Front Cover
 Cold protection by heating
 Cold protection with wind...
 Cold protection by overhead...
 Back Cover

Title: Protecting citrus from cold damage
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00084317/00001
 Material Information
Title: Protecting citrus from cold damage
Series Title: Protecting citrus from cold damage
Physical Description: Book
Creator: Gerber, John Francis,
Publisher: Agricultural Extension Service, University of Florida
 Record Information
Bibliographic ID: UF00084317
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 79089220

Table of Contents
    Front Cover
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
    Cold protection by heating
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
    Cold protection with wind machines
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
    Cold protection by overhead sprinklers
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
    Back Cover
        Page 30
Full Text






JUNE, 1965

Dr. J. F. Gerber is Assistant Climatologist, Department
of Fruit Crops, Florida Agricultural Experiment Stations;
J. D. Martsolf is Assistant Extension Horticulturist, De-
partment of Fruit Crops, Florida Agricultural Extension

Protecting Citrus from Cold Damage

Reliable, accurate records of cold
weather in Florida have been col-
lected since 1937. These records,
even for this short period of time,
show that cold winter weather is a
constant and ever-present threat to
the state's citrus. Parts of the pen-
insula are relatively free from cold
weather; however, at least half of
the citrus producing area can expect
some damagingly low temperatures
during any one particular season.
Cold weather in Florida. Low
temperature in Florida results from
several simultaneous occurrences.
The most disastrous cold waves oc-
cur when a massive high-pressure
area sweeps out of the Arctic, across

the mid North American continent
and into the southeastern United
States, while at the same time, a
low pressure area or storm is moving
northeast along the Atlantic coast.
The counter clockwise circulation
around the center of the low pres-
sure combines to form a wind flow
pattern that funnels cold air from
the Arctic and Canada into Florida,
as shown in Fig 1.
Fortunately the State is oriented
so that only on rare occasions does
cold, dry, Arctic air invade the state
without first flowing across the Gulf
of Mexico or the Atlantic Ocean.
These two large bodies of water are
warmer than the Arctic air and

Fig. 1 The clockwise circulation of air around the center of high pressure
(H) and the counter-clockwise circulation around the center of low pres-
sure (L) combine to form a wind flow pattern that funnels cold air from
the Arctic and Canada into Florida.

Fig. 2 A grove enclosed in an imaginary box. The arrow through the sides
represent heat loss by cold wind advectionn), the arrow through the top
represents heat loss by radiation.

modify its temperature. It is only
when air passes into the state with-
out flowing over these bodies of
water that extremely low tempera-
tures occur.
After the state is overrun by cold,
dry air, the skies clear and large
quantities of heat are lost to the sky
by radiation. Radiant energy is in
all ways similar to the light from a
bulb or heat from the glowing stack
of a grove heater. Cold weather
and low temperatures in Florida re-
sult from these two conditions:
Cold, dry air overrunning the
state displacing the warm
moist, maritime air from ad-
jacent water, and
rapid loss of heat by radiation
at night.
Heat balance and cold protec-
tion. In order to study gains and
losses of heat from a grove, it is use-
ful to visualize the grove enclosed in
an imaginary rectangular box shown
in Fig. 2. The arrow entering the
side represents advection or over-
running of the grove by cold, dry
air from the continental United
This cold air absorbs heat as it

passes through the grove and be-
comes warmer. This warmer air is
indicated by the arrow leaving the
grove. As the air moves southward,
it is progressively warmed by heat
from the surface. At night heat
passes through the top of the box
as radiant energy and is lost to the
cold sky. During the daytime, heat
is gained as radiant energy from the
sun, and this warming reduces the
duration and severity of cold weath-
er in the state.
For ideal cold protection, as
much heat must be supplied to the
imaginary box at night as is lost to
the wind and by radiation. If all
the heat supplied by cold protection
devices were retained inside the
imaginary box the efficiency would
be very high; but this is impossible.
The heat produced by burning fuel
warms the air, which may rise out
of the grove through the top of the
box. Heat may also be lost through
the sides by horizontal wind move-
Still more heat is lost when radia-
tion from heaters passes directly to
the sky without warming the trees.
When the total heat lost is balanced
against the heat supplied, it is

often found that not more than 35
percent of the heat is utilized for
cold protection.
Cold protection requires a favor-
able balance between heat loss and
heat gain. If heat loss can be off-
set by heat gain from cold protec-
tion devices, then protection will
be successful. Large quantities of
money, equipment, and labor may
be involved, so it is desirable that
the system succeed, even under se-
vere conditions.
Following freezes with wide-
spread damage, the value of citrus
increases markedly. Cold protec-
tion under such conditions offers
the grower the best opportunity of
larger economic returns. Cold pro-
tection systems should be designed
to provide protection at least to the
trees, if not to the fruit, under the
most severe conditions anticipated.
Measuring the microclimate
and cold protection effects. The
United States Weather Bureau and
the Florida Agricultural Experiment
Stations forecast minimum temper-
atures for all cold nights in Florida.
Forecasts are for air temperature in
shelters such as shown in Figures
3 and 5.
A grove may demonstrate indi-
vidual climatic characteristics that
are different from adjacent areas.
These differences which occur over
short distances are due to micro-
climatic effects. The forecasts can
alert an individual to the possibility
of cold weather, but the exact de-
gree of cold weather and exact
temperatures must be determined in
each individual location.
Most growers are well aware of
the effects that elevation, proximity

to water and wind have upon
temperature at individual sites.
Elevation differences produce local
temperature variations especially on
calm nights. On most nights, hill-
tops will be warmer than adjacent
lower areas. Cold air tends to col-
lect in depressions and pockets since
it is heavier than warm air, and
since light winds on many cold
nights do not penetrate into the
pockets. Broad flat areas usually
are cold regardless of the elevation.
The temperature is usually uni-
form on windy nights. If it is se-
verely cold and very windy, hill-
tops-which are normally warmer
than adjacent low areas-will be as
cold or colder. Most windy nights
will be warmer than if there is little
Areas immediately on the down-
wind side of lakes are warmer than
other adjacent areas. The amount
of temperature advantage depends
upon the size of the lake, the dis-
tance from the lake, the water
temperature and wind speed and
direction. Small lakes provide little
protection on very windy nights;
large lakes afford good protection.
The presence of many lakes in a
region may provide some protection
for a whole area. In order to adapt
forecasts to individual areas, it is
necessary to measure minimum
temperatures and observe the wind
on a series of cold nights and cor-
relate these with forecast tempera-
Various meteorological conditions
strongly influence the efficiency and
effectiveness of cold protection sys-
tems. Predominant among these
are cold air advection (discussed

