipley Jan. 17 Feb. 1 Feb. 18 Mar. 7 Mar. 22
Clermont Jan. 14 Feb. 5
Cross City Jan. 13 Jan. 29 Feb. 15 Mar. 5 Mar. 21
Everglades *
Fellsmere Dec. 18 Jan. 19 Feb. 4
Fernandina Beach Jan. 19 Feb. 8 Feb. 23
Homestead ** ** ** *
Jacksonville Beach Dec. 27 Jan. 19 Feb. 5 Feb. 19
Lake Alfred Jan. 12 Feb. 1 Feb. 16
Lake City Jan. 16 Feb. 1 Feb. 17 Mar. 5 Mar. 18
Lakeland WSO Jan. 26
Madison Dec. 24 Jan. 16 Feb. 4 Feb. 21 Mar. 9
Milton Jan. 31 Feb. 11 Feb. 23 Mar. 7 Mar. 18
Monticello Jan. 4 Jan. 27 Feb. 14 Mar. 3 Mar. 18
Palatka Jan. 6 Jan. 30 Feb. 18
Quincy Jan. 4 Jan.23 Feb.11 Feb.28 Mar. 16
St. Leo Dec. 21 Jan. 24 Feb. 13
Tallahassee WSO Dec. 30 Jan. 29 Feb. 26 Mar. 21
Tampa WSO Dec. 23 Jan. 28
Titusville Jan. 1 Jan. 26 Feb. 12
* The chances are less than the indicated probability that a temperature of
28F or lower will occur.
** Fewer than five occurrences.




















TABLE 9. SPRING DATES AFTER WHICH THE LAST TEMPERA-
TURE OF 240F OR LOWER WILL OCCUR FOR SE-
LECTED PROBABILITIES.
Probabilities
Station 0.90 0.75 0.50 0.25 0.10
Avon Park ** ** ** ** **
Bushnell Jan.11 Feb. 2 Feb. 20
Carrabelle Nov. 30 Jan. 12 Feb. 4 Feb. 22
Chipley Dec. 30 Jan. 18 Feb. 5 Feb. 20
Clermont ** ** ** :* **
Cross City Jan. 6 Jan. 22 Feb. 6 Feb. 19
Everglades ** ** ** ** **
Fellsmere ** ** ** ** **
Fernandina Beach Jan. 20 Feb. 10
Homestead ** ** ** ** **
Jacksonville Beach Jan. 15 Feb. 11
Lake Alfred Jan. 14
Lake City Dec. 25 Jan. 17 Feb. 3 Feb. 18
Lakeland WSO ** ** ** ** **
Madison Dec. 19 Jan.14 Feb. 3 Feb. 19
Milton Jan. 8 Jan. 28 Feb. 15 Mar. 2
Monticello Dec. 30 Jan. 22 Feb. 11 Mar. 1
Palatka Jan. 4 Feb. 3
Quincy Oct. 28 Dec. 21 Jan. 15 Feb. 6 Feb. 25
St. Leo Jan. 7
Tallahassee WSO Jan. 15 Feb. 10 Feb. 28
Tampa WSO ** ** ** ** **
Titusville ** ** ** ** **
* The chances are less than the indicated probability that a temperature of
24F or lower will occur.
** Fewer than five occurrences.




















TABLE 10.


SPRING DATES AFTER WHICH THE LAST TEM-
PERATURE OF 200F OR LOWER WILL OCCUR FOR
SELECTED PROBABILITIES.


Probabilities

Station 0.90 0.75 0.50 0.25 0.10
Avon Park ** ** ** ** **
Bushnell Jan. 11
Carrabelle Jan.24 Feb.9
Chipley Jan. 10 Jan.29 Feb. 13
Clermont ** ** ** ** **
Cross City Dec. 31 Jan. 23 Feb. 7
Everglades ** ** ** ** **
Fellsmere ** ** ** ** **
Fernandina Beach ** ** **
Homestead ** ** ** ** **
Jacksonville Beach ** ** ** ** **
Lake Alfred ** ** ** ** **
Lake City Jan. 15 Feb. 6
Lakeland WSO ** ** ** ** **
Madison Jan. 20 Feb. 5
Milton Jan. 7 Jan. 26 Feb. 8
Monticello Jan. 11 Jan. 29 Feb. 12
Palatka ** ** ** ** **
Quincy Dec. 7 Jan.15 Feb.4
St. Leo ** ** ** ** **
Tallahassee WSO Jan. 9 Feb. 2
Tampa WSO ** ** ** ** **
Titusville ** ** ** **
* The chances are less than the indicated probability that a temperature of
20 F or lower will occur.
** Fewer than five occurrences.






















