Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/taylor-francis/land-use-and-land-cover-dynamics-in-the-melap-forest-reserve-west-pXUrVO52ey
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Taylor & Francis
- Copyright
- © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON).
- ISSN
- 2474-9508
- DOI
- 10.1080/24749508.2021.1923269
- Publisher site
- See Article on Publisher Site
Abstract
GEOLOGY, ECOLOGY, AND LANDSCAPES 2022, VOL. 6, NO. 4, 305–315 INWASCON https://doi.org/10.1080/24749508.2021.1923269 RESEARCH ARTICLE Land use and land cover dynamics in the Melap Forest Reserve, West Cameroon: implications for sustainable management Lucie Félicité Temgoua , Alvine Larissa Meyabeme Elono, Younchahou Mfonkwet Njiaghait, Amadou Ngouh and Clovis Nzuta Kengne Department of Forestry, Faculty of Agronomy and Agricultural Sciences, University of Dschang, Dschang, Cameroon ABSTRACT ARTICLE HISTORY Received 30 July 2020 The present study aims to analyse the dynamics of land use and land cover in the Melap Forest Accepted 24 April 2021 Reserve in West Cameroon on the basis of a diachronic analysis of Landsat 4 TM, Landsat 7 ETM + and Landsat 8 OLI_TIRS images for the years 1988, 2000 and 2018 respectively. Satellite KEYWORDS images were processed using ENVI and ArcGIS software. The results showed that The Melap Spatio-temporal dynamic; Forest Reserve consists of five main classes of land use/land cover; namely, forest, savannah, forest cover lost; land use/ bare soil, cropland and built-up area. The analysis showed an ongoing deforestation and land cover; sustainable degradation of the forest. The forest class has been steadily decreasing from 1345 ha in 1988 management; melap forest reserve to 664 ha in 2018, corresponding to a total loss of 680.9 ha (around 49%) over the 30-year period. The decrease in forest has led to an increase in savannah (+ 315 ha from 1988 to 2018); cropland (+ 351 ha); as well as built-up (+ 9 ha) and bare soil (+ 6 ha). The local communities are the main actors of these changes, principally through agriculture, wood extraction and breed- ing. Then, the present study suggests participatory management, which includes local com- munities, for the restoration and the management of the study area. Introduction and background of the study indirect factors of this change are wood extraction, The world’s forests cover an area of 4.06 billion hec- infrastructure extension, the development of the tares (ha), of which tropical forests account for 45% mining sector and other activities that change the (FAO, 2020). Forest ecosystems provide several ser- physical attributes of the land cover (Gillet et al., vices such as climate regulation, provisioning of plant 2016; Lambin et al., 2003; Momo Solefack et al., and animal resources, recreation and socio-cultural 2018; Temgoua et al., 2018a). In Congo basin, defor- services (Carnol & Verheyen, 2010; Chevassus-au- estation and forest degradation are concentrated louis & Pirad, 2011). Forests and their products play around densely populated areas (Megevand, 2013). a critical role in the improvement of lives of the local Changes in forest cover negatively affect the supplying communities. However, global demand for forest and of ecosystem services, such as climate regulation, bio- food resources increase with population growth, lead- diversity conservation, and carbon storage (Foley ing to overexploitation and conversion of forest land et al., 2005; Giam, 2017; Lucas et al., 2015; Pan et al., to other uses (Temgoua, 2011). According to FAO 2011). This land use change is one of the major causes (2020), 10 million hectares per year of global forest of global climate change (IPCC, 2014). cover have been lost since 2015 and Africa had the The forests of the Congo basin constitute largest annual rate of net forest loss in 2010–2020, at the second largest continuous forest massif after the 3.9 million ha. Anthropogenic activities are the main Amazon. Despite their importance, these forests are causes of forest lost (Harris et al., 2012; Malhi et al., also threatened by deforestation (Megevand, 2013). 2014; Le Quéré et al., 2016). Between 2000 and 2005, annual net deforestation in In tropical regions, extensive conversion of forests the Congo Basin was estimated at 0.17% and annual and agricultural intensification are typically identified net degradation at 0.09% over the same period (Ernst among the most proximate causes of deforestation et al., 2013; De Wasseige et al., 2012). Recent studies (Geist & Lambin, 2002). Agricultural expansion with show that even classified forests are facing many the degradation of natural vegetation cover, is the anthropogenic pressures that lead to their degradation most dominant trajectory of land use and land cover and deforestation (Djiongo et al., 2020; Fokeng et al., change tropical regions. FAO (2016) reports an 2020; Kyale et al., 2019; Temgoua et al., 2018a; Zekeng increase in agricultural