-Ai ..l"


Fig. 3 Instrument shelter and instruments used by the United States
Weather Bureau, Federal-State Frost Warning Service, Lakeland, Florida.

previously) and nocturnal inver-
sions. On clear, calm nights, radiant
heat is lost through the top of the
imaginary box in Fig. 2 from the
soil and the exposed leaves. The air
in contact with these is cooled by
conduction and eventually cold air
accumulates among the trees and
particularly near the soil. This air
is colder than that above the tops
of the trees. Such a condition is
called a nocturnal inversion and
occurs on all calm, clear nights.
If an inversion -is present, the
light, heated gases from firing are
slowed by the warmer air above and
may be restricted by the inversion
to lower layers of the atmosphere.
Cold protection with a wind ma-
chine depends on the presence of an
inversion, with warmer air aloft
available for mixing with colder air
near the ground.
It is desirable to measure the
temperature of leaves and fruit to
test effectiveness of the cold pro-
tection operation while it is con-
ducted. The next day is too late to

surmise that not enough heaters
were lit, the wind machines were
turned on too late, or, too little
water was used in sprinkling. On
the other hand, it is unwise to over-
protect and expend fuel resources
that may be badly needed on suc-
ceeding nights.
Temperature measuring devices
and their care. Temperatures of
leaves, fruit and inversions can be
measured with a remote indicating
thermistor or thermocouple ther-
mometer of the type shown in Fig.
4. Such devices permit measure-
ment of air temperatures (aloft and
near the ground), and temperatures
of exposed leaves and fruit. Ther-
mistor systems are usually less ex-
pensive and -more portable than
thermocouple measuring systems.
Remote-indicating thermometers,
such as thermistor systems, must be
properly cared for to give satisfac-
tory service. The meter movement
must be protected from shocks and
dust, the batteries must be checked
prior to use and replaced if defec-



tive, or accurate readings will not
be obtained. Plugs, wires and
probes must be correctly installed
and good connections maintained.
One of the common tools used to
measure temperature in the grove is
the familiar glass and alcohol mini-
mum indicating thermometer, Fig.
5. These instruments are inexpen-

sive, rugged and reliable-if proper-
ly used and maintained.
The temperature indicated is the
ambient or air temperature, and not
the temperature of trees and fruits
being protected. There is a distinct
tendency for the alcohol column in
the thermometer to separate, and
unless this is corrected the indicated

1ig. 4 Inermistor thermometer and remote probes which can be used to
measure inversion, fruit and leaf temperatures. Proper placement of
probe for measurement of leaf temperature.


jf aiz


Fig. 5 Minimum indicating thermometer in a commonly used grove shelter.

temperature will be too low by sev-
eral degrees.
The alcohol column separation
can be checked visually and cor-
rected by placing the bulb in warm
water and gradually heating the
water until the column is again
united (Fig. 6). Never heat the
bulb directly with a flame as the
intense heat will boil the alcohol

and rupture the glass or will cause
the glass bulb to stretch and ruin
the calibration and accuracy of the
Exposing the thermometer to full
sunlight will cause deterioration by
a loss of color in the alcohol, re-
peated separation of the column
and eventual loss of calibration.

Fig. 6 Separated column in a minimum
indicating thermometer and correct pro-
cedure for repair.




10 ~



Cold Protection by Heating

Firing has been used to protect
plants from cold for at least 2,000
years. It is successful if enough
fuel is burned in the grove to keep
plant and air temperatures above
danger levels. Fires are of two
types, those that burn as open
flames or open fires, and those that
heat metal objects such as a stack
which radiates heat. Fig. 7. shows
examples of some heating devices
used for cold protection in Florida.
How firing provides cold protec-
tion. Firing provides cold protec-
tion by replacing heat that is lost
from the grove. If heat lost is
Si' t *'"


exactly equal to heat gained from
burning fuel, the temperature re-
mains constant. Part of the energy
released by the fuel burned in the
grove takes the form of hot gases
and heated air which provide cold
protection to the plant as convec-
tive heat. Convective heat pro-
duced must be distributed through-
out the entire grove and warm the
air in order to be effective.
Another part of the heat from
burning fuel is released as radiation
from the flame and from objects
heated by the flame, such as the
heater stacks. Plants may be




q 6t&1 P


.7 Some heaters com-
ily used in Florida.

L A ,",^a* .;.^ ." .,..'..,

warmed directly by radiation with-
out heating the air. Trees close to
heaters are warmed by radiation
and are usually warmer than the
If all the heat produced by
burning fuel could be kept within
the grove, heating would be effi-
cient. However, this is not possible
since radiation passes directly to the
sky from the top of the grove, and
light, hot gases produced by burning
fuel rise above the trees and are
blown away by the wind.
On cold, calm nights radiant loss
of heat from a grove ranges from
20 to 40 BTU's3 per square foot per
hour in Florida. Heat loss by con-
vection is relatively small, usually
less than 5 percent of the radiant
heat loss. Under conditions when
the leaves are colder than the air,
heat is gained by convection rather
than lost.
The total heat loss from unpro-
tected citrus groves ranges from 0.9
to 1.8 million BTU's per acre per
hour. However, in order for heating
systems to provide cold protection,
at least 3 to 5 million BTU's per
acre per hour are required. Heat
requirement is greater than natural
heat loss from an unprotected grove
because heating efficiency is low.
The theoretical amount of heat
required for cold protection as in-
fluenced by wind speed is shown in
Table 1. This is the amount of heat
required to provide cold protection
to an exposed leaf. By dividing the
heat required for cold protection by
the heat produced from heating de-

3British Thermal Units (BTU) are a
measure of heat-the quantity required
to increase the temperature of one pound
of water one degree fahrenheit.