TABLE 11.


SPRING DATES AFTER WHICH THE LAST TEM-
PERATURE OF 160F OR LOWER WILL OCCUR FOR
SELECTED PROBABILITIES.


Probabilities

Station 0.90 0.75 0.50 0.25 0.10
Avon Park ** ** ** ** **
Bushnell ** ** ** ** **
Carrabelle ** ** ** ** **
Chipley Jan. 20
Clermont ** ** ** ** **
Cross City Dec. 26
Everglades ** ** ** ** **
Fellsmere ** ** ** ** **
Fernandina Beach ** ** ** **
Homestead ** ** ** ** **
Jacksonville Beach ** ** ** ** **
Lake Alfred ** ** ** ** **
Lake City ** ** ** ** **
Lakeland WSO ** ** ** ** **
Madison ** ** ** **
Milton Dec. 23 Jan. 25
Monticello Jan. 18
Palatka ** ** ** ** **
Quincy Jan.20
St. Leo ** ** ** ** **
Tallahassee WSO NA
Tampa WSO ** ** ** **
Titusville ** ** ** ** **


* The chances are less than the
16F or lower will occur.
** Fewer than five occurrences.


indicated probability that a temperature of




Station
NORTH
Cross City 2WNW

Fernandina Beach

Jacksonville Beach
Lake City 2E
Madison

Palatka

Tallahassee WSO

NORTHWEST
Carrabelle 1NNW
Chipley 3E


Milton Exp. Stn.
Monticello 3W

Quincy 3SSW


County Elevation (ft.)


Dixie

Nassau

Duval
Columbia
Madison

Putnam

Leon



Franklin
Washington

Santa Rosa
Jefferson

Gadsden


42 Over sod on grounds of Florida Forest Service district
office. Good exposure.
25 Over sod at city water plant. Surrounded by low to
medium height shrubs. Fair exposure.
10 Over sod at Fire Dept. Good exposure.
195 Over sod in fenced enclosure. Excellent exposure.
190 Shelter over sod near observer's residence. Good ex-
posure. SRG over dirt. Excellent exposure.
20 Over sod at Fire Dept. Only a few feet away from as-
phalt parking lot. Fair exposure.
55 Over sod in low depression. Poor exposure. Much too
cold for area.

10 Over sod near observer's residence. Fair exposure.
130 Over sod at U of F Exp. Stn. Heavily wooded in gen-
eral area of shelter. Fair exposure.
217 Over sod in fenced enclosure. Excellent exposure.
148 Over sod at U of F Exp. Stn. Excellent exposure.
Ground elevation drops 15 feet 600' north.
245 Over sod in fenced enclosure. Excellent exposure. Lo-
cated at U of F Exp. Stn.


Remarks





NORTHCENTRAL
Bushnell 2E
Clermont 6SSE


St. Leo


Titusville 3NW

SOUTHCENTRAL
Avon Park

Fellsmere 7SSW


Lake Alfred Exp. Stn.
Lakeland WSO


Tampa WSO
EVERGLADES &
COAST
Everglades

LOWER EAST
COAST
T~inestead Exp. Stn.


Sumter
Lake


Pasco


Brevard



Highlands


Indian River


Polk
Polk


Hillsborough



Collier


Dade


75
125


190


30



149


20


145
214


19


Over sod in observer's back yard. Excellent exposure.
Over thin sod very near observer's large citrus grove.
Good exposure.
Over sod at St. Leo College. Surrounding area very
hilly.
Over very thin sod near observer's residence. Heavily
wooded with fair exposure.


