The Lives of Bees
Page 28
ment inside its nest cavity: by conduction, convection, evaporation, and
thermal radiation. It loses heat by conduction through the ceiling of the
cavity and through the combs, and it loses heat by convection through air
currents moving through the porous cluster and within the nest cavity. The
colony also loses heat from respiratory evaporation of the adult bees, sur-
face evaporation from the moist bodies of larval bees, and surface evapora-
tion from any damp combs inside the cluster. Finally, the colony emits
thermal radiation toward all objects around it: the unoccupied combs and
the nest cavity’s walls. The isotherm lines in this figure show us that the
loss of heat from the cluster created a pocket of relatively warm air outside
the cluster in the top of the nest cavity. Note that the temperature in the
bottom of the nest cavity was the same as outside, −21°C (−6°F), but that
in the top of the cavity, and all around the cluster, the temperature was
much warmer, −1°C (30°F).
The heat that is lost from the cluster warms not only the air around it
but also the nest cavity’s ceiling and walls, and they in turn lose heat to the
environment outside by conduction and thermal radiation. If wind outside
causes air currents to pass in and out of the nest cavity through the en-
trance, then there also can be loss of heat to the outside through convec-
tion. Wild colonies reduce heat loss from their nest cavities through con-
vection by filling cracks and small holes with seams of propolis, as discussed
in chapter 5. In late summer, some colonies will also reduce their nest’s
entrance opening by building a curtain of propolis across most of it, as
shown in Figure 5.7.
The take- home message from Figure 9.4 is that a honey bee colony has
two ways to minimize its loss of heat to the environment: 1) by reducing
Seeley.indb 224
2/21/2019 8:07:55 AM
Temperature Control 225
Conduction
Convection
Evaporation
Thermal radiation
7˚
16˚
24˚
32˚
–1˚
–18˚
–21˚
Entrance
10 cm
Fig. 9.4. Anatomy of a winter cluster of a honey bee colony inside a Langstroth
hive when the ambient temperature was −21°C (−6°F), at 0700 hours on 25
February 1951. It shows the ways in which a cluster loses heat to the nest cavity’s
air and walls: through conduction, convection, evaporation, and thermal radia-
tion. It also shows how the nest cavity’s air and walls lose heat to the outside
environment: though conduction, convection, and thermal radiation. Dark shad-
ing indicates the dense, insulating mantle of the cluster. The colony’s small brood
nest is inside the 32°C (90°F) isotherm.
heat loss from the colony to the nest cavity, and 2) by reducing heat loss
from the nest cavity to the environment outside of whatever is housing
the colony (hereafter, called the “nest enclosure”). In both stages of a
colony’s heat loss—from cluster and from cavity—the rate of heat trans-
fer from inside to outside increases approximately in proportion to the
Seeley.indb 225
2/21/2019 8:07:55 AM
226 Chapter 9
difference in temperature between the inside and the outside (T minus
I
T ). This temperature difference is the driving force for the heat loss.
O
When there is no heat transfer due to evaporation, Newton’s law of cool-
ing applies:
Rate of heat transfer = C × (T
)
I−TO
The coefficient C, the thermal conductance of a structure, is a measure of
how readily heat can move from the inside to the outside of the structure
(whether a bee cluster or a nest enclosure). A bee cluster or a nest enclo-
sure with a low C has a high resistance to loss of heat, hence it has a high
insulation value. Let’s now review what is known about how a honey bee
colony adaptively adjusts the C
for its mass of bees, and how a honey
cluster
bee colony acquires an impressively low C
for the walls of its dwell-
enclosure
ing place by occupying a thick- walled cavity in a tree.
A colony of honey bees can strongly reduce its heat loss by lowering its
thermal conductance (C
). It does so when the bees press together
cluster
tightly to form a compact, roughly spherical cluster. This process begins
when the temperature inside the nest cavity dips below about 14°C
(57°F). If the temperature outside the cluster falls further, the bees press
together more tightly, thereby shrinking the cluster’s size, though at about
−10°C (14°F), cluster contraction reaches its limit. Between 14°C and
−10°C, the volume of the cluster shrinks roughly fivefold. The structure
of a winter cluster was closely investigated in 1951, by Charles D. Owens,
in Madison, Wisconsin, while he was working for the U.S. Department
of Agriculture. He did this by taking careful measurements of the tem-
peratures throughout a colony’s cluster on a day when the ambient tem-
perature was −14°C (7°F), then killing the colony with hydrogen cyanide
gas, and finally carefully dissecting the dead colony. Figure 9.4 shows the
two- part internal organization that Owens found. There was an outer zone,
between what had been the 7°C and 16°C (44° and 61°F) isotherms, con-
sisting of several layers of densely packed bees oriented with their heads
pointed inward. They filled all the empty cells in the combs and had
pushed themselves as close together as possible in the spaces between the
Seeley.indb 226
2/21/2019 8:07:55 AM
Temperature Control 227
combs to form an insulating mantle. Bees in the inner zone, however, had
room to crawl about, feed on the honey stores, fan with their wings, and
tend the brood.