vices, efficiency can be calculated.
This efficiency ranges from 10 to
35 percent.
Factors which control the
amount of heat required for cold
protection. The amount of radiant
heat produced by the heating de
vices and the wind speed are two
important factors controlling the
amount of heat required. An in-
crease in percentage of radiant heat
will usually increase the efficiency
and the effectiveness of cold pro-
tection, especially on windy nights.
Along the edges of the heated area
efficiency is always low. The hot
gases produced by the fires are light
and rise, causing cold air inflow
around the borders. To prevent
damage to the border trees, addi-
tional heating must be supplied.
Grove size has a considerable
effect upon heating efficiency. Much
of the heat produced is blown away
by the wind. Anything which re-
duces wind speed increases heating
efficiency. Big groves reduce the
wind speed by the braking effect
of the trees. As a result, heating is
more efficient and more effective in
a big grove. Moreover, the percent-
age of the grove in border trees is
smaller, and the relative importance
of inflow is reduced.
Tree size has an important in-
fluence ur ~' heating efficiency.
Large trees reduce the wind in the
grove and increase efficiency. Small
young trees r-' sent little wind re-
sistance antm st of the heat can
be blown away easily. Radiant heat
is used to warm the tree directly;
however, the percentage of radiant
heat intercepted and absorbed by
the tree depends directly upon the
size of the tree and distance to the

Table 1.- Theoretical minimum quantity of heat required for cold protec-
tion of citrus in BTU per acre per hour.

T. Temperature difference between
Wind mph heated and non-heated leaves *F.
1o 2" 3 4 5"
0- 2 250,000 496,000 745,000 933,000 1,241,000
4- 6 355,000 710,000 1,066,000 1,412,000 1,777,000
8 10 642,000 1,284,000 1,926,000 2,568,000 3,210,000

heater. Canopies of large trees
intercept more of the radiant heat
than the canopies of small trees.
The resulting efficiency is higher
with large trees.
Number of heaters required.
Enough heaters are required to pro-
vide 3 to 5 million BTU's per acre
per hour. This can be supplied by
35 heaters burning one gallon of

number 2 diesel fuel. If fewer
heaters are used, very little protec-
tion can be expected under windy,
cold conditions.
To estimate the number of
heaters required with any system,
the total heat production should be
known. The minimum number of
heaters required per acre can be cal-
culated by the following equation:

Heaters required per acre =
BTU required per hour per acre
(gallons burned per heater per hour) X (BTU per gallon of fuel)

For example hea
which burn one gall
fuel per hour; 5 mill
acre per hour musl
The heat of combus
fuels used for cold
shown in Table 2.
quired, 5 million B
divided by the heat
per gallon for the fue
sult is an estimate of
heaters required pe
minimum system if
burns one gallon per

37 heaters per acre -

Most fuels do not
ly and produce some
fore the number of hi
should be increased I
cent over the estin
This estimate does ni

ters are used der heaters. Experience at the Uni-
on of distillate versity of Florida has shown that at
ion BTU's per least one heater per tree for the
t be supplied, border rows is required for adequate
tion of various protection; if possible, heaters for
protection is two border rows are needed on the
The heat re- upwind sides.
rU's, must be The number of heaters required
of combustion on an individual night depends
1 used. The re- upon the wind, radiant heat loss,
the number of heater placement, and heater per-
'r arre for a formance. Heat loss by convection
each heater on windy nights is so great that
hour. very little difference in air tempera-
5,000,000 ture can be maintained by heating.
(135,000) Under such conditions, radiation
burn complete- produced by the hot, glowing stacks
smoke, there- of heaters provides most of the pro-
eaters required tection.
y 5 or 10 per- Heater Placement. For heat to
hated number. be effective, it must be distributed
ot include bor- throughout the grove. Convective

heat is distributed best by spacing
heaters evenly through the grove.
Distribution of radiant heat is also
optimum with this spacing. The
intensity of radiation decreases in-
versely with the square of the dis-
tance. A tree 20 feet from a heater
stack will receive about 1/4 the
amount of radiant heat received by
a tree 10 feet from the heater stack.
Best cold protection can be ob-
tained by using all the heaters avail-
able and reducing the burning rate.
This is true for both convective and
radiant heating.
With small trees it is best to place
heaters in the row between the
trees. This minimizes the distance

from the heater to the tree. Since
losses due to wind are large in a
young grove and since the canopy is
small, placing the heater between
the trees results in the highest pos-
sible efficiency of cold protection.
Best results will be obtained with
trees which have canopies 10 feet
or more in diameter by placing the
heater in the middle of the row
between trees.
A heater placement configuration
for maximum heating efficiency and
best cold protection is shown in Fig.
8. This system is superior to
placing heaters in alternate rows
because the distribution of radia-
tion and convection is more nearly

Fig. 8 A heater placement pattern for maximum efficiency of heating and
best cold protection.

D00 o0D0 0

* 0 0 0 0 0

*0 0 0 0* 0
0* 0 0 0
00000 Q






0 0
0 0 (




Table 2.- Total heat of combustion of various fuels.

Kerosene 19,800 137,000
Distillate 19,500 135,000
(grades 1-4)
Liquefied Petroleum
Gas (Butane-Propane) 95,500
Coke 12,000
Wood (pine) 7,900
Rubber 6,000

uniform. Basically, the pattern con-
sists of one heater for every 2 trees.
The heaters are spaced in alternate
middles in each row. Spacing is
staggered in the adjacent rows. This
system is superior to placing heaters
in alternate rows. While some diffi-
culties may be encountered in re-
fueling and lighting, they are prob-
ably worth the effort in terms of
cold protection.
Heaters may be placed under the
canopy of the tree. However, some
problems arise from this practice.
The tree must be shielded from the
direct flames or heat damage and
desiccation will occur. If a flame
shield is used, the hazard of injury
to the tree is reduced. Cultivation
beneath the skirt of the tree is diffi-
cult with permanent under canopy
devices; weeds and grass that col-
lect in the vicinity of the burner
present a fire hazard; and lighting
is inconvenient if tree canopies are
Performance characteristics of
heaters. Heaters have definite char-
acteristics of performance which
have been determined and which
can be specified with some certain-
ty. Stack heaters burn fuel more
efficiently than open pots and pro-
duce more radiant heat. LP gas

burners usually have a high effi-
ciency of combustion-i.e. most of
the fuel is oxidized to carbon di-
oxide and water-but the percent-
age of radiant heat is very low for
open gas flames. The characteristics
of solid fuels and random sized open
pots is very difficult to specify be-
cause the size of the burning area
and flame temperature are so
The percentage of radiant heat
produced by various heaters used
for cold protection is given in Table
3. There is wide variation among
the types. Gas fired tile or metal
mantle heaters have been used only
on an experimental basis, and are
included as examples of high effi-
ciency radiant heaters. Of the com-
monly used heaters, the return
stack and jumbo cone produce more
radiant heat than do other types.
This is an advantage on windy
nights when the tree can be warmed
directly by the radiant heat.
Burning regulation trials with the
jumbo cone and return stack have
shown that the lowest burning rate
of the jumbo cone is lower than the
return stack. This allows heating
for a longer duration with the
jumbo cone than with the return
stack. At the lowest draft setting