Over sod across street from Fire Dept. Excellent ex-
posure.
Over sod at Diamond G Farm. Completely open with
excellent exposure.
Over sod in fenced enclosure. Excellent exposure.
On roof of City Hall, in immediate downtown area.
Very poor exposure.
Over sod on airport property. Excellent exposure.


5 Over hardpan and coral rock. Chokoloskee Bay 20 feet
to the south. Good exposure for this area.



11 Over sod in fenced enclosure. Excellent exposure.









LITERATURE CITED


1. Butson, K. D. and J. F. Gerber. 1964. Temperature hazards to peaches
in Florida. Proc. Fla. State Hort. Soc. 77:395-401.
2. Hopp, R. J., K. E. Varney, and R. E. Lautzenheiser. 1964. Late spring
and early fall low temperatures in Vermont. Univ. Vt. Agr. Exp.
Stat. Bull. 639, 23 pp.
3. Johnson, W. O. 1970. Minimum temperatures in the agricultural areas
of peninsular Florida, Summary of seasons 1937-67. Univ. Fla. Inst.
Food Agr. Sci. Public. 9, 154 pp.
4. Lilliefors, H. W. 1967. On the Kolmogorov-Smirnov test for normality
with mean and variance unknown. Amer. Stat. Assoc. 62(318):
399-402.
5. Thom, H. C. S. 1959. The distribution of freeze-date and freeze-free
period for climatological series with freezeless years. Monthly
Weather Rev. 87(4):136-144.
6. Thom, H. C. S., and R. H. Shaw. 1958. Climatological analysis of freeze
data for Iowa. Monthly Weather Rev. 86(7):251-257.
7. Vestal, C. K. 1971. First and last occurrence of low temperatures during
cold season. Monthly Weather Rev. 99(8):650-652.








ACKNOWLEDGMENTS

The author thanks W. O. Johnson and J. Georg of the Lakeland
Weather Service Office for many fruitful discussions and T.
Clarke, the National Weather Service Substation Network Spe-
cialist, for compiling Table 12. Finally, we acknowledge our debt
to the many dedicated and public-spirited Cooperative Weather
Observers, whose daily observations form the basis for this work.









LITERATURE CITED


1. Butson, K. D. and J. F. Gerber. 1964. Temperature hazards to peaches
in Florida. Proc. Fla. State Hort. Soc. 77:395-401.
2. Hopp, R. J., K. E. Varney, and R. E. Lautzenheiser. 1964. Late spring
and early fall low temperatures in Vermont. Univ. Vt. Agr. Exp.
Stat. Bull. 639, 23 pp.
3. Johnson, W. O. 1970. Minimum temperatures in the agricultural areas
of peninsular Florida, Summary of seasons 1937-67. Univ. Fla. Inst.
Food Agr. Sci. Public. 9, 154 pp.
4. Lilliefors, H. W. 1967. On the Kolmogorov-Smirnov test for normality
with mean and variance unknown. Amer. Stat. Assoc. 62(318):
399-402.
5. Thom, H. C. S. 1959. The distribution of freeze-date and freeze-free
period for climatological series with freezeless years. Monthly
Weather Rev. 87(4):136-144.
6. Thom, H. C. S., and R. H. Shaw. 1958. Climatological analysis of freeze
data for Iowa. Monthly Weather Rev. 86(7):251-257.
7. Vestal, C. K. 1971. First and last occurrence of low temperatures during
cold season. Monthly Weather Rev. 99(8):650-652.








ACKNOWLEDGMENTS

The author thanks W. O. Johnson and J. Georg of the Lakeland
Weather Service Office for many fruitful discussions and T.
Clarke, the National Weather Service Substation Network Spe-
cialist, for compiling Table 12. Finally, we acknowledge our debt
to the many dedicated and public-spirited Cooperative Weather
Observers, whose daily observations form the basis for this work.






A JUW 7W


This public document was promulgated at a cost of $688.20, or
34 cents per copy to provide information on climatological
probabilities of occurrence of freezing temperatures in Florida.


All programs and related activities sponsored or assisted by the Florida
Agricultural Experiment Stations are open to all persons regardless of race,
:olor, national origin, age, sex, or handicap.


I TEACHING RESEARCH EXTENSI




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs