Insights for Integrated Conservation from Attitudes of People toward Protected Areas Near Hwange National Park, Zimbabwe

Abstract

Increase in human settlements at the edge of protected areas (PAs) is perceived as a major threat to conservation of biodiversity. Although it is crucial to integrate the interests of surrounding communities into PA management, key drivers of changes in local populations and the effects of conservation on local livelihoods and perceptions remain poorly understood. We assessed population changes from 1990 to 2010 in 9 villages located between 2 PAs with different management policies (access to natural resources or not). We conducted semi-directive interviews at the household level (n =217) to document reasons for settlement in the area and villager's attitudes toward the PAs. We examined drivers of these attitudes relative to household typology, feelings about conservation, and concerns for the future with mixed linear models. Population increased by 61% from 2000 to 2010, a period of political and economic crisis in Zimbabwe. Forty-seven percent of immigrants were attracted by the area; others had been resettled from other villages or were returning to family lands. Attitudes toward PAs were generally positive, but immigrants attracted by the area and who used resources within the PA with fewer restrictions expressed more negative attitudes toward PAs. Household location, losses due to wild animals, and restrictions on access to natural resources were the main drivers of this negative attitude. Profit-seeking migrants did not expect these constraints and were particularly concerned with local overpopulation and access to natural resources. To avoid socio-ecological traps near PAs (i.e., unforeseen reduced adaptive capacity) integrated conservation should address mismatches between management policy and local expectations. This requires accounting for endogenous processes, for example, local socio-ecological dynamics and values that shape the coexistence between humans and wildlife.

Full text

Contributed Paper
Insights for Integrated Conservation from Attitudes of
People toward Protected Areas Near Hwange National
Park, Zimbabwe
CHLOE GUERBOIS,∗†‡ ANNE-BEATRICE DUFOUR,‡ GODFREY MTARE,§∗∗ AND HERVE FRITZ†‡
∗UMR 7204 MNHN-CNRS-UPMC, Conservation des esp`
eces, restauration et suivi des populations, Mus´
eum National d’Histoire
Naturelle, 55 rue Buffon, Paris F-75005, France, email chloe.guerbois@univ-lyon1.fr
†CNRS HERD Programme–Hwange LTER, Main Camp research, Hwange National Park, BP 62, Dete, Zimbabwe
‡Universit´
e Lyon 1, CNRS, UMR 5558, Laboratoire de Biom´
etrie et Biologie Evolutive, 43 bd du 11 nov 1918, Villeurbanne F-69622,
France
§Zimbabwe Parks and Wildlife Management Authority, P. Bag 140 CY, Causeway, Harare, Zimbabwe
∗∗University of Cambridge, Geography Department, Downing Place, CB2 3EN, Cambridge, United Kingdom
Abstract: Increase in human settlements at the edge of protected areas (PAs) is perceived as a major threat
to conservation of biodiversity. Although it is crucial to integrate the interests of surrounding communities
into PA management, key drivers of changes in local populations and the effects of conservation on local
livelihoods and perceptions remain poorly understood. We assessed population changes from 1990 to 2010 in
9 villages located between 2 PAs with different management policies (access to natural resources or not). We
conducted semi-directive interviews at the household level (n =217) to document reasons for settlement in
the area and villager’s attitudes toward the PAs. We examined drivers of these attitudes relative to household
typology, feelings about conservation, and concerns for the future with mixed linear models. Population
increased by 61% from 2000 to 2010, a period of political and economic crisis in Zimbabwe. Forty-seven
percent of immigrants were attracted by the area; others had been resettled from other villages or were
returning to family lands. Attitudes toward PAs were generally positive, but immigrants attracted by the area
and who used resources within the PA with fewer restrictions expressed more negative attitudes toward PAs.
Household location, losses due to wild animals, and restrictions on access to natural resources were the main
drivers of this negative attitude. Profit-seeking migrants did not expect these constraints and were particularly
concerned with local overpopulation and access to natural resources. To avoid socio-ecological traps near
PAs (i.e., unforeseen reduced adaptive capacity) integrated conservation should address mismatches between
management policy and local expectations. This requires accounting for endogenous processes, for example,
local socio-ecological dynamics and values that shape the coexistence between humans and wildlife.
Keywords: perception, population growth, protected areas, rural livelihoods, social and cultural values, socio
ecological systems, socio ecological traps
Percepciones para Conservaci´
on Integrada a Partir de las Actitudes de las Personas hacia ´
Areas Protegidas cerca
del Parque Nacional Hwangem Zimbabwe
Resumen: El incremento de asentamientos humanos a la orilla de ´
areas protegidas (AP) es percibido como
una amenaza mayor para la conservaci´
on de la biodiversidad. Aunque es crucial integrar los intereses de las
comunidades circundantes en el manejo de las AP, los conductores clave del cambio en poblaciones locales
y los efectos de la conservaci´
on de sustentos locales y las percepciones permanecen pobremente entendidos.
Estudiamos cambios poblacionales desde 1990 hasta 2010 en 9 aldeas ubicadas entre 2 ´
areas protegidas con
diferentes pol´
ıticas de manejo (acceso a recursos naturales o no). Llevamos a cabo entrevistas semidirigidas
Paper submitted June 4, 2012; revised manuscript accepted October 24, 2012.
844
Conservation Biology, Volume 27, No. 4, 844–855
C
2013 Society for Conservation Biology
DOI: 10.1111/cobi.12108

Guerbois et al. 845
aniveldecasas(n=217) para documentar las razones del asentamiento en el ´
area y la actitud de los
aldeanos hacia las AP. Examinamos los conductores de estas actitudes en relaci´
on con la tipolog´
ıa de la casa,
sentimientos hacia la conservaci´
on y preocupaciones por el futuro con modelos lineales mixtos. La poblaci´
on
increment´
o en 61% desde 2000 hasta 2010, un periodo de crisis pol´
ıtica y econ´
omica en Zimbabwe. El 47% de
los inmigrantes fueron atra´
ıdos por el ´
area; otros hab´
ıan sido reubicados de otras aldeas o estaban regresando
a tierras familiares. Las actitudes hacia las AP eran generalmente positivas, pero los inmigrantes atra´
ıdos por
el ´
area y que usaban recursos dentro de la AP con menos restricciones expresaron m´
as actitudes negativas
hacia las AP. La ubicaci´
on de las casas, la p´
erdida debido a animales silvestres y las restricciones sobre el
acceso a los recursos naturales fueron los principales conductores de estas actitudes negativas. Los migrantes
que buscaban ganancias no esperaban estas restricciones y estaban particularmente preocupados con la
sobrepoblaci´
on local y el acceso a los recursos naturales. Para evitar trampas socioecol´
ogicas cercanas a las
AP (p.ej.: capacidad adaptiva reducida no prevista) la conservaci´
on integrada debe dirigirse a los desajustes
entre las pol´
ıticas de manejo y las expectaciones locales. Esto requiere un arreglo de cuentas para los procesos
end´
ogenos, por ejemplo, din´
amicas socioecol´
ogicas locales y los valores que forman la coexistencia entre
humanos y vida silvestre.
Palabras Clave: ´
areas protegidas, crecimiento poblacional, percepci´
on, sistemas socioecol´
ogicos, sustentos
rurales, trampas socioecol´
ogicas, valores sociales y culturales
Introduction
Whereas the consequences of immigration in terms of
conservation are well defined, the causes of changes in
human settlement around protected areas (PAs) are not
well understood (Oglethorpe et al. 2007). Wittemyer et
al.’s (2008) suggestion that accelerated human popula-
tion growth occurs on the edge of PAs raises the issues
of the effects of conservation-related funding on PA ef-
fectiveness and on the values of PAs for local people.
Using the same data sets as Wittemyer et al. (2008), Joppa
et al. (2009) found no evidence of increasing human
populations near PAs. This finding brings into question
the use of large-scale meta-analyses to address population
change on the edge of PAs. In particular, Hoffman et al.
(2011) underline that by amalgamating all human com-
munities as “populations,” large-scale analyses of popula-
tion growth at the edge of PAs often neglect local land
tenure systems and diversity of values that may facili-
tate or impede migration. In a PLoS ONE “Reader Com-
ment” (http://www.plosone.org/annotation/listThread.
action?root=13%2C089), J. Igoe argues that without a
better understanding of the dynamics of human com-
munities living on the boundaries of PAs, interventions
targeting these communities are likely to misfire in terms
of addressing social equity and in terms of protecting
biodiversity. Overall, these researchers call for critical
analyses of the interactions between people and PAs for
use in addressing factors relevant to conservation plan-
ning. We investigated the drivers of population changes
and how attitudes toward PAs vary among different kinds
of settlers near Hwange National Park and Sikumi Forest
Area, Zimbabwe. We also documented the concerns of
people living near these PAs for their future livelihoods.
Often, human communities surrounding PAs have
been excluded or prohibited from using natural re-
sources, and their needs are regarded as incompatible
with conservation (Infield & Namara 2001). Conventional
exclusionary approaches to conservation can have high
social costs, especially when affected indigenous people
and local communities were marginalized even before
establishment of the PAs (Borrini-Feyerabend et al. 2004).
These “fortress conservation” models have been criti-
cized because the associated opportunity costs are borne
by the rural poor and the benefits accrue to governments,
national elites, and wealthy foreign tourists, especially in
Africa (Brockington 2002).
Integrated conservation and development projects
(ICDP) have been conducted since the early 1990s with
the aim of reducing opportunity costs of conservation
or reestablishing property rights for indigenous people
(Naughton-Treves et al. 2005). Some of these initiatives
have been successful, but in general they do not appear
to improve standards of living or to match indigenous
visions of development (Alexander & McGregor 2000;
Brockington 2002; Naughton-Treves et al. 2005). The
ICDPs can also affect peoples’ decisions to immigrate
to places near PAs, and immigration to such areas may
threaten initial conservation objectives (Scholte 2003).
Because natural resources can contribute substantially to
rural incomes (Cavendish 2000), PAs may have a great ef-
fect on people living nearby (West & Brockington 2006).
However, drivers of human movement, often referred to
as attraction models, cannot be explained by perceived
economic opportunities alone (Hoffman et al. 2011).
Scholte and De Groot (2010) built on the push-and-pull
framework of Oglethorpe et al. (2007) and proposed 2
models and an incidental category to describe immigra-
tion to PA edges. Their frontier-engulfment model cor-
responds to a situation where population increases as
a result of resource extraction at the edge of the PA
and subsequent development of agriculture. In the at-
traction model, people settle near the PA for the services
it provides (employment, ICDPs, resources). Incidental
Conservation Biology
Volume 27, No. 4, 2013

846 Immigration and Integrated Conservation
mechanisms of settlement near PAs involve movement
of people toward PAs during times of conflict or nat-
ural disaster. This category also includes movement of
people evicted from PAs or lands set aside for commer-
cial development and people in need of resettlement.
Because some PAs have the dual purpose of protect-
ing biodiversity and contributing to poverty alleviation
(Naughton-Treves et al. 2005), it is crucial to under-
stand links between people and PAs, drivers of attrac-
tiveness of PAs to people, and related peoples’ atti-
tudes toward PAs (Salafsky & Wollenberg 2000; J. Igoe
PLoS ONE “Reader Comment” [http://www.plosone.org/
annotation/listThread.action?root=13%2C089]).
Our study area in Zimbabwe allowed us to explore the
relations between population changes around PAs and
peoples’ attitudes toward conservation, in a context of
diverse livelihood strategies and linkages to PAs, and to
analyze the conditions for integrated management. The
9 villages we studied are located between 2 state-owned
unfenced PAs, Hwange National Park (HNP) and Sikumi
Forest Area (SFA). These areas have different histories
and different restrictions on access to natural resources
and hence different strategies to integrate rural commu-
nities in conservation efforts. Anyone who attempts to
extract natural resources from HNP can be prosecuted;
however, the boundaries of HNP are not fully controlled.
On specific occasions extraction of thatching grass has
been allowed under close supervision by HNP staff. In
contrast to HNP, access to natural resources in SFA is
more formally permitted and organized. Following severe
drought in the early 1990s, local communities were au-
thorized to graze cattle up to 3 km within the SFA bound-
aries. Firewood collection (dead wood only) is allowed
but regulated, and wood harvesting can be authorized
by the Forestry offices. In the early 2000s, coinciding
with the Zimbabwean economic crisis, the veterinary
fence separating SFA and the villages (households and
associated fields) was dismantled. This led to an increase
in encroachment of rural communities into the PAs and
created major sources of conflict among stakeholders.
We investigated population changes in the 9 study
villages and expected that, under the attraction model,
immigrants would preferably settle close to the PAs on
a gradient of distances. We tested correlations between
level of immigration per village and proxies of attractants
to PAs, including level of employment, access to facilities,
and access to natural resources. We assessed whether
people benefited living near the PAs and how their de-
rived benefits were associated with a positive perception
of the PA and conservation in general. We expected that
people attracted to the area and that benefited from the
PA would have a more positive perception of the PA
than those who did not benefit. We explored plausible
explanations for the negative perceptions of immigrants
toward PAs by testing candidate variables related to 3
hypothetical sets of drivers: household typology, feelings
and expectations toward conservation and future liveli-
hoods, and location of settlement relative to PAs.
Methods
Study Area
Located in a communal area (i.e., an area dedicated to hu-
man settlement with lands allocated by traditional author-
ities) within the Hwange District (Matabeleland North,
Zimbabwe), our study area (200 km2) included 9 villages
bordered to the southwest by the Main Camp area of
Hwange National Park (HNP) (14 651 km2), to the east
by Sikumi Forest Area (SFA) (1100 km2), and to the south
by the town of Dete (26◦87’E, 18◦62’S) (Fig. 1). First
designated as a game reserve in 1928, HNP is now under
the management of the Zimbabwe Parks and Wildlife
Management Authority, whereas SFA, designated in 1968,
is under the management of the Forestry Commission of
Zimbabwe. Human settlements are on the edge of SFA,
whereas a small portion of land (i.e., a buffer area) used by
people and wildlife exists between HNP and the commu-
nal area (Fig. 1). Hwange National Park and SFA are part of
the Kavango–Zambezi Trans Frontier Conservation Area
(TFCA), and HNP hosts one of the highest densities of
free-ranging elephants in Africa, particularly in the Main
Camp area (Chamaill´
e-Jammes et al. 2009).
The study area, classified as agroecological region IV, is
characterized by low-fertility soils (mostly Kalahari sands)
and erratic annual rainfall (606 mm, interannual CV =
25%). The villagers rely primarily on subsistence farm-
ing and natural resource harvesting. Maize (Zea mays),
sorghum (Sorghum bicolor),andpearlmillet(Pennise-
tum glaucum) are the main crops. Economically, the area
has historically benefited from tourism (safaris), which
declined during the Zimbabwean economic crisis in the
2000s (Bond & Cumming 2006).
In the communal area, access to natural resources and
land use are ruled internally and enforced primarily by
traditional leaders. There was no ICDP in place that could
have affected people’s use of natural resources (Scholte
2003; Wittemyer et al. 2008), except CAMPFIRE (Com-
munal Areas Management Programme For Indigenous Re-
sources), but this program has not yet been particularly
successful in this area (C.G., unpublished data; see also
Alexander & McGregor 2000 for a study of the distrust
of CAMPFIRE in neighboring districts). The few local
non-governmental organizations that operate in this area
concentrate on food security, medical assistance, and
education.
Data Collection and Processing
We collected village histories through group discus-
sions with key informants (village heads, headmen, and
Conservation Biology
Volume 27, No. 4, 2013

Guerbois et al. 847
°
°
°
°
°
°
°°
°°
°°
°
°
°°
°°
°
°
°
°
°°°
°
°
°°
°°
°°
°
°
°°
°
°
°
°
°
°
°°
°
°
°
°°
°
°°
°
°°
°
°
°°
°
°
°°
°°
°
°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°
°°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°
°
°
°
°
°°°°°
°
°
°
°
°
°°
°°
°
°
°
°
°
°
°
°
°
°
°
°°
°
°°
°
°°
°
°
°
°
°°°
°°
°
°
°
°
°
°
°°
°
°
°°°°
°°
°
°
°
°
°°
°
°
°
°
°
°°
°
°
°
°
°°
°
°
°
°
°°
°
°
°°
°
°°°°
°
°
°°°
°
°
°
°
°
°
°
°°
°°°
VF−BYO Road
HWANGE NP
SIKUMI FOREST
0 2500 5000 m
N
BUFFER AREA
Chezhou
Dingani Jwape
Magoli
Makwandara
Malonga
Mambanje
Nyagara
Sialwindi
Dete
*
Figure 1. Map of the study area in Hwange District, Zimbabwe (circles, sampled households [n =217]). Hwange
National Park and Sikumi Forest Area are the 2 protected areas.
elders). The number of households per village in 1990,
2000, and 2010 was extracted from records provided by
the local traditional leaders (village heads and headmen
[Supporting Information]). From 1990 to 2000, the Zim-
babwean economy was growing, whereas 2000–2010
was characterized by high inflation and economic re-
cession. We conducted semi-directive interviews from
May to July 2011 following a stratified random sampling
of 217 households in 9 villages distributed at different
distances from HNP and SFA (Fig. 1). The interviews
focused on livelihoods made at the periphery of PAs,
peoples’ history, education, perceptions of conservation
and PAs, and concerns for the future; production sys-
tems used; and natural resources used. Interviews were
conducted in local languages (Nambiya, Tonga, or Nde-
bele) by trained local assistants. The position of each
household was recorded with a geographic positioning
system.
To avoid a priori biases in data collection, responses to
open questions were recorded exhaustively in writing.
For the statistical analyses, we classified the responses a
posteriori to minimize the number of modalities in each
variable but glean as much information as possible. Thus,
our results may be partly affected by our interpretation
of the responses because qualitative analyses are always
liable to subjectivity. However, having lived within the
community for several years, we believe our qualitative
assessment is meaningful (Drury et al. 2011; for the de-
bate on qualitative data in conservation research) and
reflects the complexity of drivers of the diverse local
attitudes toward PAs.
Ethnolinguistic categories were Nambiya, Ndebele,
Dombe, Shona, Tonga, and other. Where respondents’
parents lived was recorded and used to build a binary
variable such that respondents whose parents lived in
the same village or in HNP and SFA were considered
Conservation Biology
Volume 27, No. 4, 2013

848 Immigration and Integrated Conservation
locals and those whose parents lived farther away were
considered immigrants. We separated settlers into 3 cat-
egories on the basis of their responses to the question,
why do you live here? We considered that people who
mentioned the area was a good place to live because
natural resources are more available than elsewhere lived
in the area because of its attractiveness. People who were
evicted from the PAs or were removed from other places
and resettled by local authorities lived in the area due
to resettlement. Some people lived in the area because
family land was available or to be close to relatives. Eight
respondents mentioned other reasons for living in the
area, and we removed them from our analyses because
they did not constitute a substantial or homogeneous
group. Education level categories were no education,
primary education, and secondary education.
Over 25 different classes of natural resources were
mentioned by respondents. We used access to wood
products (i.e., timber and firewood) as a proxy of natural
resources access because wood products were harvested
where people collected most natural resources. Places of
collection were communal areas, SFA, HNP, and buffer
area.
We categorized people’s perceptions of PAs as good
or bad and the reasons for this perception as services,
tourism, conservation, no benefit, problem animals, and
restrict access. Because 95% of the respondents answered
positively when asked if protecting wildlife was impor-
tant, we focused on why they thought so and categorized
their answers as tourism, heritage, conservation, and pro-
tect people. We asked villagers their main concern for the
future and summarized the answers into 5 categories: wa-
ter, natural resources, overpopulation, problem animals,
and development. We included a no-idea category when
questions were not answered.
Data Analyses
We described the relative proportion of immigrants ver-
sus locals according to the reasons for settling at the
periphery of HNP (attractiveness, resettlement, and fam-
ily land) and used chi-square tests to examine differences
in contingency tables. We used Spearman tests to assess
rank correlations between village characteristics and ap-
parent attractiveness of the area per village. We expected
attractiveness to be negatively correlated with distance
to PAs (used as a proxy of access to natural resources)
and to distance to roads (proxies of access to facilities)
but positively correlated with the level of employment
per village.
A multiple correspondence analysis (MCA) was per-
formed on household-wealth characteristics (ownership
of goats, cattle, and transport vehicle; ploughing practice;
employment; and access to meat). We used the first 2
principal axes, which explained 47% of total variance, to
build livelihood indices: PC1 (axis 1) was level of assets
owned by the households (assets) and PC2 (axis 2) was
level of poverty of the household (poverty). Details of
this analysis are in Supporting Information.
We computed distances of household to SFA and HNP
(distance from the center of the yard to the edge of the
PA) and distances of villages to the closest dirt, tarred, and
main roads (A8 Bulawayo-Victoria Falls) (distance from
the center of the village to the road) with geographic
information system (GIS) software (Quantum GIS 1.6,
2010 Quantum GIS development team). For each village,
we defined the type of edge shared with PAs as hard
(no buffer area between settlement and PA), soft (buffer
area between settlement and PA), and no boundary (no
boundary shared with a PA) (Supporting Information).
We performed an MCA to investigate the link between
people’s access to natural resources and their attitude
toward PAs to illustrate how different settler types were
represented. We expected access to natural resources
within the PAs would correlate with a better attitude
toward PAs, in particular when access was officially au-
thorized (i.e., SFA), and hence with a better integration
of community functioning in the management of PAs.
Because households were nested within villages, we used
a mixed linear model (village as random effect) to inves-
tigate determinants of negative attitudes toward PAs. In
this analysis, we tested 3 hypothetical sets of attitude
drivers that related to household typology (e.g., educa-
tion, history, assets), feelings toward conservation and
concerns for the future, and location of household. To
avoid overdispersion in the MCA, we did not include no-
idea responses in any model. Less than 15 households
provided this response.
We performed statistical analyses with R software
(version 2.13) (R Development Core Team 2011), in par-
ticular the package ade4 for multivariate analyses (Dray &
Dufour 2007).
Results
Village History, Population Growth, and Drivers
of Immigration
Most of the villages in the study area were created after
Nambiya were forcibly resettled by colonial authorities,
first to allow for European settlers to farm the land and
second to create the PAs (DNPWLM 1999) (Supporting
Information). Except for Jwape village, which had a 35%
decrease in population from 2000 to 2010, village popu-
lations increased 16% (SD 9) from 1990 to 2000 and 61%
(SD 42) from 2000 to 2010. Household density ranged
from 2.05 to 16.1 households/km2in 2010 (Supporting
Information).
Of the 209 households included in our analyses, 72%
were locals (n=151). Seventy-six percent of locals were
in the family-land category, 13% were in the resettlement
Conservation Biology
Volume 27, No. 4, 2013

Guerbois et al. 849
°
°
°
°
°
°
°°
°°
°°
°
°°°°
°
°
°
°
°
°°°
°
°
°°
°°
°°
°
°
°°
°
°
°
°
°
°
°°
°
°
°
°°
°
°°
°
°°
°
°
°°
°
°
°°
°°
°
°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°
°
°
°
°
°
°
°°
°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°°
°
°
°
°
°
°
°°
°°
°
°
°
°
°
°
°
°
°
°
°
°°
°
°°
°
°°
°
°
°
°
°°
°
°°
°
°
°
°
°°
°°
°°
°°°°
°°
°
°
°
°
°°
°
°
°
°
°
°°
°
°
°
°
°°
°
°
°
°
°°
°
°
°°
°
°°°°
°
°
°
°
°
°
°
°
°
°
°
°
°°
°°°
0 2750 5500
m
N
VF−BYO Road
a
−0.1 0.1 0.3
°
°
°
°
°
°
°°
°°
°°
°
°°°°
°
°
°
°
°
°°°
°
°
°°
°°
°°
°
°
°°
°
°
°
°
°
°
°°
°
°
°
°°
°
°°
°
°°
°
°
°°
°
°
°°
°°
°
°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°
°
°
°
°
°
°
°°
°
°
°
°
°
°°
°
°
°
°
°
°
°
°
°°
°
°
°
°
°
°
°°
°°
°
°
°
°
°
°
°
°
°
°
°
°°
°
°°
°
°°
°
°
°
°
°°
°
°°
°
°
°
°
°°
°°
°°
°°°°
°°
°
°
°
°
°°
°
°
°
°
°
°°
°
°
°
°
°°
°
°
°
°
°°
°
°
°°
°
°°°°
°
°
°
°
°
°
°
°
°
°
°
°
°°
°°°
b
−0.5 −0.3 −0.1 0.1 0.3
Figure 2. Relative proportion of (a) sampled households (n =209) per village citing attractiveness of the area
and (b) immigrants citing attractiveness as a driver of settlement decision (black squares, villages with higher
relative proportions of households mentioning attractiveness; white squares, villages with lower relative
proportions of households mentioning attractiveness; circles, sampled households).
category, and 11% were in the attractiveness category for
the reason to settle in the area. Immigrants (n=58) said
they moved to the area due to family land (32%), resettle-
ment (21%), and attractiveness (47%). The proportion of
attractiveness respondents was higher in villages close to
SFA (χ2=14.96, df =2, p=0.0006) and higher in vil-
lages close to either of the PAs for attractiveness migrants
(χ2=12.28, df =2, p=0.0021) (Fig. 2). Magoli and
Chezhou had the highest population growth from 2000
to 2010 (133% and 103%, respectively), and household
density was highest in villages closest to PAs (up to 16.11
household/km2for Dingani) (Supporting Information).
The proportion of immigrants in the villages was neg-
atively correlated with distance to PAs (rs=−0.71, p<
0.05) and particularly with distance to SFA (rs=−0.83, p
<0.05). Correlations between the distance to SFA and the
proportion of respondents mentioning attractiveness per
village (rs=−0.72, p<0.05) and between the distance
to PAs and the proportion of attractiveness migrants per
village (rs=−0.71, p<0.05) were significant and neg-
ative. Neither employment level per village nor distance
to roads (access to facilities) were significantly correlated
with proportion of immigrants, attractiveness, or attrac-
tivenes migrants per village. Level of employment was
significantly correlated with the proximity of households
to the main Victoria Falls-Bulawayo road (rs=–0.79, p<
0.05).
Attitudes Toward PAs
Respondents had on average a good attitude of the PAs,
and there were significant differences in relative percep-
tion among the 3 categories of settlers for both PAs (HNP:
χ2=13.26, df =2, p=0.0013; SFA: χ2=7.801, df
=2, p=0.0203). Twenty-seven percent of those who
were attracted to the area, 15% who resettled, and 4%
living on family land had a negative attitude toward HNP,
whereas 32%, 12%, and 11%, respectively, had a negative
attitude toward SFA. Immigrants had a higher proportion
of negative attitudes than locals (respectively 17% vs. 8%
for HNP and 22% vs. 12% for SFA).
The PC1 of the MCA explained 29% of the total
variance and distinguished people with a negative atti-
tude toward PAs who benefited from SFA from others
(Fig. 3a). Individuals attracted to the area were more
likely to express a bad attitude of the PAs (Fig. 3b). This
result was mainly due to extreme values for attractive-
ness migrants (Fig. 3c). For most of the respondents, the
perceived benefits of HNP were related to the conserva-
tion of natural resources (Table 1). The distribution of
Conservation Biology
Volume 27, No. 4, 2013

850 Immigration and Integrated Conservation
Figure 3. Results of multiple-correspondence analysis of attitudes toward protected areas (HNP, Hwange National
Park; SFA, Sikumi Forest Area) and places of natural resource collection (CA, communal area; buffer, buffer area)
(n =185): representation of (a) attitude to protected areas (good or bad) and places of collection of wood and
firewood on the first 2 axes explaining 51% of the total variance, (b) number of individuals on the first 2 axes
with reason of settlement as an illustrative variable (circles), and (c) individuals on the first 2 axes with
attractiveness immigrants as a driver of settlement decision as an illustrative variable (gray squares, individuals;
size of square indicates relative number of individuals with similar coordinates).
answers among settler categories (χ2=3684, df =12,
p=0.0002) and their immigration category (χ2=15.91,
df =6, p=0.014) differed significantly, particularly
due to negative perceptions associated with problem
animals that were more frequent than expected for immi-
grants and attractiveness settlers (Table 1). Attitudes to-
ward SFA did not differ significantly among respondents,
and positive attitudes were related to conservation and
services (such as providing access to natural resources).
Feelings about wildlife protection differed significantly
among settlers of different types (χ2=15.91, df =6,
p=0.014) (Table 1). Resettlement respondents primarily
mentioned the importance of protection for tourism pur-
poses, whereas other types of respondents mainly said
wildlife was part of their heritage (Table 1). Respondents
attracted to the area thought protecting wildlife also
Conservation Biology
Volume 27, No. 4, 2013

Guerbois et al. 851
Table 1. Attitudes of villagers to protected areas and wildlife protection and concerns for the future as a proportion in each category of settlers and
immigration status (n=209).
Reason for settlement
Attitude element Modalities Attractiveness Resettlement Family land Locals Immigrants
Hwange National Park services 0.09 0.06 0.03 0.03 0.09
tourism 0.14 0.03 0.14 0.12 0.14
conservation 0.39∗0.68∗0.66∗0.64 0.52
no idea 0.11 0.06 0.13 0.14∗0.05∗
no benefit 0.07 0.00 0.01 0.01 0.05
problem animals 0.20∗0.13 0.02∗0.05∗0.16∗
restrict access 0.00 0.03 0.00 0.01 0.00
Sikumi Forest Area services 0.20 0.42 0.28 0.27 0.31
conservation 0.36 0.42 0.46 0.46 0.38
no idea 0.18 0.10 0.18 0.18 0.14
no benefit 0.02 0.00 0.02 0.02 0.02
problem animals 0.11 0.00 0.02 0.03 0.07
restrict access 0.11 0.06 0.04 0.05 0.09
Wildlife protection conservation 0.11 0.13 0.16 0.15 0.16
tourism 0.11∗0.45∗0.28 0.30 0.21
heritage 0.43 0.23∗0.46∗0.42 0.41
no idea 0.00 0.00 0.02 0.01 0.02
protect people 0.34∗0.19 0.07∗0.13 0.21
Concerns for the future water 0.16∗0.35 0.48∗0.45∗0.24∗
development 0.11 0.16 0.13 0.15 0.10
no idea 0.07 0.06 0.04 0.04 0.07
natural resources 0.41∗0.29 0.25∗0.26∗0.36∗
overpopulation 0.16∗0.13 0.06∗0.07∗0.16∗
problem animals 0.09∗0.00∗0.04 0.04 0.07
∗Actual value differs significantly from expected value (chi-square test on contingency table, p <0.05).
protected people and their livestock from wild animals
(i.e., manage wildlife to reduce the risks). Locals (χ2=
20.33, df =10, p=0.026) and immigrants (χ2=11.94,
df =5, p=0.036) differed significantly in their concerns
for the future. Locals were more concerned with climate
uncertainties and water availability, whereas immigrants
were more concerned with natural resources and over-
population (Table 1).
Factors Affecting Attitudes Toward PAs
On the basis of household typology, attractiveness set-
tlers had significantly more negative attitudes toward
PAs while accessing resources in SFA than respondents
in other categories (Table 2). Conversely, the poverty
index (households with no employment, low protein ac-
cess) was inversely correlated with this attitude toward
PAs. Other variables such as ethnic group, assets, and
education level were not significant.
For respondent feelings about PAs, problem animals
were correlated to negative attitudes to both PAs and
restricted access to natural resources. Restricted access
significantly increased negative attitudes toward SFA
(Table 2). Similarly, seeing wildlife protection as a way
to protect people from wild animals (i.e., spatial control)
was also significantly correlated with a negative attitude
toward PAs. A negative attitude was significantly affected
by respondents whose concerns for the future were cen-
tered on problem animals, natural resources, and over-
population (Table 2).
Household location was also associated with negative
attitudes of people who accessed resources in SFA. The
distance to PAs was significantly and negatively corre-
lated, which suggests respondents living closer to the PAs
were more likely than respondents living farther away
to have negative attitudes toward PAs despite access to
natural resources. This correlation between distance and
attitude seemed to be stronger for households in villages
that shared a hard edge with the PAs relative to house-
holds in villages with a soft edge (Table 2).
Discussion
Population Growth on the Edge of HNP
Annual population growth in Zimbabwe was on average
1.78% and 0.06% for the 1990–2000 and the 2000–2010
decades, respectively (United Nations 2011), our results
thus suggest that the huge increase in the population
observed on the edge of HNP resulted from immigration.
Part of this immigration was related to the national politi-
cal and economic crisis that occurred in the early 2000s.
Conservation Biology
Volume 27, No. 4, 2013

852 Immigration and Integrated Conservation
Table 2. Results from mixed linear models for explaining the negative attitudes toward protected areas of people who access natural resources
within Sikumi Forest Area, with all candidate drivers of this attitude tested, including village identity as a random effect.a
Drivers of negative attitude Variables Fbp Modalities EstimatescSE p
Household typology immigrant F1,175 =0.17 0.679
family lande0.006 0.183 0.974
reason for settlement F2,174 =4.28 0.015 attractiveness −0.302 0.103 0.004
resettlement −0.066 0.115 0.568
attractiveness migrantsdF1,175 =2.37 0.126
ethnic group F5,171 =1.44 0.213
assets F1,175 =0.01 0.926
poverty F1,175 =3.69 0.057 0.183 0.096 0.057
education level F2,174 =0.21 0.808
Attitude element Hwange National Park F5,160 =9.914 <0.0001 servicese0.041 0.216 0.85
tourism −0.057 0.186 0.76
conservation 0.064 0.171 0.711
no benefit −0.283 0.261 0.28
problem animals −0.899 0.202 <0.0001
restrict access −1.221 0.482 0.012
Sikumi Forest Area F4,157 =12.671 <0.0001 servicese0.073 0.175 0.678
conservation 0.019 0.085 0.819
no benefit −0.055 0.241 0.821
problem animals −0.821 0.172 <0.0001
restrict access −0.76 0.153 <0.0001
wildlife protection F3,170 =3.755 0.012 conservatione−0.064 0.203 0.752
tourism 0.117 0.119 0.327
heritage 0.01 0.116 0.93
protect people −0.301 0.14 0.033
concerns for the future F4,165 =2.039 0.091 watere0.092 0.172 0.592
development −0.012 0.138 0.933
natural resources −0.274 0.139 0.051
overpopulation −0.321 0.183 0.081
problem animals −0.465 0.206 0.025
Location edge PA F2,6 =10.735 0.01 softe0.342 0.206 0.099
hard −0.926 0.255 0.011
no 0.024 0.274 0.935
distance to PA F1,175 =22.776 <0.0001 0.0001 0.00003 <0.0001
an=209, households responding “no idea” were not included in the models.
bThe degrees of freedom of the residuals vary due to different number of households responding "no idea" and to the covariation between the
variables and the random effect (Village).
cNegative estimates correspond to a more negative attitude toward protected areas, that is, more negative values on the PC1 in Figure 3.
dMigrants who settled in the area for its attractiveness (cf. "Methods" section).
eModalities used as references in the intercept.
Displacement, as a political and survival strategy, consti-
tuted a major dimension of the Zimbabwe crisis given its
unprecedented scale and the depths of physical, eco-
nomic, institutional, social, and personal dislocation
(Hammar et al. 2010). This crisis is described by Scholte
and De Groot’s (2010) incidental-mechanisms model in
which people are pushed from towns by increasing un-
employment and a worsening urban economy. Our re-
sults thus demonstrate that immigration to the edge of
the PAs did not result only from the attractiveness of the
area but also included resettlement and people moving
back to family land.
However, access to natural resources appeared to be
the main driver of the attractiveness of the PA. We
think that HNP and SFA provided refuges that allowed
inventive strategies for survival, adaptation to circum-
stances, and prosperity in a context of what Jones (2010)
called a Kukiya-kiya economy (i.e, a system that is based
on opportunistic, marginally legal practices that exploit
distortions of a hyperinflationary, quickly contracting
economy). Moving to the PA margins thus appeared
to be a coping strategy (i.e., a short-term adjustment
and adaptation to extreme events for maintaining the
future income-generating capacity of the household in-
tact) (Adger 2000) for people in a subsistence econ-
omy that relied on agriculture and the use of natural
resources.
Paradoxical Perception of PAs
Allendorf et al. (2006) suggest that positive attitudes
toward PAs are highly correlated with a perception of
benefits resulting from conservation and the manage-
ment of these areas. This is consistent with our results;
Conservation Biology
Volume 27, No. 4, 2013

Guerbois et al. 853
attitudes toward PAs were generally good and associated
with conservation benefits and tourism. However, we
were surprised by the negative attitudes of people ac-
cessing resources from SFA, where greater access to nat-
ural resources is allowed. This negative attitude may be a
protest response to the rules of access, but because the
access to NR is greater than is officially authorized, we be-
lieve this paradoxical negative attitude arises from other
causes.
Respondents who settled in the area for attractiveness
reasons, in particular the wealthiest respondents, had this
paradoxically negative attitude toward PAs. This suggests
that if the conservation framework draws profit-seeking
individuals close to PAs (Igoe & Brockington 2007), it
may also create conditions for greater frustration if the ex-
pectations associated with the motivation to migrate are
not met and hence promote more conflicts between PA
managers and surrounding communities. Furthermore,
immigration may lead to a dilution of traditional values
that enforce sustainable use of shared resources (Berkes
et al. 2000) and ultimately lead to more conflicts about
natural resource use and conservation in general.
Our results are consistent with the thought that prob-
lem animals are the primary factor in negative attitudes
toward PAs (de Boer & Baquete 1998). In our study area,
elephant damage occurs up to 4400 m from the edges
of the PAs (Guerbois et al. 2012). Hence, people settling
close to PAs to access natural resources expose them-
selves to more problems with wildlife, which may lead
to frustration relative to the net benefits derived from
the PAs (Brockington 2002). Furthermore, respondents
in our study with a paradoxically negative attitude toward
PAs believed wildlife protection is necessary to protect
them from wildlife through increased wildlife confine-
ment and control. These correlations imply that problem
animals may act as a deterrent to settling at the margins
of PAs.
More importantly, people expressing paradoxically
negative attitudes were concerned with the availability
of natural resources and overpopulation. These concerns
likely reflect a perceived lack of sustainability of their
way of life as a result of increasing competition with
wildlife and other people for space and natural resources.
This conclusion is supported by the fact that restricted
access significantly affected respondents’ perceptions of
both PAs (when access was totally banned or when it
was regulated), which is consistent with the debate sur-
rounding costs of conventional exclusionary approaches
(Brockington 2002; Borrini-Feyerabend et al. 2004). From
a landscape perspective, the more paradoxically negative
attitude at the hard edge probably arises from the fact that
hard edges are more abrupt physical transitions between
wildlife areas and villages and hence promote close con-
tacts and greater interactions between people and wild
animals.
Integrated Management of PAs
Integrated protected area management implicitly bor-
rows from management of socio-ecological systems, char-
acterized by multiple social dimensions (conservation
and development objectives), and driven by complex
interrelated ecological and social dynamics. Outsiders’
prominent claims and interests in forests and wildlife
areas often lead to locals’ claims being reformulated,
sometimes modified, in regards to these new stakeholder
desires (Baldus 2009). Population changes, profit-seeking
migrants in particular, may ultimately dilute local and
traditional value of wildlife conservation and natural re-
source use and thus increase social pressures on man-
agement of PAs. We believe the future trajectory of the
Hwange socio-ecological system depends on its capac-
ity to cope with changes such as population growth,
aridification, dilution of local values, social crisis, and
drought. In our study area, we believe that the Zim-
babwe economic crisis and its associated migrations led
to increased social tensions in relation to conflict with
wildlife, restricted access to natural resources, dilution
of local values, and overpopulation. These tensions af-
fected the perception of PAs, creating the paradoxi-
cally negative attitude of some toward PAs. This strongly
suggests that integrated conservation initiatives should
consider the balance between conservation and services
and disservices provided by the PAs to the neighboring
communities.
Moreover, apparent attractiveness of PAs could lead to
poverty or rigidity traps, defined respectively as states
preventing adaptation through loose connections and
poor resources or lack of flexibility induced by strong
self-reinforcing controls (Carpenter & Brock 2008). In
our study, we refer to these as socio-ecological traps
because the benefits experienced by local communities
through access to PAs may not be sustainable as a result
of enforced de jure protection of PA edges (Joppa et al.
2008); weak control of problem animals which increases
the costs of living at the edge of PAs; increasing popula-
tion at PA edges; and a loose definition of shared values
that could allow coexistence amongst stakeholders (PA
managers, local authorities, and villagers). Mismatches
between the scales of ecological processes and the insti-
tutions that are responsible for managing them can con-
tribute to a decrease in socioecological resilience (Cum-
ming et al. 2006). This decrease in resilience can also
arise from the mismatch between management policies
and local expectations and from mismatches between
traditional values of coexistence and business-driven con-
servation frameworks promoting profit-seeking attitudes
toward conservation areas (Igoe & Brockington 2007).
Furthermore, protecting the values that shape coexis-
tence between people and wildlife, such as a subsis-
tence economy that is based on endogenous ecological
Conservation Biology
Volume 27, No. 4, 2013

