Bio control of Opuntia aurantiaca (Prickly Pear)

Post-release evaluation of the effect of Dactylopius austrinus (cochineal)
and Cactoblastis cactorum (Cactus moth) on Opuntia aurantiaca (Jointed
Cactus)



Compiled by: William Keith Knorr; Marloth Park Conservancy



September 2013






Abstract.

Biological control agents have been used in certain
circumstances to reduce the extent and distribution of alien
invasive weeds. It is unfortunate that biological agents are
normally used as a last resort and then they are expected to
resolve the problem in a short period of time. It must be borne
in mind that in many cases biological control may still require
the assistance of conventional weed control methods (e.g.,
chemical and or mechanical control). In most cases biological
control has proved to be effective and there is generally an
ongoing and more importantly a continual financial cost saving.
In some cases this may be difficult to quantify due to a lack of
post release quantitative data.

The primary aim's of biological control is to reduce both the
chemical contamination of the environment as well as reduce
mechanical control which often results in erosion. The resultant
benefit assists in ecosystems being restored to their original
native state.

Biological control in many cases results in a cyclical
interaction between alien weed and the agent. This can be due to
climate or the typical predator prey relationship. Occasionally
the biological control agent can be so successful that manual re-
infestation of the agent is sometimes required to maintain
control of the alien weed.

Only by actively following up and quantifying the results on a
regular basis, can one determine the effectiveness of the
released agent in a particular area. It can not be assumed that
effectiveness or lack thereof in one area, will confirm the same
result in another.

Not only should the effectiveness of the biological control
agent be evaluated, but the response of non-target species
should also be evaluated as it is possible that other alien
pioneer plants may start to infect the area.

Without quantitative assessments being carried out prior to the
release of biological control agents and also post release
evaluation, the impact on the ecosystem can not be correctly
evaluated.

This is one of the reasons why post release evaluation is so
critical.







Keywords: Agent; Biological control; Cladode; Host-
specificity; Invasive plants; Oligophagous;
Phytophagous










Introduction.

Many alien plants do not become invasive due to existing climate
conditions restricting their survival and sometimes also the
local insects predating on them. In other cases some alien
plants become invasive because they have no natural enemies to
regulate their numbers.

The use of biological control agents is used to reduce the
density of alien plant infestations where chemical and
mechanical controls have not proved to be cost effective. There
are often varying degrees of doubts from Governments, NGO's and
the public as to the effectiveness and the risk of introducing
biological agents to control alien plants. It appears that only
mistakes are remembered and not the successes.

This in part is due to "people forgetting" the effectiveness of
the biological agents role in the reduction of the alien plants
where it has been extremely successful.

Biological control agents are usually very host-specific and
result in less damage to the environment than either chemical or
mechanical control methods.

Careful surveys of infested areas should be carried out and
documented prior to an agent being released. Surveys should
again be carried out after the agent has been released and
documented so that the quantitative data on direct and indirect
benefits or lack thereof can of can be fully analysed.




Brief history.

It is generally accepted that Opuntia aurantiaca Lindley arrived
in South Africa from the United Kingdom during the 1840s. O.
aurantiaca originated in South America however its exact origin
has not been comprehensively proved (H.G. Zimmermann, 1974). It
was only in the early 1900s that it became an invasive weed of
national importance in South Africa.

Until the late 1930s only mechanical and chemical controls were
being used and was not adequately keeping the infestation under
control. At this stage, live stock, grazing and agricultural
land was under serious threat.

During 1933 the Phycitid Moth, Cactoblastis cactorum Bergroth
was released. This particular Moth was hugely successful in
Australia resulting in the collapse of Opuntia stricta Haworth
(Dodd, 1940; Mann, 1970). In South Africa the same success was
not achieved due to predation of the eggs by ants (Robertson,
1985a) and also the predation of the larvae by baboons. This
does not necessarily mean that C. cactorum was a failure, it was
just not as successful as it was in Australia.

The release of the cochineal insect, Dactylopius austrinus De
Lotto in 1935 resulted in a wide spread reduction of O.
aurantiaca. It was so successful that during 1946 biological
control was officially the recommended practice (V.C. Moran &
H.G. Zimmermann, 1991). The reduction of O. aurantiaca was so
successful, it resulted in that the numbers of D. austrinus
insect rapidly declining. This also proved the extent of the
insects oligophagous nature. As a biological control agent D.
austrinus is also more successful in the drier parts of the
country (V.C. Moran & H.G. Zimmermann, 1991).

