Gabon
Location
of Gabon
Unlike surrounding countries,
Gabon has a stable government, a low human population
density, and large tracts of intact and pristine
forests. Consequently, Gabon may well be the stronghold
of the forest elephant in central Africa, and it
is therefore imperative that this population is
protected.
We will be starting two research
projects in Gabon during 2007.
The first will be in Loango
National Park. In common with the other National
Parks in Gabon, all very recently created (2002),
effective legislation is still lacking. Consequently,
a Chinese company has been granted an oil concession
that includes 85% of Loango. Oil exploration and
extraction could have severe effects on the elephants
and other wildlife. Exploration involves seismic
testing, which may be harmful to elephants because
they respond behaviorally to both seismic (O’Connell-Rodwell
et al. 2006) and low frequency acoustic signals
(Langbauer et al. 1991) and may use both for long-distance
communication. In addition, the habitat destruction
and infrastructure development (camps, roads etc.)
that result from oil exploration and extraction
could affect elephant behavior and habitat use.
In early 2007, we will be mounting Autonomous Recording
Units (ARUs) in the area that will record elephant
calls and other acoustic signals such as gunshots
and dynamite activity over a period of several months.
In addition, the ARUs will record seismic disturbances
from oil exploration. Using these data, we will
quantify disturbances from oil exploration and the
potential impacts on habitat use by forest elephants
in Loango NP.
Our other project will center
on Ivindo National Park. Ivindo is home to Langoue
bai where gorillas, forest elephants, and other
wildlife congregate to drink water and feed on plants
that are rich in minerals. We plan to begin fieldwork
in the bai in summer 2007 with an intensive 2-3
month study during which ARUs will record elephant
calls and other sounds such as gunshots. We will
also make visual observations and collect data on
a number of factors, such as rainfall and human
activity, that might affect elephant use of the
bai. Both gunshots and elephant calls will be extracted
from recordings and the spatial and temporal patterns
of elephant activity will be related to both ecological
(e.g., rainfall, mineral availability in bai waters)
and human variables (distance to human settlement,
frequency of gunshots). The data will also be used
to test our existing model for estimating elephant
densities from the number of calls (link to Dzanga
sections), this time using a bai with a lower –
and more typical - elephant density. We will also
deploy ARUs to monitor elephant and human activity
at bais in two or three other National Parks in
north-eastern Gabon: Minkebe, Mwagne, and/or Batéké.
Each of these locations will also be sampled for
the mineral composition of water in pools in bais
and data on rainfall will be collected over the
sampling period of 2-3 months. Our objective is
to further understand what factors affect the use
of bais by forest elephants.
References
Langbauer, W. R. Jr., K. Payne, R. Charif, E. Rappaport,
and F. Osborn. 1991. African elephants respond to
distant playbacks of low-frequency conspecific calls.
Journal of Experimental Biology 157: 35-46.
O’Connell-Rodwell,
C. E., J. D. Wood, T. C. Rodwell, S. Puria, S. R.
Partan, R. Keefe, D. Shriver, B. T. Arnason, and
L. A. Hart. 2006. Wild elephant (Loxodonta africana)
breeding herds respond to artificially transmitted
seismic stimuli. Behavioral Ecology and Sociobiology
59:842-850
|
Dzanga
Bai, Central African Republic
 |
| Location
of Central African Republic's Dzanga National
Park. |
 |
| Forest
elephants in the Dzanga clearing. |
|
In
2000-2002, ELPs research efforts were focused at
the Dzanga forest clearing where up to 100 elephants
gather daily. Andrea
Turkalo, a member of ELP, has studied the elephants
in this clearing for over 15 years via support from
the Wildlife Conservation Society. During this time,
Andrea has identified over 3000 elephants using
the clearing. Thus, Andrea’s expertise on
the elephant population at Dzanga and the high concentration
of elephants made this site the ideal location for
comparing elephant vocalizations with counts of
elephants in the clearing (Figure 1 below) and for
determining the distance in which elephant calls
travel. Calling elephants were identified by localizing
the source of sound via an acoustic array (See “Tools
and Techniques”). Recording units, placed
along a 1 kilometer line going into the forest,
recorded the progressive reduction in sound intensity.
This information has allowed a statistical model
to be developed to predict the area over which elephant
calls can be detected surrounding a single recording
unit. Results from Dzanga suggest that low frequency
calls could be detected up to 860 meters corresponding
to 2.3 square kilometer area of detection surrounding
a recording unit.
In addition, recordings
of elephant vocalizations were gathered at Dzanga
from individual elephants of various sex/age classes
and in various behavioral contexts. These data are
currently being used by PhD student Mya
Thompson to identify quantifiable measures of
elephant calls that can be used to identify behaviors
(e.g., mating or reproductive condition) and/or
age/sex class of calling elephants. This information
can then be used to determine demography (age and
sex) and population health (via reproductive behavior)
in long recordings made where elephants cannot be
observed |
