Temperature and the Reef Aquarium

I know what you have to be thinking… An article about
temperature. For Pete’s sake, reef aquaria have been popular
for at least 20 years in the U.S., and one would think we’ve
got a handle on proper temperature values. Just keep the water
temperature between 70° F and 80° F, right?

Now, imagine this scenario: One visits the doctor’s
office, and it is time to get your temperature taken. The nurse
waves the thermometer in the air, and records the number. She has
to be kidding – that’s not representative of your
temperature! But this is analogous, in some cases, to what
we’re doing when we simply make a water temperature
measurement in an aquarium! Granted, the water temperature will
be valid and truly represents the water at that particular spot
and likely the temperature of most corals within the aquarium.
However, there may be coral ‘hot spots’ if one is
lighting the captive reef with high wattage lamps using pendent
fixtures (especially those pendents made of aluminum and/or with
a polished internal reflective surface). Not surprisingly, those
corals most affected will be those directly under and closest to
the lamp and they can be much warmer than the ambient water
temperature indicates.

The implications are obvious. Coral bleaching could possibly
occur in a few corals while the simply measured metric – water
temperature – indicates conditions are within an acceptable

This article will convey the results of simple experiments and
will combine this information with important results recently
published in the respected journal Limnology and
. We will also briefly discuss observations made
of pendent luminaire temperatures and how to potentially limit
thermal transfer from the lamp to the aquarium.



A 100-gallon RubberMaid tub (holding 75 gallons of natural
seawater) was maintained in a darkened, air-conditioned room. Two
400-watt metal halide lamps in pendent luminaires provided light
to captive corals (mostly ‘lobe coral’ – Porites
– and Pavona varians). These corals were
maintained for an experiment examining the effects of red light
wavelengths on zooxanthellae. While checking different parameters
for standardization, it was discovered that the two luminaires
were operating at different temperatures. Not only that, but the
corals directly beneath the one of the light fixtures appeared to
be warmer (sometimes much warmer) than the water temperature.

This procedure utilized an infrared thermometer to monitor
temperatures. While there was little doubt that temperatures
measured on the luminaires were correct, there was some question
about the ability of this device to accurately gage temperatures
of objects submerged in water (albeit shallow water. Only 3 to 6
inches of water covered these corals. The bottom of the light
fixture was another 6 inches or so above the water). This
information (concerning temperature) was presented at the
Washington DC MACNA. The testing protocol was questioned during
the presentation.

Although I felt comfortable with the information, I decided to
repeat the procedure. Two experiments were conducted – one
experiment monitored coral and water temperatures with
‘good’ water motion, and the second focused on
coral/water temps with ‘reduced’ water flow. Instead of
using an infrared thermometer, I used two thermistors and a data
logger from Spectrum Technologies. Holes just slightly larger
than the outside diameter of the thermistor were drilled with a
masonry bit into the bottom of thePorites corals. These
holes extended to within an estimated 3-4 mm of the top coral
surface. The thermistor fit snugly into the hole and measured
temperature just below the porous tissue/skeletal interface of
Porites corals. The corals rested on eggcrate material
which conveniently allowed water circulation (provided by a
5-gallon Carlson Surge Device, or CSD) around, over and under the
corals. It also allowed the thermistor wiring to remain under,
and out of the way, during the experiment. The data logger was
programmed to measure temperature and PAR every minute for the
duration of the 3.0 – 3.5 hour experiments (this article will
discuss only the temperature readings). Once everything was in
place, the data logger began recording information and the 400
watt lamp was ignited a few minutes later. Temperatures were
measured within the coral and immediately below it (in a shadow).
Water temperature was also monitored with a calibrated mercury
‘lab’ thermometer. Accuracy of the two thermistors had
previously been checked by placing the two sensors into a stirred
and heated water bath and observing delta temperature across a
broad range. The sensors’ responses as displayed in the data
logger LCD were compared to a calibrated laboratory thermometer.
Both thermistors were in agreement with the ‘lab’
thermometer to within 0.1°C (0.18° F) from ~20°C
to ~40ºC. Engineers at Spectrum approved the thermistors and
cords for immersion in seawater for a period not to exceed ~24


Coral temperature rose more quickly than water temperature
under the conditions of these experiments. Figure 1 charts the
temperatures of the coral and water with ‘good’ water
motion. Figure 2 demonstrates temperatures with ‘reduced’
discharges from the Carlson Surge Device.


Figure 1. Time course temperature of
aquarium water and a coral beneath a high wattage lamp. These
are with ‘good’ water flow.



Figure 2. Same as Figure 1, except with
‘reduced’ water motion.


It is apparent that the coral warmed more quickly than the
water during the course of both experiments, and the temperature
differed by as much as 1.6°C (2.88°F, with
‘reduced’ flow) and 0.8C (1.4°F, with ‘good’
flow). Note that, under the circumstances of these experiments,
it was possible to modify only the periodicity of the CSD
discharge, and not the velocity (which, of course, remains the

A digital water velocity meter recorded velocity at a maximum
of 42 cm/sec (~17 in/sec) and a period of 30 seconds for
‘good’ flow and ~1.5 minutes periodicity for
‘reduced’ flow. The periodicity of the ‘good’ CSD
discharge is approximately that seen on Hawaiian reefs on a calm
day, but velocity is 3 times that experienced by corals under
‘normal’ conditions. It is not unreasonable to have
anticipated that the coral temperature and water temperature
would have remained the same – they did not.