However during the late 1940s O. aurantiaca started to spread
again and mechanical removal was restarted. This did not prove
to be successful and from 1957 landowners were compelled at
great economic cost to utilise chemical controls (Zimmermann et
al.,1982).

The democratically elected government of 1994 decided that the
cost was to great and therefore stopped the supply of the
chemical herbicide. It was thought that O. aurantiaca densities
would rapidly increase, however this was not the case. What did
occur is the typical sine wave graph that is representative of
many predator/ prey relationships (Zimmermann & Malan, 1989).

In some cases it can occur that the density of O. aurantiaca
becomes too low to sustain a viable population of D. austrinus.
The D. austrinus crawlers are mainly wind dispersed and as the
O. aurantiaca is low growing the distances between plants for
effective wind dispersal may exceed a minimum critical level
resulting in the local disappearance of the cochineal.

If a resurgence of O. aurantiaca starts to be detected again, it
is normally a relatively simple exercise to reintroduce a number
of cochineal infected cladodes in that area. At present O.
aurantiaca is considered to be under control.




Hypothesis.

To confirm that the area which was originally badly infested
with the O. aurantiaca plants was now under better control since
the biological control agents D. austrinus and C. cactorum had
been released. It is assumed that any remaining plants that may
be found, would generally be infected with one or both of the
above mentioned biological control agents, indicating that the
biological control programme is effective. Remaining plants
found should be showing signs of stress and will therefore have
a lower survival and reproductive rate than plants which have
not been infected by one or both of the above biological control
agents.




Location of the research undertaken.

The post evaluation research took place on 4 September 2013 at
the Sam Knott Nature Reserve which is part of the Great Fish
River Reserve, located in the Eastern Cape Province, South
Africa.




Materials and methods.

The basic materials used were, GPS, camera, 50 meter tape, 5
meter tape, cactus tongs, data sheet, note book and pen.

A transect of 50 meters in length was laid out and 2 meters on
either side of the transect was checked for the presence of O.
aurantiaca. GPS coordinates were taken and recorded for the
beginning and end of each of the 50 meter transect. (A total of
200m2)

The distance's between each O. aurantiaca plant/ cladode was
then recorded.

The identified O. aurantiaca were also further classified into
three different sizes, namely;

"Size A" = a single cladode
"Size B" = two to five cladodes
"Size C" = more than five cladodes

Further noted was the presence or lack thereof of the biological
control agent and also if present the type of biological control
agent on the cladode. i.e. Dactylopius austrinus and/or
Cactoblastis cactorum.

A total of 5 transects were carried out by five different teams.
It is unfortunate that the data collected by one team does not
provide the correct spacing between individual plants, however
it will be used in this report and overall it should not impact
negatively on the result.













Results.

The results were all recorded and a spread sheet compiled. The
following figure, Fig. 1.1 shows the five transects and the
results tabulated to indicate the plant density per square meter
and also the percentage of infection of the plant by the
individual biological control agents.




Plant density per square meter and percentage of biocontrol
infected plants

Transect "Size A/m2 "%Coch "%Cact "Size B/m2 "%Coch "%Cact "Size
C/m2 "%Coch "%Cact " " " " " " " " " " " " "1 " 0.51 " 0.98 " 0 "
0.27 " 0 " 0 " 0.02 " 50 " 50 " "2 " 0.19 " 0 " 0 " 0.02
" 0 " 0 " 0.02 " 0 " 0 " "3 " 0.675 " 0 " 0 " 0.125 "
0 " 0 " 0.035 " 0 " 0 " "4 " 0.61 " 0.82 " 0 " 0.035 "
0 " 0 " 0.025 " 20 " 0 " "5 " 0.2 " 0 " 0 " 0.14 " 0 "
0 " 0.02 " 0 " 0 " " " " " " " " " " " " "Fig. 1.1 Coch =
Dactylopius austrinus; Cact = Cactoblastis cactorum
Size A /m2 = 1 cladode; Size B = 2 to 5 cladodes;
Size C = more than 5 cladodes


From Fig. 1.1 it can be seen that the single cladode "Size A"
has a very low incidence of infection by D. austrinus and no
infection by C. cactorum. "Size B" two to five cladodes, had no
biological control infection. "Size C" more than five cladodes,
had the highest percentage of biological control agents and was
also the only size to be infected by both D. austrinus and C.
cactorum.



The following graph shows the density of O. aurantiaca per
transect. It shows that "Size A" the single cladode, is the most
numerous found, "Size C" which indicates plants with more than
five cladodes being the least numerous found. This indicates
that the majority of the "Size A" plants do not manage to grow
into the larger "Size C" plants. The results indicate that the
biological control agents become more effective as the O.
aurantiaca increases in size.





The following graph shows the percentage of biological control
taking place in "Size A" the single cladode. Only D. austrinus,
the cochineal was found on those single cladode's and then at a
density of less than one percent. This indicates that the vast
majority of the single cladode plants were not being infected by
the biological control agents.







The following graph shows the percentage of biological control
taking place in the "Size C" plants with more than five
cladode's. Both biological control agents, D. austrinus and C.
cactorum were found on these slightly larger plants. The
percentage of infection was much higher. In the previous graph
there was less than one percent of the plants infected and here
we have up to 50 percent of the plants infected. This indicates
that the biological control is much more effective the larger
the plant becomes and also showed that both biological control
agents were active on the larger plants.







Discussion.

Even though there was little evidence of the biological control
agents, they were effective in reducing the plant infestation,
from a distance of only a few meters it appeared that there was
O. aurantiaca whatsoever. However a much closer inspection
revealed that there was indeed some very small O. aurantiaca
present, the vast majority being single cladode's.

Their presence was at a very low density with an average of a
single cladode per two square meters. Interestingly, less than
1% of the single cladodes were actually infected by one of the
biological control agents.

On the larger plants there was up to a 50% infestation of the
biological control agents (both agents being present), however
the density of these plants was less than 0.1%.

This indicated that the smaller plants were generally not being
infected by the biological control agents but that the larger
plants were being more frequently infected. This was in
contradiction to our hypothesis, where we expected to see the
majority of all sizes of plants being affected by either one or
both of the biological control agents.

The factor that seems to limit the infestation density of the
biological control agent is distance. D. austrinus is dispersed
mainly by wind. As the current plants are very small with the
largest on our transect only about 15 cm tall, it is obvious
that the D. austrinus cannot be wind dispersed very far.
According to Gunn where plants were one meter or smaller, less
than 20% of the cochineal crawlers were blown more than 2.5 m
from the source plant. He placed crawlers on an infected plant
on top of a 5 m high pole, crawlers were then found up to 100 m
away from the source (Gunn, 1979). This confirms why the
crawlers were not spreading in the area under investigation.

Upon investigating the surrounding area, we found a few O.
aurantiaca which were larger than 1 m in height and it was found
that almost all these plants were infected by both D. austrinus
and C. cactorum. These plants were approximately 20 m away from
the area that we were investigating.

This indicates that the larger O. aurantiaca are acting as sinks
for the biological control agents. This means that these larger
plants should not be removed mechanically or herbicides sprayed
on them, as to do so would result in the loss of the sink for
the biological control. Should this occur then the smaller
plants would merely grow and ultimately replace those which were
removed. More importantly, should the biological control agents
have been eliminated due to the removal of the larger plants,
then there will be no biological control agents present to
reduce the plant numbers. In a short period of time the area may
become infected by a high density of large O. aurantiaca plants.




Conclusion.

This indicates that the biological control agents will never
fully eliminate O. aurantiaca. A typical predator/ prey
relationship exists between the biological control agents and O.
aurantiaca. At present the current infestation of O. aurantiaca
is well below any economic or ecological damage threshold.

The only slight economic conflict that exists is that the
farmers of prickly pear must now spray their crops to keep both
D. austrinus and C. cactorum at bay. Overall this is a small
price for the farmers to pay as the alternative would have
resulted in country wide infestations of O. aurantiaca which
would have resulted in costly chemical and/ or mechanical
removal programmes. However the result would have been very
different if the prickly pear farmers had objected sufficiently,
resulting in the biological control programme from not going
ahead.

As a point in passing, a similar scenario exists with the Acacia
mearnsii (Black wattle) and this was resolved by using a
biological control agent that only stops the spread of the trees
but does not affect existing trees.

D. austrinus and C. cactorum are not entirely host specific
biological control agents, however they only attack Cactus genus
of which there are have been many different species introduced
to South Africa. As there are no indigenous cactus species in
Africa, the use of the above mentioned non specific biological
control agents is not a problem.