land of 6 million hectares et al., 2019). per year in tropical countries. But other direct and CONTACT Lucie Félicité Temgoua temgoualucie@yahoo.fr Department of Forestry, Faculty of Agronomy and Agricultural Sciences, University of Dschang, Dschang 222, Cameroon © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON). This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 306 L. F. TEMGOUA ET AL. The Cameroonian forest, represents the second lar- the same line, evaluating forest cover loss in forest gest tropical forest of Congo basin (after those of the reserves can provide a baseline for sustainable forest Democratic Republic of Congo), covers an area of management and conservation. The MFR has a lack of 22.5 million hectares, which represents about 48% of studies on land cover change and factors of deforesta- the national territory (De Wasseige et al., 2009). In tion and forest degradation. In fact, no monitoring of Cameroon, De Wasseige et al. (2012) estimated the vegetation dynamics has been carried out in the average annual rate of gross deforestation at 0.10% for reserve since his creation. Therefore, this study the period 1990–2000 and 0.17% for the period which aims at assessing the evolution of land use and 2000–2005. This shows that during these two periods, land cover in the reserve through a diachronic analysis the rate of deforestation has increased. Between 1990 of satellite images over three different periods (1988; and 2010, FAO (2011) estimated that Cameroon lost 2000 and 2018). In addition, the activities and the 4400 ha of forests. For the planning and sustainable perceptions of the surrounding populations on the management of the natural forest resources, detailed changes that this reserve has undergone over the knowledge of land use and land cover, as well as the years have been assessed. This constitutes a basic ele- detection of changes, is considered as key parameter ment for the new orientations of strategies and policies for the assessment of the environment and its evolu- for the management of forest reserves in general and tion over time (Verburg et al., 2011; Usman et al., the MFR reserve in particular. 2015). Thus, remote sensing emerges as a fundamental technology to assist in this spatial and Material and methods temporal planning of land use types and vegetation cover evolution (Hansen et al., 2013; Loveland & Study site Dwyer, 2012; Temgoua et al., 2018a; Tsayem, 2010). The Melap Forest Reserve is located in the Foumban In Cameroon, since colonial epoch, in response to Sub-division, Noun Division and West Cameroon forest degradation, several forest reserves and pro- region. It is located between latitudes 5° 44’ 30”-5° tected areas have been created in the interest of 48’ 30”N and longitudes 10° 52’ 0”- 10° 54’ 30”E biodiversity conservation (Temgoua, 2011). With (Figure 1). The climate is tropical sudano-guinean a view to land-use planning, a zoning plan for the type characterized by two seasons; one rainy season southern part of Cameroon has been drawn up from mid-March to mid-November and one dry sea- (Côte, 1993). This zoning plan divides the forest son from mid-November to mid-March. The average area and makes a fundamental distinction between rainfall is 1907 mm/year and the average annual tem- permanent and non-permanent forests. Forest perature is 21.4°C (Njoukam et al., 1996). The soils fall reserves are part of the permanent forest estate that into two main categories: poorly evolved soils (low is designated to remain forested over the long term. proportion of lithosols and alluvial soils), which are Thus, within this permanent forest estate, agricul- mainly found along the Nchi River; and ferralitic soils, ture is prohibited and the use of forest resources is which occupy most of the reserve (Faure, 1986). The subject to restrictions. Today, however, in natural vegetation is dominated by shrub savannah Cameroon, these forest reserves are undergoing sig- characterised by the presence of Terminalia glauces- nificant degradation due to strong pressure on nat- cens, Annona senegalensis and Daniellia oliveri. ural resources by human activities (agriculture, However, the MFR has been reforested with exotic timber extraction, grazing, road and housing con- species, Pinus (about 15 species) and Eucalyptus struction), causing land use change (Fokeng et al., (about 30 species) by the forest services (Njoukam 2020; Fokeng & Meli, 2015; Temgoua et al., 2010; et al., 1996). The population living around the reserve Wafo et al., 2005). is mainly composed of Bamoun tribe. This population Like most reserves, the Melap Forest Reserve mainly practices agriculture, handicraft and petty (MFR) in West Cameroon is encroached by local trade. populations for agriculture and wood extraction. This reserve is also the main source of firewood supply of the surrounding cities and gradually the activities Data collection and analysis carried out there have led to