Why do bees form a tight cluster at low ambient temperatures? Clus-
tering reduces a colony’s heat loss in part by shrinking its surface area for
heat loss by thermal radiation. Clustering also reduces heat loss by convec-
tion (air currents), since by pressing together the bees reduce the porosity
of their cluster and therefore its loss of heat by air currents. Most impor-
tant, the dense outer layer of a cluster forms an effective blanket of insula-
tion that reduces its heat loss by conduction. Based on measurements of a
colony’s metabolic rate as a function of ambient temperature, Edward E.
Southwick has estimated the heat conductance from a 17,000- bee (hence,
a 2.2- kilogram/4.9- pound) winter cluster to be: 0.10 watts per kilogram
per °C. Rather amazingly, this low heat conductance matches, or is even
below, that of birds and mammals of the same body weight. Evidently,
the insulation effectiveness of the dense outer layer of bees in a winter
cluster is as good as, or better than, that of the feathers of birds or the fur
of mammals.
A honey bee colony living in the wild also reduces its heat loss by oc-<
br />
cupying a thick- walled tree cavity. The thick wooden walls have low heat
conductance (low C
) so they impede heat loss by conduction from
enclosure
the nest cavity to the general environment. At present, nobody has inves-
tigated whether nest- site scouts assess the thickness of a prospective nest
cavity’s walls and factor this into their assessments of the overall quality of
a prospective homesite. I bet they do. What has been studied, however, is
how strongly the total heat conductance of a honey bee colony’s nest en-
closure differs between a thick- walled tree cavity and a conventional, thin-
walled wooden hive. This is the work of Derek Mitchell in England. The
thick walls of tree cavities differ from those of wooden hives in both heat
capacity and heat conductance, but Mitchell decided to study the effects of
differences in heat conductance per se, so he worked with models of tree
cavities that he constructed using sheets of polyisocyanurate foam, a mate-
rial that has very little heat capacity. He built a foam model of a tree cavity
Seeley.indb 227
2/21/2019 8:07:55 AM
228 Chapter 9
whose walls had the same thermal conductance per unit height (watts per
°C per meter) as the thick- walled tree cavities that Roger Morse and I had
reported on in the 1970s, in our study of the nests of wild colonies. He
also built his tree- cavity model so that its size (40 liters/10.6 gallons),
shape (a tall cylinder), and entrance properties—a 15- centimeter- (6-
inch- ) long passageway, 5 centimeters (2 inches) in diameter—matched
the average values of these variables for the tree- cavity nest sites that we
had studied. Furthermore, Mitchell worked with various kinds of wooden
hives, including British National and Warré. Inside his tree- cavity model
and each of his hives, he installed an array of temperature sensors, and he
suspended in its upper region a heating element for simulating the heat
production of a colony of bees. Finally, he powered everything up and let
the heating elements run and the temperature sensors take data for several
hours, until the conditions inside each structure came to equilibrium.
What he discovered was that his model of a thick- walled tree cavity had a
value of total (lumped) enclosure heat conductance (C
) that was only
enclosure
about 0.5 watts per °C, whereas his British National hives (for example)
had values of C
that were approximately five times higher, around
enclosure
2.5 watts per °C.
What are the effects of such a strong difference in heat conductance
between the walls of natural tree cavities and conventional wooden hives?
One is that the bees in a strong colony inhabiting a well- insulated tree
cavity can stay mobile inside their nest well into winter, possibly even
throughout it. Mitchell’s analysis shows that a 1- kilogram (2.2- pound)
colony of bees, producing heat at what is a normal rate of 20 watts at a
nest- cavity temperature of 20°C (68°F), will not need to form a tight,
well- insulated cluster until the temperature outside the cavity drops below
−30° or −40°C (−22° or −40°F), if it is living inside a thick- walled, well-
insulated tree cavity. In contrast, a colony of the same size that is living in
a standard, thin- walled wooden hive will need to go into a tight cluster
soon after the temperature outside the hive drops below about 10°C
(50°F), because its nest cavity is so poorly insulated (Fig. 9.5). This striking
difference may translate into better winter survival for colonies residing
Seeley.indb 228
2/21/2019 8:07:55 AM
Temperature Control 229
hives
trees
20
0
1
2
3
4 Kg • W–1 • ˚C
of clustering (˚C) –20
T outside
–40
Fig. 9.5. Graphical analysis of the threshold temperature at which a 1.0- kilogram
(2.2-pound) colony will need to form a cluster to stay warm depending on the
total heat conductance of its nest enclosure: a thin- walled wooden hive (British
National) vs. a thick- walled tree cavity. Tree cavities have relatively low total heat
conductance, so they have a high ratio of colony mass to nest enclosure heat
conductance (kg•W− 1•°C). Therefore, a kilogram colony of bees living in a tree
cavity will experience a warm, 20+°C (68+°F) microclimate inside its nest cavity
until T
drops below
outside
− 40°C (− 40°F), but a kilogram colony of bees living in
a British National hive will experience such a warm microclimate only when the
outside temperature is above about 10°C (50°F).