It is therefore essential that there is always a follow up done
on a regular basis wherever biological control agents have been
released. By doing this we will over a period of time have a
historical record of the effectiveness or lack thereof of the
biological control agents. This will allow us to decide whether
we need to manually re-infect the area with biological control
agents. It is also possible that newer and perhaps better
biological control agents have been developed which could then
also be introduced.

It has been already been identified that D. austrinus has a
higher survival rate in warm climates with low rainfall. Parts
of the Eastern Cape have cool and relatively wet climates, so
should a more effective biological control agent be found that
is more suitable to that type of climate, it may then be prudent
to release it in that area.

In most cases biological control cannot merely be implemented
and then walk away from it. There are many variables which could
inhibit its effectiveness, and if these are not addressed in
time then the whole biological control programme could be placed
in jeopardy. As an example a farmer who perhaps thinks that he
can speed up the process of eliminating a particular alien weed
could decide to spray it with herbicides and or insecticides
thus unknowingly eliminating the biological control agents.

Ecosystems are dynamic and when the presence of one plant is
removed often a replacement will take its place. Again this
needs to be monitored as the replacement plant could be another
alien plant and not necessarily an indigenous pioneer plant.

Another aspect of biological control, especially in this
particular scenario (Game reserve with Black Rhino), is that
there is little or no danger to people when compared to the
amount of time that would need to be spent in this potentially
dangerous environment on mechanical and or chemical removal.

The management of the Sam Knott Nature Reserve, need to be
advised that the area inspected is under good biological control
and no further action is required. A follow up investigation
will be carried out in the future.




Future research recommendations.

It appears that the majority of the biological control programme
focuses on the acquiring of the correct agent, which is
necessary and to be expected. However should this not be
properly followed up after the agents release by long term
monitoring, then the successfulness of the biological control
programme is likely to be compromised.


Prior to the introduction of any biological control programme it
is essential that at a minimum the following be carried out:

Current status of the vegetation (alien and indigenous) be
assessed and quantified.

The current use of the land. (natural or altered by man)
how, etc?

The history of the land and man's involvement. E.g.
farming, mining etc. This may have resulted in an
unnatural disturbance in the past.

Previous use (if any) of chemical and mechanical control
and its effectiveness.


After the biological control agent has been released it is
essential to follow up on the effectiveness or non-effectiveness
of the biological control agent. This will enable us to more
accurately predict the effect of the biological control agent
and also how non target species respond to the removal of the
alien plant. The outcomes and results which are nearly always
positive to some degree will further enhance funding, create
employment and promote further use of biological control agents.




References.

DENSLOW, J.S. & D'ANTONIO, C.M. 2004. After biocontrol:
Assessing indirect effects of insect releases. USDA Forrest
Service, USA.

DODD, A.P. 1940. The biological campaign against Prickly Pear.
Commonwealth Prickly Pear Board Bulletin, Brisbane, Australia,
177pp.

MANN, J. 1970. Cacti naturalised in Australia and their control.
Department of Lands, Brisbane, Australia, 115pp.

GUNN, B.H. 1979. Dispersal of the cochineal insect Dactylopius
austrinus De Lotto (Homoptera: Dactylopiidae). PhD. Thesis,
Rhodes University, Grahamstown. 188 pp., (unpublished).

MORAN, V.C. & ZIMMERMANN, H.G. 1991. Biological control of
jointed cactus, Opuntia aurantiaca (Cactaceae), in South Africa.



ROBERTSON, H.G. 1985a. The ecology of Cactoblastis cactorum
(Berg) (Lepidoptera: Phycitidae) in relation to its
effectiveness as a biological control agent of prickly pear and
jointed cactus in South Africa. PhD. Thesis. Rhodes University,
Grahamstown. 181 pp., (unpublished).

ZIMMERMANN, H.G. 1974. The biological control of jointed cactus
in South Africa. Papers presented at the First National Weeds
Conference, 1974, at Pretoria, South Africa. Pp. 204 - 211.

ZIMMERMANN, H.G. & MALAN, D.E. 1989. Population biology of a
jointed cactus infestation. Proceedings of the Conference of the
Weeds Sciences Society of South Africa, 1989, at Port Edward,
South Africa, 1p (unpublished).

ZIMMERMANN, H.G. & MORAN, V.C. 1982. Ecology and management of
cactus weeds in South Africa. S. Afr. J. Sci., 78: 314-320.