its degradation (Chako, 2016). Due to the failure of the management of forest Land use/land cover change reserves by the government in Cameroun, the state The study was based on the Landsat 4 TM, Landsat 7 decided to entrust decentralized entities such as non ETM+ and Landsat 8 OLI_TIRS satellite images of the governmental Organizations (NGOs). Thus, the MFR, study site for the years 1988, 2000 and 2018, respec- which was created in 1947, was retroceded in 2015 to tively. These images were downloaded via http://glo the National Forest Development Support Agency vis.usgs.gov website of the United States Geological (ANAFOR). However, this transfer of management Survey-Global Visualization Viewer’s. In order to requires the elaboration of a management plan. On reduce cloud cover, images from the dry season period GEOLOGY, ECOLOGY, AND LANDSCAPES 307 Figure 1. Localization of study site in the Western Region of Cameroon. were selected. ENVI 5.2 and ArcGIS 10.5 software and the Kappa index (ratio of well-classified pixels were used to perform image processing. These images to the total number of surveyed pixels). The finali - have been pre-processed to facilitate classification. zation, perfection and cartographic layout of the This consisted in combining the spectral bands to land cover maps resulting from the processing of obtain a multispectral image, resampling of the pixels satellite images were carried out using ArcGIS 10.5 to overlay and combine the images (Mama et al., software. The different steps of satellite images and 2013), enhancing of images to improve its appearance production of maps are shown in Figure 2. and facilitate the visual interpretation and analysis of Change detection consisted in highlighting the the scenes by playing on the dynamics of the radio- change in land cover, quantifying the land cover metric values in the frequency histogram(Donnay, change and developing the transition matrix. The 2000), colour composition, geometric correction and simultaneous image analyses of the three studied extraction of the study area. years (1988, 2000 and 2018), using the spatial analysis The Super Vector Machine (SVM) algorithm was functions of the GIS (Spatial analyst extension of the used for the supervised classification of Landsat ArcGIS software), made it possible to develop the scenes. SVM works by considering each pixel as a two- Land use/Land cover change matrix. dimensional system vector, where each class has The quantification of the annual change was based a pixel as a support vector to establish the class bound- on the calculation of the annual rate of change (R) ary (Boser et al., 1992; Cortes & Vapnik, 1995). In fact, given by Peng et al. (2008) formula below: several authors have demonstrated that the SVM algo- rithm is more accurate than the Maximum Livelihood A2 A1 1 R ¼ X X100 (1) Classification (MLC) and Neural Network (NN) algo- A1 T rithms (Arun et al., 2012; Naguib et al., 2009). where R is the annual rate of land cover dynamic, Subsequently, the classification results were com- measuring the change rate of the target land cover pared with GPS points collected in the field. Thus, type; A1 and A2 are the area of the target land cover in order to evaluate the accuracy of the classifica - type at the beginning and end of the study period, tion, two classification validation indices were respectively and T is the study period, which is usually determined: the overall accuracy (proportion of measured in year. well-classified pixels, calculated as a percentage) 308 L. F. TEMGOUA ET AL. Figure 2. Flowchart showing the different stages of satellite images processing to obtain the land cover map. Activities and perception of the degradation of the Results forest reserve by local populations Image classification accuracies In order to assess the perception of the local popu- lations on the state and the management modalities The computed error matrices for the classified images of the reserve, semi-structured interviews were con- revealed an overall accuracy of more than 80% for the ducted and mainly involved 42 household heads three classification dates (Table 1). The classification including village Chiefs. The interviews were con- of LULC using the Landsat 8 OLI for the year 2018 ducted in five villages: Mfentame, Njiketnkiet, showed the highest classification accuracy (90%) and Njiloum, Njinka and Njintout. They were selected kappa index (0.83) compared to the Landsat 4 TM because of their proximity to the forest reserve. The images used for the years 1988 which gave the lowest household heads were interviewed on their activ- values (85% for accuracy and 0.80 for kappa index). ities carried out in the reserve and the current state of the reserve, in order to get their perception of Land use/land cover dynamics from 1988 to 2018 the changes that have taken place over time. These changes refer to the evolution of land use patterns. In 1988, the land cover mapping