Table 3.-Average performance characteristics

of heaters used for cold

return stack heaters, which are not
equipped with dampers, will burn 9
gallons of fuel in 8 to 9 hours;
jumbo cone heaters adjusted simi-
larly will burn 9 gallons in 12 to 14
hours. However, jumbo cone heaters
tend to burn more slowly as the fuel
is depleted. This is usually not
serious since regulation of burning
rates cannot be sufficiently precise
for such an exacting requirement.
Lighting and regulating heating
devices. Heaters should be lit be-
fore the temperature of exposed
leaves or fruit reaches dangerously
low levels. The best method for
determining when heaters should be
lit is to measure the temperatures

of exposed leaves and fruit.
Remote indicating thermometers,
such as thermistors or thermo-
couples, are well adapted for this
purpose. If these are not available
a minimum indicating thermometer
-exposed in the grove on top of a
small flat board 4 to 5 feet above
the surface-may be used as an
index of exposed leaf temperature.
(Fig. 9)
Heaters should be lit when the
exposed leaves or the exposed ther-
mometer reaches 22 or, if fruit is
to be protected, when the exposed
thermometer reaches 26 To esti-
mate in advance when to light, the
exposed thermometer or leaf tem-

Fig. 9 A minimum indicating thermometer exposed on top of a small
board, 4 to 5 feet above the ground may be used as an index of exposed
leaf temperature,

Heater Type Input 1000 BTU per hour % Radiant Heat
Gas heaters 50-100 45-55
(tile or metal mantle type)
Coke heaters 50-100 30-35
Jumbo cone 75-150 18-26
Return Stack 75-150 15-22
Lazy Flame or
Short stack 75-150 10-15
Barrels highly variable 5-15 (estimate)
Open pots highly variable 5-15
Solid fuels 25 2-5
Open gas jets 25-75 0-5

(Heaters should be lit at this temperature)


8 9 10 II 12 I 2 3 4 5 6

Fig. 10 A method for estimating the earliest time firing will be required.
Heaters are to he lit at 24o. Temperatures are read hourly and plotted when
30' is reached. Lines joining these hourly temperature plots intersect the
24 degree, when projected, at estimated earliest time for firing.

perature should be read at sunset
and hourly thereafter. If this tem-
perature is plotted against time, as
shown in Fig. 10, and a line con-
necting the points is drawn, an esti-
mate can be made of the earliest
time firing will be required. This
estimate is not precise but will
usually predict a firing time earlier
than actually required. It is helpful
in alerting firing crews and assem-
bling equipment and personnel.
Heating is more effective on calm
nights with strong temperature in-
versions than on windy nights.
Warm heated gases from the fires
will rise above the top of the grove
and may cause temperature modifi-
cation 40 to 50 feet above the
ground before temperature changes
occur at the surface. The burning

rate should not be increased if
temperature changes are observed
above the grove immediately after
lighting heaters. After 15 to 30
minutes have elapsed and the tem-
peratures near the surface still re-
main low, the burning rate may be
On windy nights, radiant heat is
required for adequate protection.
The absolute quantity of radiant
heat produced will be more impor-
tant than the efficiency with which
it is produced. High burning rates
may be desirable. Even under such
conditions, all heaters should be lit
to provide good distribution of radi-
ant heat.
As a general rule, heaters should
not be burned at excessively high
rates. Burning in excess of one

gallon per hour tends to reduce
heater efficiency. It is better to in-
crease the number of heaters than
the burning rate. High burning
rates tend to shorten the life of the
heater by oxidation and subsequent
rusting of stacks. The most effec-
tive and efficient heating can usual-
ly be accomplished with heaters
that produce few flames beyond the
top of the heater stack.
When cold conditions with high
winds are expected, the temperature
of the air in the grove will remain
low even with firing. The purpose
of heating under such conditions is
to slow the loss of heat from the
It is desirable to have the maxi-
mum amount of heat at the coldest
time. Minimum temperatures com-
monly occur near sunrise. Heaters
should be lit so that maximum heat
production will occur during the
period of lowest temperatures. If
the total amount of heating equip-
ment is limited, heating may be de-
layed until temperatures are dan-

gerously low, but then lit to provide
heat during the period of lowest
temperature. Such a practice is
known as timed firing, because the
length of time that fires can be kept
burning determines the lighting
It requires less fuel to maintain
temperatures at satisfactory levels
than to raise them. Trees have a
large heat capacity and cool more
slowly than air-especially the
fruit, trunks and large limbs. If
they become cold, large quantities
of heat are required to raise their
temperatures. Best results will be
obtained with a minimum amount
of heat. Once the temperature has
fallen below the satisfactory level,
more than twice as much heat is
required to raise the temperature
one degree than to maintain it.
Extinguishing and refueling
heaters. Heaters may be extin-
guished when leaf temperatures
reach 24' or fruit temperatures
reach 28 This temperature can be
measured with an exposed thermom-

for rapid refueling of heaters.