854 Immigration and Integrated Conservation
processes, should help strengthen resilience in socio-
ecological systems including PAs.
Acknowledgments
This research was conducted under the Hwange Environ-
mental Research Development program from the Centre
National de la Recheche Scientifique (CNRS), within the
CNRS-Hwange Long Term Ecological Research Site and
the research platform Production and Conservation in
Partnership. We thank the Parks and Wildlife Manage-
ment Authority, Research Council of Zimbabwe, Hwange
Rural District Council, and the Chief D. Nelukoba for
providing authorizations to carry out this research. C.G.
was supported by an ERAMET PhD grant. We are grateful
to D. Cumming for his comments and P. Duncan for
his in-depth review of the manuscript. We also thank
E. Chapanda, V. Bitu, and the 4 local enumerators for
their dedicated involvement in this project, and the local
community for their open participation.
Supporting Information
History and trends in population size of the 9 villages in
which we conducted surveys (Appendix S1) and the MCA
for building assets and poverty indices (Appendix S2) are
available online. The authors are solely responsible for
the content and functionality of these materials. Queries
should be directed to the corresponding author.
Literature Cited
Adger, W. N. 2000. Social and ecological resilience: Are they related?
Progress in Human Geography 24:347–364.
Alexander, J., and J. McGregor. 2000. Wildlife and politics: CAMPFIRE
in Zimbabwe. Development and Change 31:605–627.
Allendorf, T., K. K. Swe, T. Oo, Y. Htut, M. Aung, K. Allendorf, LA.
Hayek, P. Leimgruber, and C. Wemmer. 2006. Community attitudes
toward three protected areas in Upper Myanmar (Burma). Environ-
mental Conservation 33:344–352.
Baldus, R. D. 2009. A practical summary of experiences after three
decades of community-based wildlife conservation in Africa “What
are the Lessons Learnt”? Food and Agriculture Organization and In-
ternational Council for Game and Wildlife Conservation, Budapest.
Berkes, F., J. Colding, and C. Folke. 2000. Rediscovery of traditional eco-
logical knowledge as adaptive management. Ecological Applications
10:1251–1262.
Bond, I., and D. H. M. Cumming. 2006. Wildlife research and de-
velopment. Pages 465–496 in M. Rukuni, P. Tawonezi, C. Eicher,
M. Munyuki-Hungwe, and P. Matondi, editors. In Zimbabwe’s agri-
cultural revolution revisited. University of Zimbabwe Publications,
Harare, Zimbabwe.
Borrini-Feyerabend, G., A. Kothari, and G. Oviedo. 2004. Indigenous
and local communities and protected areas: towards equity and
enhanced conservation. International Union for the Conservation
of Nature, Gland, Switzerland.
Brockington, D. 2002. Fortress conservation. The preservation of the
Mkomazi Game Reserve, Tanzania. James Currey. African Issues se-
ries Oxford, Oxford, United Kingdom.
Carpenter, S. R., and W. A. Brock. 2008. Adaptive capacity and
traps. Ecology and Society 13:http://www.ecologyandsociety.
org/vol13/iss2/art40/ (accessed April 2012).
Cavendish, W. 2000. Empirical regularities in the poverty-environment
relationship of rural households: evidence from Zimbabwe. World
Development 28:1979–2003.
Chamaill´
e-Jammes, S., M. Valeix, M. Bourgarel, F. Murindagomo, and H.
Fritz. 2009. Seasonal density estimates of common large herbivores
in Hwange National Park, Zimbabwe. African Journal of Ecology
47:804–808.
Cumming, G. S., D. H. M. Cumming, and C. L. Redman. 2006. Scale
mismatches in social-ecological systems: causes, consequences, and
solutions. Ecology and Society 11:http://www.ecologyandsociety.
org/vol11/iss1/art14/ (accessed December 2011).
de Boer, F., and D. S. Baquete. 1998. Natural resource use, crop dam-
age and attitudes of rural people in the vicinity of the Maputo
Elephant Reserve, Mozambique. Environmental Conservation 25:
208–218.
Dray, S., and A. B. Dufour. 2007. The ade4 package: implementing the
duality diagram for ecologists. Journal of Statistical Software 22:
1–20.
DNPWLM (Department of National Parks and Wildlife Manage-
ment). 1999. Hwange National Park management plan 1999–2003.
Technical report. United Nations Development Program, Harare,
Zimbabwe.
Drury, R., K. Homewood, and S. Randall. 2011. Less is more: the po-
tential of qualitative approaches in conservation research. Animal
Conservation 14:18–24.
Guerbois, C., E. Chapanda, and H. Fritz. 2012. Combining multi-
scale socio-ecological approaches to understand the suscepti-
bility of subsistence farmers to elephant crop raiding on the
edge of a protected area. Journal of Applied Ecology 49:
1149–1158.
Hammar, A., J. McGregor, and L. Landau. 2010. Displacing Zimbabwe:
crisis and construction in Southern Africa. Journal of Southern
African Studies 36:263–283.
Hoffman, D. M., D. Fay, and L. Joppa. 2011. Human migration to pro-
tected area edges in Africa and America: questioning large-scale
statistical analysis. Conservation and Society 9:1–7.
Igoe, J., and D. Brockington. 2007. Neoliberal conservation: a brief intro-
duction. Conservation and Society 5:432–449. http://www.plosone.
org/annotation/listThread.action?root=13%2C089 (accessed March
2013).
Infield, M., and A. Namara. 2001. Community attitudes and behaviour
towards conservation: an assessment of community conservation
programme around Lake Mburo National Park, Uganda. Oryx 35:48–
60.
Jones, J. L. 2010. “Nothing is straight in Zimbabwe”: the rise of the
Kukiya-kiya economy 2000–2008. Journal of Southern African Stud-
ies 36:285–299.
Joppa, L. N., S. R. Loarie, and S. L. Pimm. 2008. On the protection of
“protected areas.” Proceedings of the National Academy of Sciences
105:6673–6678.
Joppa, L. N., S. R. Loarie, and S. L. Pimm. 2009. On population
growth near protected areas. Public Library of Science ONE 4doi:
10.1371/journal.pone.0004279.
Naughton-Treves, L., M. B. Holland, and K. Brandon. 2005. The role
of protected areas in conserving biodiversity and sustaining local
livelihoods. Annual Review of Environment and Resources 30:219–
252.
Oglethorpe, J., J. Ericson, R. E. Bilsborrow, and J. Edmond. 2007. Peo-
ple on the move: reducing the impacts of human migration on
biodiversity. World Wildlife Fund and Conservation International
Foundation, Washington, D.C.
Conservation Biology
Volume 27, No. 4, 2013

Guerbois et al. 855
Salafsky, N., and E. Wollenberg. 2000. Linking livelihoods and conserva-
tion: a conceptual framework and scale for assessing the integration
of human needs and biodiversity. World Development 28:1421–
1438.
Scholte, P. 2003. Immigration! A potential time bomb under the inte-
gration of conservation and development. Ambio 32:58–64.
Scholte, P., and W. T. De Groot. 2010. From debate to insight: three
models of immigration to protected areas. Conservation Biology
24:630–632.
United Nations. 2011. World population prospects: the 2010 revision.
Department of Economic and Social Affairs, Population Division,
New York.
West, P., and D. Brockington. 2006. An anthropological perspective on
some unexpected consequences of protected areas. Conservation
Biology 20:609–616.
Wittemyer, G., P. Elsen, W. T. Bean, A. C. O. Burton, and J. S. Brashares.
2008. Accelerated human population growth at protected area
edges. Science 321:123–126.
Conservation Biology
Volume 27, No. 4, 2013