identified four The interviews were supplemented by participatory classes: bare soil, built up area, savannah and forest. observations. The difference in perceptions between These different land cover classes have evolved in size respondents was tested using the kruskal-wallis over the years, either by gaining or by losing surface ANOVA test with the STATISTICA 12 software. area to give way to other land cover such as croplands In case of significant differences between the tested that appeared only in 2000 and 2018 (Figure 3). Out of variables, the Mann–Whitney U Test was used to the four classes of land cover of 1988 (Figure 3A), the separate the means. forest occupied 1345.2 ha, corresponding to 77.31% of GEOLOGY, ECOLOGY, AND LANDSCAPES 309 Table 1. Values of overall accuracies and Kappa index. cropland, 8.29% into savannah, 0.13% into built up Overall errors Overall accuracies Kappa area and 0.85% into bare soil. As for the savannah Landsat imagery (%) (%) index zone, 4.22% remained intact and 9.48% were con- 4TM (1988) 15 85 0.80 verted into cropland, 0.3% into forest, 0.13% into 7 ETM (2000) 12 88 0.82 8 OLI_TIRS 10 90 0.83 built up area and 0.03% into bare soil. (2018) Between 2000 and 2018, 0.21% of bare soil remained in this category and 0.89% was converted to other forms of land use. 0.33% of built up area remained unchanged and 0.31% was converted to the total surface area of the reserve, followed by savan- other categories. Cropland area underwent a change nah with 246.21 ha (14.15%), bare soils with 145.39 ha of 12.68% to other forms of land use. With regard to (8.36%), and traces of human activities (mainly built the forest, of the 63.44% in 2000, 26.5% was converted up areas) with 3.2 ha (0,18%) (Figure 3A). Cropland to other forms of land use and mainly to savannah areas were absent (Figure 3A and Table 2). (14.48%). 10.21% of the savannah zone remained Changes in land cover became noticeable from unchanged against 3.22% that were converted into the year 2000 (Figure 3B). The forest gradually other forms of land use. decreased from 1345.20 ha in 1988 to 1103.77 ha in 2000 to reach 664.29 ha in 2018. Between 1988 and 2000, the forest cover decreased from 77.31% to Local perception of degradation of Melap Forest 63.44% (corresponding to an annual change of Reserve −1.50%). Similarly, and during the same period, the bare soil decreased from 8.36% to 1.10% (annual An analysis of respondents’ perceptions about changes change of −7.23%) while the built up area showed in the MFR reveals that for only 7.14% of respondents, an exponential increase with an annual change of the reserve has changed over the years compared to 20.81%. Cropland which did not exist in 1988, 78.58% who did not perceive any change. Besides, gained 372.23 ha corresponding to 21.39% of the 14.29% of respondents have no opinion on changes total surface area of the reserve in 2000; then, in the reserve. Statistical analysis shows that there is a slight regression was observed in 2018 with an a significant difference in the mean of respondents’ annual change of −0.32%. Between 2000 and 2018, perceptions about changes in the MFR (p = 0.0069). the forest lost 11% of its surface, corresponding to The proportion of respondents who say that the MFR an annual change of −2.21%. This forest lost has remained intact is higher than those who say they occurred simultaneously with an increase of bare have no idea (p = 0.012) on one hand; and also higher soils and savannah which presented the annual than those who say there have been changes changes of 38.12% and 7.79% respectively. (p = 0.010) on the other hand. The annual rate of change was negative in all per- The explanation of those who don’t perceive the iods for forest cover (Figure 4). Built up showed change is that, the MFR continues to provide goods a continuous positive dynamic which was much and services as before. The changes observed and more marked between 1988 and 2018 (Figure 4). evoked by a low proportion of respondents is the Considering the whole period 1988–2018, the forest reduction of the forest cover due to the earthworks was the only land use class whose surface area carried out for the asphalting of the national road regressed: 22.7 ha/year corresponding to a total loss number 6 (Foumban-Manki section), which crosses of 680.9 ha over the 30-year period the MRF for 5 km. In addition, other reasons were The transition matrices (Table 3) highlight the dif- raised to justify change cover, there were the cut ferent forms of conversion that land use classes have down of trees for firewood and the increase in agricul- undergone between 1988 and 2018. tural areas inside the reserve. The transition matrix for the period 1988–2000 shows that 3.82 ha (0.22%) of estimated 8.36% Activities practiced within the Melap Forest (145.39 ha) bare soil remained intact and 8.13% Reserve underwent conversion to other forms of land use. Similarly, 1.76 ha (0.10%) of built up area remained Respondents reported the main activities practiced in unchanged and 0.8% (1.44 ha) underwent conversion the forest reserve. As shown in Figure 5, agriculture is to other forms of land use. Cropland is the new form the most common activity, cited by 48% of respondents, of land use (21.38%) and was created during this followed by wood extraction (24%), livestock raising period as a result of the conversion of the other cate- (7%), and hunting (5%). 