11 Equipment

'L".? -...
It~~- -~;~;-'~;?5~:~::;;~i,~-~cl~"*JI-~:~3~c
i~sr`t~~'WC~Rg,~*-~*~j~..4 i

eter that is shielded from the sun's
direct rays. During daylight hours,
if the air temperature is 26 or
above, heaters can be extinguished.
After a night of use heaters must
be refueled. Equipment for rapid
refueling and removal of oil at the
end of the cold season is shown in
Fig. 11. This equipment consists
of a tank with a power take-off
pump that can be reversed for oil
withdrawal. Hot or burning heaters
should not be refueled. Pouring oil
into a hot heater produces vapors
which ignite easily. Personnel en-
gaged in refueling should avoid
spilling oil and contaminating
Heaters should be given good
care. The tops should be securely
closed to prevent rain water from
entering. Water in the oil will cause
operating difficulty. Oil floats upon
water and will ignite with a torch.
Once the oil becomes hot, the water
in the heater will boil and escaping
steam will either extinguish the
heater or reduce the burning rate
and may occasionally cause the
stack to be blown off. Water must
be removed before proper burning
can be obtained.
If heaters have been operated at
high burning rates or for several
nights, the galvanized stack will be
oxidized. Unless these stacks are
treated with a rust preventive, se-
vere rusting will occur. Rust pre-
ventatives are available from most
major oil companies and should be
secured by consulting their techni-
cal representatives. If heater stacks
become coated with a thick layer of
soot, the radiant heat production
is reduced and the stacks should be

Heaters should not be stored in
the grove or under trees in Florida.
Rain water enters the heaters and
normal expansion of the oil and
capillarity during warm summer
days will result in oil spillage. Fur-
thermore, the high humidity be-
neath the tree canopy contributes
to rusting and deterioration of the
heater stack. Heater stacks which
are properly cared for will last for
approximately 10 years. The life of
the bowl is greater than 10 years.
Precautions when using heaters.
Precautions for lighting heaters
should be practiced at all times.
The crew should be impressed to
keep dry and free of oil. Normal
body temperature will warm oil so
that it becomes easily ignited. Oil
soaked clothing is a great fire haz-
ard in the presence of open flames
from lighting torches which provide
excellent opportunity for ignition of
the soaked clothing.
Flame screens in the spouts of
lighting torches shown in Fig. 12
should always be kept in place.
These screens prevent the flame

Fig. 12 The flame screen in the spouts
of the lighting torch prevents the flame
from striking back into the torch.

from striking back into the reservoir
of the lighting torch. Lighting mix-
ture is commonly composed of equal
parts gasoline and heater oil. The
gasoline increases the flamability of
the mixture and unless the screen
is kept in place, the torch is a fire
and explosion hazard. Screens
should be inspected to make sure
they are in place. A stopped screen
should never be pierced but should
be cleaned. The lighting crew
should be aware of the necessity of
keeping the screen in place at all
Lighting hot heaters is a definite
hazard. The grey vaporized oil
cloud is flamable and will ignite
easily. Since heaters must be lit in
this condition occasionally, the grey
vapor should be ignited first and the
operator remain as far from the
heater while lighting as possible
since flash ignition will occur. If
possible the heater should be al-
lowed to cool before lighting is at-
tempted. A definite procedure to
care for a fire victim should be
drawn up and the entire crew should
be made aware of this procedure.
Fire blankets and instructions for
their use should be available, and
the means of emergency treatment
and medical care be available and
Evaluating a new system. Since
new heating systems become avail-
able annually, some criteria are
needed to evaluate their effective-
ness. The most important one is the
total heat production by the system.
Systems which cannot produce
more than 4 to 5 million BTU's per
acre per hour do not meet minimum
heat requirements and should not
be considered for cold protection.

Heat distribution should be uni-
form throughout the grove and pref-
erably the system should produce
radiant heat. It is doubtful that
good results will ever be obtained
by systems that do not provide
good heat distribution throughout
the grove.
Tests of performance data sup-
plied by the manufacturer should
be critically evaluated. In order for
a device to be effective, the mini-
mum amount of heat required for
protection must be possible on a
sustained basis. Due to overall
variations from year to year, no
single year of testing demonstrates
satisfactory performance. Tests
should include either performance
under critically cold conditions or
valid theoretical computations
which demonstrate its effectiveness.
In order for performance data to be
valid the conditions of the test must
be carefully specified and should be
available for detailed study. Data
obtained under short test periods
or under conditions where cold in-
jury was not anticipated are of
limited value.
An ideal system of cold protec-
tion by firing may some day be de-
vised. At the present time such a
system does not exist. However,
definite criteria of an ideal system
are known. The system should be
permanent, have a central fuel sup-
ply, have a high performance effi-
ciency, be self-lighting, provide cold
protection under the most severe
conditions envisioned, and be eco-
nomically justifiable. If a system
can be devised to satisfy these con-
ditions, it will constitute a great
improvement in heating for cold


Fig. 13 Wind machines used for cold protection in Florida.

Cold Protection with Wind Machines

Wind machines offer some excel-
lent advantages in cold protection
because they minimize labor re-
quirements, require less refueling
and less fuel storage than heaters,
are permanently located in the
grove, have a low operational cost
per acre and do not produce smoke
and air pollution.
These advantages must be
weighed against the disadvantages
of rather high capital costs and the
failure of the wind machine to pro-
vide adequate cold protection under
all conditions.
How wind machines provide
cold protection. Wind machines

provide cold protection by mixing
warmer air above the trees with
colder air among the trees. They
produce little or no heat but de-
pend on the presence of warmer air
aloft-the temperature inversion
discussed previously-to provide
cold protection. Under cold, windy
conditions, the wind machine will
not provide protection.
Protection is provided by using
heat stored in the upper air. By
mixing the warmer air above the
trees with the colder air around the
trees, part of the heat lost by radi-
ation cooling of the surface and the
trees is replaced. The net effect is

to slow the rate of temperature fall
and maintain the temperature
higher than in unprotected areas on
clear, calm nights. The wind ma-
chine provides cold protection by
causing vertical motions in the
atmosphere. On nights when wind
speeds are higher than 5 mph, no
protection will be provided by this
Where the wind machine pro-
vides cold protection. Because of
the climatic nature of cold weather
in Florida, a fairly high percentage
of cold nights is accompanied with
wind. The northern half or two-
thirds of the state has a sufficiently
high incidence of such nights that
wind machines alone cannot provide
adequate protection.
Effectiveness of wind machine
systems increases as one moves
southeastward in the state. Even
though wind machines may not pro-
vide complete cold protection under
all conditions, they will often mod-
erate the effects of cold weather.
If complete protection is desired,
additional cold protection methods,
such as heaters, must be used in
conjunction with the wind machine.
The amount of cold protection
that can be expected. The exact
amount of cold protection that can
be expected from a wind machine
depends on the strength of the in-
version, the power of the machine,
the wind speed, and the distance
from the wind machine. Results ob-
tained with a wind machine used in
a research program at the Univer-
sity of Florida showed that under
clear, calm conditions, at least 2
of cold protection was obtained on
an area of 8 to 10 acres.
The machine used delivered ap-

