Sviluppo e validazione di un modello d'idoneità ambientale su una specia minacciata: il leone in Ghana
Author(s)
Petrozzi, Fabio
Date Issued
March 20, 2012
Type
Doctoral Thesis
Abstract
During the past century, African lion (Panthera leo) suffered a strong population decline
because of several factors as loss of habitat (exploitation of the land for agricultural
purposes, habitat destruction due to wars, extraction of raw materials from the
underground, etc.), and direct persecution, both as an "harmful" animal for human
activities, both for the use of some parts of his body, related to tribal rituals and / or
animism.
Studying the estimation of African population, since the second half of the past century
until today, we can notice a dramatic decrease. In fact, at the beginning of the XX century,
in the whole Africa, lion population was about 400,000 units; in 1990 it was estimated
between 30,000 and 100,000 lions.
Historical range occupied the whole Africa, except for the Sahara and the rainforests in
Congo; the current range, highly fragmented, is estimated about 3,000,000 km2, that is less
than a quarter of the historical distribution.
One of the last group of Western lion population is represented by the small group in the
Mole National Park (perhaps 12-28 lions), which would be almost the entire population of
them in Ghana.
Because of its proximity to other protected areas in neighboring countries, it is extremely
important the role played by this park, a vast protected area (about 4,800 km2) that can
ensure the conservation and management of lions, once they are identified and monitored.
Lion population in Mole National Park has never been studied, with the exception of some
recent general studies about the presence of carnivores.
The aim of this research is to create and validate an environmental suitability model for the
lion.
The model has been developed by analyzing historical sightings of lion in the park and
linking them with the types of habitats in the area, then building a new model and
examining where the lion is more likely to be.
This was done considering also the different habitat where the species lives in West Africa,
and particularly in the Mole National Park, whose extension is mainly occupied by very
closed savannah (Western savannah) and surrounded by forests, a high discriminating
factor.
Once the model was created, I proceeded to validate it with two methods: By using camera
traps, and by carrying out transects on foot, or by car (4 wheel drive) to collect direct and /
or indirect records, as tracks of a lion (footprints, remains of predation, scats).
Through transept, both by car and on foot, we covered in total about 900 km.
About transects done by car, the vehicle proceeded at an average speed of 20 km/h, with 5
observers. The transects carried out on foot had a range of distance from 2 to 4 km each.
Camera-traps were placed in 18 different areas of the park, both at the regional level and in
terms of potential environmental suitability, for a total amount of 3600 days / trap and 23
stations, shifting 5 camera traps.
The places where they were located were mainly in areas near ponds, rivers or watering
areas, taking advantage of the fact that, especially during the dry season, the animals tend
to concentrate in these areas.
The statistical analysis, applying both backward stepwise model, and Von Bertalannfy
model, resulted in five variables that are related to the presence of the lion and that show
how population selects positively closed savannah and areas near forests - (P <0.033),
model backward stepwise AIC = 75,006 and Von Bertalannfy model AIC = 75,117; ΔAIC =
0.111-, avoiding the open savannah. They also demonstrate how the presence of the
population in the park is positively influenced by the rainfall - (P <0.001), with models
backward stepwise (AIC = 48,663) and Von Bertalannfy (AIC = 44,714) (ΔAIC = 3949)
which have a similar result-, by the presence of water basins - (P <0.045), with model
fitting almost similar in both models (AIC = 77,114 and 77,361 respectively; ΔAIC =0.247,
probably because potential preys prefer those areas.
These distinctive features are somewhat discordant with those of lion populations in
Southern and Eastern Africa. In fact, these ones prefer open places (open savannah), and
they usually don’t require the presence of plenty of water for their survival.
The GIS model, in which all positive variables of the statistical analysis have been used,
corroborated these results, the percentage of historical sightings that falls within high
environmental suitability is about 88, 7% of total sightings (P <0.002). Furthermore, the
development of this model indicated the suitability areas, divided into three categories
(high, medium, and low).
We chose the places where carry out transects and locate camera-traps from these three
categories of suitable areas.
The transects permitted to record a lion footprint, near a river, and a roar, while a cameratrap
recorded a video of a young lion moving in a densely wooded area.
Study results have a strong implication on the conservation of lion in the park, especially
because its population is limited and because of the strategic geographical location of Mole
NP.
Furthermore, this methodology may be applicable in other areas where the species (or
comparable species from an ecological point of view) has elusive habits, as in the study
area, in which any record of the cat was not recorded for 11 years (1998 -2009).
Additional information
Dottorato di ricerca in Evoluzione biologica e biochimica
File(s)![Thumbnail Image]()
Name
fpetrozzi_tesid.pdf
Size
23.26 MB
Format
Adobe PDF
Checksum (MD5)
cb087865bb4a19d0e4335aab2f4d45c8