in tree cavities relative to those in hives, because colonies overwintering
inside trees can stay mobile on their combs for longer and so are probably
better at staying in contact with their honey stores. This subject requires
much more investigation, however, because the thick walls of tree cavities
have high heat capacities—not just high insulation values—relative to the
thin walls of conventional hives. This means that once the walls of a tree
cavity get cold over winter, these cold walls will delay the warming of the
cavity air to the temperature, ca. 14°C (59°F), at which a colony living
inside the tree cavity can break its cluster. Therefore, colonies living in
thick- walled tree cavities may be delayed in becoming active in spring rela-
tive to colonies living in thin- walled wooden hives.
Seeley.indb 229
2/21/2019 8:07:56 AM
230 Chapter 9
To explore further the microclimate differences that exist between a
thick- walled tree cavity and a thin- walled hive, I recently started a study
with two colleagues, Robin Radcliffe and Hailey Scofield, to describe the
microclimates inside both types of structure, without any bees, through-
out a year. We began by building two nest cavities, one made of standard
pine lumber and the other cut with a chain saw into a large sugar maple
tree (Fig. 9.6). The two cavities are identical in shape (tall and narrow),
volume (50 liters/13.2 gallons), and entrance size (5 centimeters/2
inches in diameter), but they differ greatly in wall thickness: ca. 2 centi-
meters (0.75 inches) vs. 36 centimeters (14 inches). Therefore, both cavi-
ties have the shape and size of a natural nest cavity, but only one has the
wall thickness of a natural nest cavity. They are located side by side, and
each contains two temperature sensors/recorders, positioned in the cen-
ter of the cavity, 20 centimeters (8 inches) from cavity top and bottom.
There is also a temperature sensor/recorder mounted in a shady spot
between the two structures for measuring ambient temperature. Our aim
is to compare the temperature dynamics inside both cavities across two
years, for one year with each cavity unheated and for one year with each
containing a 40- watt heating element, installed inside to simulate the heat-
ing it would have if it were occupied by a 2
- kilogram (4.4- pound) colony
of honey bees.
The graph in Figure 9.6 shows the temperature readings from the two
cavities and the ambient temperature sensor for two weeks in April 2018.
It shows that during this early- spring time period, the readings for the tree
cavity are far more stable than those for either the ambient temperature
or the hive box. We also see that the latter two sets of readings are nearly
identical, though on sunny days the hive box interior became warmer than
the air outside it (ambient). Furthermore, we see that during every night,
the tree cavity was warmer than the box. It is too early to know what this
study will tell us overall, but already it has revealed a stunning difference
in the stability of the nest- cavity temperatures experienced by wild colo-
nies that live deep inside massive trees vs. managed colonies that reside in
thin- walled hives.
Seeley.indb 230
2/21/2019 8:07:56 AM
Temperature Control 231
Ambient
Upper box
Upper tree
35
30
) 25
20
15
10
mperature (˚CTe 5
0
–5
4/8
4/10
4/12
4/14
4/16
4/18
4/20
Day in April, 2018
Fig. 9.6. Top: The two artificial nest cavities being used to compare, over a year,
the microclimate inside a thick- walled tree cavity and a thin- walled wooden box
of the same dimensions and with the same size entrance opening. Bottom: Sample
of temperature recordings from inside the thick- walled tree cavity and inside the
thin- walled wooden box and from outside both structures.
Seeley.indb 231
2/21/2019 8:07:57 AM
232 Chapter 9
The best insulation in the world will not, however, keep a living system
warm unless the system generates heat, so it is not surprising that the
adaptations of honey bee colonies for heat retention are complemented by
an impressive capacity for heat production. A respectable amount of heat
is generated by the resting metabolism of brood and adults: about 8 and
20 watts per kilogram (3.6 and 9 watts per pound), respectively, at the
brood- nest temperature of 35°C (95°F). However, by isometrically con-
tracting their flight muscles—so they burn energy without producing