proximately 85 horsepower to the
propeller. There can never be any
assurance that the temperature ad-
vantage due to the machine will be
adequate for cold protection. If
temperatures become sufficiently
low, damage will occur even though
the protected area may be warmer
than non-protected areas.
The area of protection and its
shape. The exact temperature at
any particular point in the pro-
tected area depends on the distance
from the machine, and wind speed.
Wind tends to distort the area of
protection and displaces it down-
wind from the machine. On calm
nights, the protected area is roughly
circular with the machine at. the
center. On nights with wind drift,
the pattern is usually oval with the
wind machine in the northwest
Protection obtained decreases as
one moves away from the wind
machine. Unless temperatures in
the protected area are more than 20
higher than the unprotected area,
this advantage should not be as-
signed to the machine since it may
be due to microclimatic influences.
Fig. 14 shows protection patterns
obtained with a machine delivering
85 brake horsepower to a 15 foot
fan turning 590 rpm and rotating
through 360 each 41/ to 5 min-
utes. These results are typical of
those obtained on other nights. If
multiple wind machine installations
are used, additional protection is
obtained where the protected areas
overlap. This reduces the power re-
quired for cold protection.
Approximately 10 brake horse-
power delivered to the propeller is
required per acre for cold protection

with a single machine and approxi-
mately 8 brake horsepower per acre
with multiple machine installations.
Use of 2 engines and propellers on
the same tower increases the area
protected by about 60 percent over
the single engine machine. Some-
what higher temperatures may be
expected in the immediate vicinity
of a 2-engine machine than around
a single engine machine.
Wind machine location. Since
the wind machine protects a
roughly circular area, it can never
give complete protection for a*
square or rectangular block. The
protection pattern changes as the
wind drift changes in speed and
direction. Because the wind ma- 0
chine provides maximum protection

Fig. 14 Protection patterns obtained
with a machine delivering 85 brake
horsepower to the fan.

S, Check Minimum 29.50
Darkly shaded = 5.0 acres Engine speed: 2600 rpm
Lightly shaded = 10.0 acres uare co indic
Square corners indicate
Meteorological data: boundaries of test plot.
Sky: clear
Wind: NNW 3-5 mph
Maximum inversion (5-55 ft.): 5.5 F.

290 29

0--- 65'
Darkly shaded = 5.0 acres
Lightly shaded = 10.0 acres
Meteorological data:
Sky: clear
Wind: NNE 0-3 mph
Maximum inversion (5-55 ft.): 7.00 F.

k Mni mum 29

Engine speed: 2600 ipn
Square (oJners: indicate
boundaieis Lest plot

j65' Check Minimum 29.50
Darkly shaded = 5.0 acres Wind=4.0mph
Lightly shaded = 10.0 acres Engine speed: 2600 rpm
Meteorological data: Square corners indicate
Sky: clear boundaries of test plot.
Wind: WW 4.0 mph entire night
Maximum inversion (5-55 fP.): 3.90 F.



4 6866O1'--

Fig. 15 The area which is satisfactory for location of wind machine with
predominately northwest drift.

on clear, calm nights, the location
should be fairly near the center and
slightly to the northwest. If the
predominate direction of drift on
cold nights is known to be from one
direction, then the machine should
be located closer to the updrift edge.
The areas satisfactory for loca-
tion of the wind machine with pre-
dominantly northwest drift condi-
tions are shown in Fig. 15. Deciding
where to put a machine is easier
when multiple installations are
made because displacement of the
protected area of one machine is re-
placed by the next machine. Topo-
graphical features often dictate the
location of single installations when
they are used to specifically protect
small cold areas.
When the wind machine should
be started. Since the purpose of a

wind machine is to mix air and take
advantage of heat stored in the
atmosphere, it should be started
early in the evening to maintain
temperatures as high as possible.
Many researchers report it is diffi-
cult to regain a previous tempera-
ture advantage if the wind machine
is stopped. For that reason, it
should be started when tempera-
tures are near 320 when protection
will be needed.
An additional reason for starting
the machine when air temperatures
are near 32' is that exposed leaves
will be considerably below air tem-
perature and may be approaching
critical levels. This practice also
helps assure that all wind machines
can be started before temperature
levels have fallen critically low.
Wind machines used in combi-

nation with heaters. On nights
when temperatures are so low that
the wind machine alone cannot pro-
vide adequate protection, or when
inversions are very weak, heaters
can be used to improve the amount
of cold protection obtained. More
protection is provided by the com-
bined system than by either used
separately. The wind machine
mixes and distributes hot gases from
heaters throughout the grove and
makes protection more uniform and
effective. Heat rising from the
heaters is mixed throughout the
grove and thus is used more effi-
ciently. On very windy nights, wind
machines used with heaters will not
provide additional protection.
The number of heaters and
placement about the wind ma-
chine. When heaters are used with
a wind machine, approximately 15
to 25 heaters per acre or their
equivalent should be used; prefer-
ably 'these should be return stack
or jumbo cone heaters because they
are more easily regulated and not
as subject to poor burning under
the windy conditions provided by
the wind machine.
Heaters should be placed uni-
formly throughout the grove, ex-
cept that they should not be placed
closer than 100 to 150 feet from the
wind machine. Not only is it diffi-
cult to keep heaters lit close to the
machine, but also they tend to lift
the wind machine jet and cause it
to rise above the trees, thereby de-
creasing its effectiveness. Additional
heaters should be used on the
borders, especially on the updrift
sides. The extra heaters can be
used to make the protection uni-
form over the entire block.