This raises the question about the effects of solar warming on
corals on natural reefs. Katharina Fabricius (a name familiar to
serious hobbyists for her work with soft corals) reports in a
2006 article that corals can indeed become warmer than the
surrounding water. Further, she reports that the color of the
coral can determine how it absorbs irradiance and hence influence
the magnitude of temperature gain. The implications of her work
are staggering – seawater surface temperature has been the metric
to determine thresholds of coral bleaching. Fabricius’ work
suggests that other factors are in play, and coral pigmentation
could play an important role in determining if a coral survives
an El Nino warming episode.

Luminaire Construction and Heat Transfer

Luminaire construction can make a profound difference in
focusing of light and reflection of ultraviolet radiation and
infrared radiation. This became very apparent during recent
experiments. I used two pendent luminaires – one had a painted
interior, and the other internal surface was polished aluminum.
The polished aluminum, possibly along with minor differences in
pendent construction geometry, reflected visible radiation at a
much more efficient rate than the pendent with the white paint
interior. Apparently, it also reflected infrared radiation at a
much higher rate (as evidenced by higher temperatures of corals
kept under it).


Clearly, this arena requires further research before we have a
good understanding of thermal transfer from lamps to corals, and
how we can avoid problems. For hobbyists, it is not possible to
utilize an infrared thermometer in most cases and it is certainly
impractical to use thermistors to monitor temperatures. At this
point, the best we can do is look at those things we can do to
minimize heat transfer to the aquarium. This leads us to a
discussion of other things noted during this series of

The luminaire with the polished interior surface was
not originally equipped with an acrylic ‘splash
guard.’ This is unfortunate since no protection is afforded
against lamp breakage due to thermal shocks caused by splashing
water (it also allows potentially harmful ultraviolet radiation
into the aquarium). The lens or splash guard will also absorb
some infrared radiation. In order to standardize UV radiation
(that is to say eliminate it) during the experiments, an
appropriately sized lens was cut from a sheet of Lexan and
installed on the luminaire.

With the chance of updraft through the lighting fixture
eliminated by the installation of the lens, the temperature of
the luminaire quickly rose and was as much as 97.7º C
(209°F) at the top. This problem was overcome by simply
drilling five ¼” holes in the lens and thus allowing
airflow upwards through the luminaire and venting the hot air
through the top. Figures 3 and 4 are representative of
temperatures with a ‘closed’ and ‘open’
luminaire. Figure 3 shows the luminaire temperatures (°F)
‘before’ the holes were drilled and Figure 4 demonstrates
°F ‘after’ the holes were drilled, and the resulting
drop in luminaire temperature.


Figure 3.


Figure 4.

Perhaps the most dramatic observation was the superior
efficiency of the polished aluminum surface as opposed to the
luminaire with the internal surface painted white. The
performance of the aluminum surface was great enough to allow
dropping a full wattage level (i.e. 400-watt to 250-watt) and
still get comparable light intensity. Imagine the reduction in
the utility bill this ‘watt shaving’ could provide – all
by simply choosing a lighting fixture with an efficient
reflective surface.


These observations go well beyond a coral simply being a few
degrees warmer than the ambient water temperature. It goes to the
heart of one of the true challenges of maintaining a reef
aquarium – that of expense. It is possible, under some
circumstances, that proper selection of lighting equipment could
in turn reduce or perhaps even eliminate the cost of purchasing
and maintaining a water chilling device.

Ironically, in our attempts to create what we perceive as
‘natural’ and ‘proper’ conditions for symbiotic
invertebrates, it is the use of high wattage lamps that
potentially push the boundaries of corals’ thermal tolerances
and could spell the doom of those animals unfortunate enough to
be within the luminaire’s ‘hot spot.’. The solution
is, in some cases, not a chiller but an effective and efficient
lighting fixture resulting in utilization of lower wattage

It should also be obvious that ‘good’ water motion can
ameliorate coral temperature increases. This is a double-edge
sword since submersible pumps can add heat to the aquarium water.
By the same token, it is potentially possible that low velocity
water motion resulting in poor flushing could result in
temperatures higher than those noted here. However, it is not the
intent of this article to either examine all scenarios in which
temperature could become a critical factor or look at other
available options for reducing aquarium temperature. The scope is
to establish that corals can, under certain conditions, gain heat
from lighting sources. Corals directly under high wattage (400 or
1000) should be monitored closely for signs of decreased
pigmentation. What we once attributed to bleaching caused by high
PAR values may actually be a result of a synergistic effect of
the ‘terrible 3’: Light, heat and (if the lamp isn’t
shielded as it should be) UV radiation. However, it is possible
that temperature alone could trigger a bleaching event.

Perhaps this article reflects the state of reef-keeping.
We’re focusing on minutiae – we have solved most major
problems. But it is the little things that could make the
difference in a thriving coral or a dead one, or potentially the
difference between acceptable and outrageous monthly maintenance
costs in the form of electrical utility bills. We can
congratulate ourselves on the advances the hobby has made over
the last two decades. By the same token, it reflects that we have
a lot to learn.


  1. Fabricius, K., 2006. Effects of irradiance, flow, and
    colony pigmentation on the temperature microenvironment around
    corals: Implications for coral bleaching? Limnol. Oceanogr.,
    51(1): 30-37.
  Advanced Aquarist
Dana Riddle

 Dana Riddle

  (119 articles)

Leave a Reply