General considerations for wind
machine operations. Since the ef-
fectiveness of a wind machine
depends on the presence of an inver-
sion, and since there is no way of
knowing the strength of an inver-
sion without measurements, it is
advisable to have a facility to
measure inversion strength. This
can consist of a pole 40 to 50 feet
high with a sensor at the top and
one near the 5-foot level. Tempera-
tures can be read with a remote
indicating device either of a ther-
mistor or thermocouple type.
Wind machine effectiveness de-
pends very much on proper opera-
tional speed. If the engine speed
is reduced even by 10 or 20 percent,
the power input to the propeller
may be reduced by 30 to 40 percent.
This can render the machine in-
effective. Therefore, wind machine
operational speed should be checked
frequently. Effectiveness of the
machine, compared to the inversion
present, should also be determined.
If a strong inversion is present,
i.e., a large temperature difference
between 5 feet and 40 to 50 feet,
and very little protection is ob-
tained, the operating speed of
the wind machine should be checked
The easiest method of making
this check is to determine the
length of time required for the ma-
chine to make one complete revolu-
tion about the tower. If the time
required is greater than that in the
specifications for proper operation
of the machine, it is almost certainly
operating at less than rated power
and speed. Unless this is corrected,
very little cold protection will be

Wind machines should be turned
on periodically during the cold
season to make sure they are in
good running condition and can be
started when needed. They should
not be operated in a high wind
unless the manufacturer has specifi-
cally indicated that such operation
will not damage the machine.
Operation in high winds will cause
the machine to vibrate excessively
and may produce mechanical failure
and damage. Moreover, operation

in a high wind will not provide
additional cold protection.
Power units used on wind ma-
chines should either be of the origi-
nal manufacturer's design or should
be adapted by the manufacturer or
a competent engineer to insure
proper power, speed and perform-
ance. A maintenance program
should be established with the wind
machine that will reduce the possi-
bility of machine failure on cold

Cold Protection by Overhead Sprinklers

Permanent overhead irrigation as
a cold protection system has several
attractive features. One is the
sprinkler system can be started and
stopped easily. This labor-saving
feature is a particular advantage
on cold nights when it is difficult
to obtain labor on short notice.
Another attractive feature is the
possible use of the system for regu-
lar irrigation. The adequacy for
irrigation should be determined by
a competent irrigation engineer.
Sprinkling differs from other cold
protection systems in that improper
use can result in more damage than
if the trees were left unprotected.
This emphasizes the need for intelli-
gent use of the method.
How sprinkling provides cold
protection. Heat is released when
water changes from liquid to ice.
This is called the heat of fusion.
The amount of heat released ex-
pressed as BTU's and calories is
shown in Table 4. When water is
freezing its temperature will be near

32. The heat liberated as the
water freezes maintains the temper-
ature near 32 even though the
surroundings may be colder. As
long as a mixture of both water and
ice are present, the temperature re-
mains near 32". This temperature
is an equilibrium between vapor,
liquid and ice and is known as the
triple point.
For all of the plant to be pro-
tected, it must be covered or in-
cased in the freezing ice water
mixture. A simple visual check to
determine the adequacy of the irri-
gation rate is inspection for icicles.
If icicles are forming and the under-
sides of all leaves and branches are
covered with ice, sufficient water is
being used. If, on the other hand,
only the top part of the leaves and
branches are covered with ice, dam-
age can be expected.
Result of insufficient sprinkling
rates. If less water is applied than
is necessary for protection, or if the
distribution is not good enough to

Table 4.- Comparison of heat consumed
released through freezing.

Unit Of Heat of Vaporization
Water (at 32 F.)

1 Gram

1 Pound
1 Gallon

596 Calories or
2.4 BTU
1072 BTU
8100 BTU

thoroughly wet and coat all parts
of the plant, damage will be more
severe than if water had not been
Fig. 16. is a graphic example of
using too little water. A wet leaf
may be colder than a dry leaf due
to cooling by evaporation. Since
evaporation consumes approximate-
ly 7/2 times as much heat as is
liberated by freezing, at least 71/
times as much water must be frozen
as is evaporated. If less heat is
released by freezing than is used for
evaporation of water and ice from
the leaf, the sprinkled leaf will be-
come colder than a non-sprinkled
Fig. 17 shows in graphic form the
chilling that may be expected from
evaporation if insufficient water is
used. The exact quantities of heat
released in freezing and consumed in
evaporation are shown in Table 4.
The heat consumed in evaporation
is called the heat of vaporization.
An additional hazard is encoun-
tered when sprinkling is used for
cold protection. When a citrus
plant is dry, the killing temperature
is from 2 to 4' lower than when
wet. According to University of
Florida test results dry, sweet
orange leaves can be expected to
withstand temperatures of approxi-
mately 20. When these same

through evaporation with heat

Heat of Fusion

80 Calories or
0.32 BTU
144 BTU
1200 BTU

--. -*. .,,, .-_-. -- ,.

Fig. 16 The severe damage to the trees
on the left was caused by sprinkling
at 0.1 in/hr. during the December 12-14,
1962 freeze. All the trees in the picture
are of the same age, stock and scion.
Those on the left were sprinkled while
those on the right were left unsprinkled.

leaves are wet, they are killed at
temperatures near 24. This em-
phasizes the need to have a reliable

Fig. 17 Chilling that may be expected
from evaporation if too little water is

Temperature Depression From Deficit Sprinkling
(7,-Ta) tl-

0 1,0 20
Vopor Pressure Deficit millibarss)

system if sprinkling is to be used.
Amount of water needed for
cold protection. Table 5 gives the
amount of water in inches per hour
necessary to protect a citrus grove
under various temperature and
wind conditions. It is assumed that
the water will be evenly distributed.
The temperature given is the mini-
mum expected for dry exposed
leaves which are not sprinkled.
With calm and clear conditions,
leaf temperatures may be 3 to 40
below air temperature. With a light
wind, the leaf temperature will be
only slightly below air temperature.
To estimate the sprinkling rate
required for cold protection, enter
the table with the minimum air
temperature forecast and the maxi-
mum wind expected. If the amount
of water necessary for protection is
greater than the irrigation system
can supply, the chances of failure
are very high and the system should
not be used.
For example, assume a minimum

temperature of 240 was forecast
with lowest temperatures expected
at sunrise under calm, clear condi-
tions. Under some circumstances,
the dry leaves may be 20 to 40
below air temperature due to radi-
ation heat loss. Enter Table 5 at
20 and read to the right to the 0
to 1 mph column. The sprinkling
rate required is 0.16 in. per hour
for complete cold protection.
As a second example, suppose a
minimum temperature of 24' is
forecast with winds of 5 to 8 mph.
Under these conditions the leaves
and fruit can be expected to remain
close to air temperature. Enter
Table 5 at 240 and read to the
right to the 5 to 8 mph column.
The sprinkling rate required is 0.30
in. per hour for complete cold
In a mature grove, where wind is
reduced by the trees, somewhat
more protection would be obtained
than estimated by the table. As a
general rule, the estimates in the

Table 5.- Precipitation rate in inches per hour necessary for cold protection.

Dry Leaf* 0 to 1 2 to 4 5 to 8 10 to 14 18 to 22 30
270F 0.10 0.10 0.1 0.1 0.2 0.3
26F 0.10 0.10 0.14 0.2 0.4 0.6
24F 0.10 0.16 0.3 0.4 0.8 1.6
22'F 0.12 0.24 0.5 0.6 1.2 1.8
20F 0.16 0.3 0.6 0.8 1.6 2.4
18F 0.20 0.4 0.7 1.0 2.0 3.0
150F 0.26 0.5 0.9 1.3 2.6 4.0
11FF 0.34 0.7 1.2 1.7 3.4 5.0

* The temperature of a dry leaf is the expected minimum leaf
tected leaf. This will range from 1lF. below air temperature
to 3-4F. on very calm nights.

temperature on an unpro-
on nights with light wind

table are conservative and complete
cold protection may be obtained un-
der many conditions with slightly
less water. However, in order to
assure that cold protection will be
obtained under all conditions, the
values in Table 5 should be
The temperature at which to
start the sprinklers. If the deci-
sion has been made that sprinklers
are to be used for cold protection
on a specific night, the question
arises, "When shall the sprinklers
be turned on?"
If it is calm, the system should
be started as soon as the leaf
temperature reaches 27' to 28.
Leaf temperatures will vary several
degrees due to exposure to the sky.
Consequently, exposed leaf tem-
peratures should be used to de-
termine when to start sprinkling.
If the air temperature is used to
determine when to start the system,
32' should be used for a calm night
and 290 for a windy night. The only
reason to delay the sprinkling is to
reduce the ice loading, tree break-
age and excessive soil moisture.
Turning the sprinklers off. Since
a freezing ice-water mixture is
always near 32, a melting ice-water
mixture is, for the same reason,
always near 32. Therefore, when
the ice is melting from the trees,
the irrigation system can be
stopped. Another way to determine
when the system can be turned off
is to measure the wet bulb tempera-
ture. This can be done with a sling
psychrometer (Fig. 18) or by
equipping a thermometer with a
cotton sock that dips into a reser-
voir of distilled water. Such an
arrangement is shown in Fig. 19.

The thermometer must be shielded
from the. direct rays of the sun.
When the wet bulb temperature is
320 or higher, the irrigation system
can be stopped without danger of
damage to any part of the tree.

Fig. 18 Two types of sling psychrom-
eters which may be used for determin-
ing the temperature at which the sprin-
klers may be turned off. (Note only one
thermometer may be used for measuring
wet bulb temperature. For humidity
measurement, two thermometers are re-


Fig. 19 A thermometer equipped with a cotton sock that dips into a wa-
ter reservoir.

Ice loading. A gallon of water
weighs 8 1 pounds. Consequently,
for every gallon of water that is
frozen upon the tree, another 81/'
pounds of weight is added. This
added weight is a threat to the tree.
It is difficult to predict exactly how
much breakage will occur due to ice
loading since it depends upon the
individual characteristics of a tree.
Some branches will be bent until
they rest upon the ground without
breaking, others may become par-
tially supported by the ice and will
not break until the ice begins to
Small trees which can be bent
until they rest upon the ground
without breaking will show very
little damage due to ice loading.
Ice breakage in nurseries can be
reduced by trimming back the
nursery trees to produce a compact
For every tenth of an inch of
water added to the tree as ice, one
half pound of weight is added per
square foot. A tree with a canopy
10 feet in diameter will increase in

weight by about 40 pounds for every
tenth inch of water. If 0.2 inch of
water per hour is applied for 8
hours, the added weight would be
640 pounds. A canopy that is 15
feet in diameter would increase in
weight by 1400 pounds. With large
trees, ice loading and breakage is a
serious problem. With small trees,
the problem is minor. Differences
due to growth habit and brittleness
of the wood may cause little break-
age in one grove and severe break-
age in another-even with the same
irrigation rates and the same tree
Special problems which may
arise. Under extremely windy cold
conditions, sprinkler heads can be-
come incased in ice and stop turn-
ing. This is a serious problem since
a portion of the grove will not be
sprinkled regularly and will suffer
cold damage. Sprinkler heads must
turn at a rate of / to one revolu-
tion per minute to provide adequate
distribution of water. Only those
sprinkler heads which the manu-
facturer assures will not be stopped

by ice formation should be used.
Routine inspection of the system
must be made throughout the night
to insure that none of the heads
are frozen and stopped.
Near the edges and borders of the
sprinkled area, damage may be
caused to adjacent non-sprinkled
trees. Some provisions must be
made to shield these trees from the
water, since insufficient water will
be applied on some cold nights and
the trees will be killed. Most
sprinkler layouts can be made to
avoid these trees or special shields
can be supplied to prevent sprink-
ling of adjacent trees.
Precautions for cold protection
by sprinkler irrigation. The grove
owner should:
Make sure the irrigation sys-
tem can deliver enough water to
provide protection under wind and

temperature conditions expected as
indicated in Table 5. Failure with
an irrigation system will probably
cause more damage to the tree than
if it were left unprotected.
Start the sprinkler system
when the air temperature reaches
32' on calm nights or 29 on windy
nights. If leaf temperatures are
used as criterion, start the sprinkler
system when they reach 27.
Never allow the irrigation sys-
tem to stop until the wet bulb
temperature is 320, or higher, or
when ice is melting and water is
dripping from the plant. This water
must be dripping from all parts of
the plant. It should be checked on
the shaded side of the tree.
Sprinkler heads must turn
continuously at a rate of 1/2 to one
rpm to provide good water distri-

(Acts of May 8 and June 30, 1914)
Agricultural Extension Service, University of Florida,
Florida State University and United States Department of Agriculture, Cooperating
M. O. Watkins, Director

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