17% of respondents stated gories. The forest was subject to degradation of that they do not carry out activities in the reserve. 17.53%, since out of the 77.31% of the area occupied Forest cover in the reserve has been degraded lar- by this category in 1988, only 59.78% remained gely due to agricultural activities. In 2018, 351.11 ha or unchanged in 2000. 8.26% was converted into 20.18% of the area was under cultivation. The main 310 L. F. TEMGOUA ET AL. Figure 3. Land use/land cover dynamics of Melap Forest Reserve from 1988 to 2018. GEOLOGY, ECOLOGY, AND LANDSCAPES 311 Table 2. Change in land use /land cover from 1988 to 2018 in Melap Forest Reserve. 1988 2000 2018 Annual change from 1988 to 2018 (%) Land cover ha % ha % ha % 1988–2000 2000–2018 1988–2018 Bare soil 145.39 8.36 19.21 1.10 151.03 8.68 −7.23 38.12 0.13 Built up 3.2 0.18 11.19 0.64 12.25 0.70 20.81 0.53 9.43 Cropland 0.0 0.0 372.23 21.39 351.11 20.18 - −0.32 - Forest 1345.2 77.31 1103.77 63.44 664.29 38.18 −1.50 −2.21 −1.69 Savannah 246.21 14.15 233.60 13.43 561.32 32.26 −0.43 7.79 4.27 Total 1740 100 1740 100 1740 100 - - - Figure 4. Annual rate of land use/land cover change in Melap Forest Reserve between 1988 and 2018. Table 3. Land use/Land cover change matrix in Melap Forest process within the Melap Forest Reserve. From 1988 Reserve from 1988 to 2018. to 2018, the forest class lost 680.9 ha corresponding to Proportion per class (%) 50.62% of its original surface area. Analysis of defor- LULC classes Bare soil Built up Cropland Forest Savannah estation rates for the considered time periods 1988–2000 0.22 0.30 3.58 3.34 0.91 Bare soil 1988–2000 and 2000–2018 reveals deforestation rates Built up 0.01 0.10 0.06 0.01 0.00 of 1.5% per year and 2.21% per year respectively. This Forest 0.85 0.13 8.26 59.78 8.29 high rate of deforestation, mostly observed for Savannah 0.03 0.13 9.48 0.30 4.22 2000–2018 0.21 0.00 0.53 0.28 0.08 the second period, shows that the forest reserve is Bare soil under increasing pressure over time. The rate of defor- Built up 0.04 0.33 0.24 0.00 0.04 Cropland 4.27 0.31 8.71 0.67 7.44 estation observed in our study area is higher than Forest 3.64 0.03 8.34 36.94 14.48 those reported on one hand, by De Wasseige et al. in Savannah 0.54 0.04 2.35 0.29 10.21 2012 (0.10%) for the period of 1990–2000 in Note: The unchanged area proportion of each land use class was marked in bold Cameroon, and on the other hand, by Ernst et al in 2013 (0.26%) for the Congo basin during the 2000–2005 period. crops were maize (Zea mays), groundnuts (Arachis While forests decreased, other land use classes hypogaea), potato (Ipomoea batatas) and cassava increased in the study area. Cropland and savannah (Manihot esculenta). The extraction of wood is are the classes with the highest increases: 11.7 ha/year the second most important activity practiced within and 10.5 ha/year, respectively. This trend of decreasing the forest reserve (Plate 1). Concerning livestock rais- forest cover in favour of other land use types was ing, the nomadic herders in search of pasture, graze observed by many authors in other protected areas in their animals in the reserve. They are also responsible Cameroon; some of these authors are Wafo et al. for bush fires which they use to renew the pasture. (2005) in Laf-Madjam forest reserve; Fokeng and Meli (2015) in Santchou wildlife reserve; Temgoua et al. (2018b) in the teaching and research forest of Discussion Dschang’s University; Djiongo et al. (2020) in Bouba Error matrices calculated for the classified images Ndjidda national park and Fokeng et al. (2020) in showed a general confusion of less than 15% with the Metchie-Ngoum forest reserve. However, this is not accuracy and degree of representation of ground specific to Cameroon, as similar observations have truths (Kappa index) varying from 0.80% to 0.83%. also been made in the classified forest of Tiogo in The classifications performed are reliable if we refer to Burkina Faso by Tankoano et al. (2016); in the the scale of Landis and Koch (1977). Results on land Manda national park in Chad by Benoudjita and cover change analysis show an ongoing deforestation Djinet Ignassou (2017); in the Djoli-Kera classified 312 L. F. TEMGOUA ET AL. Figure 5. Activities carried out in the Melap Forest Reserve by local communities. reserve is prohibited, local people carry out several activities there, including agriculture but also wood extraction. This situation was also reported by Fokeng and Meli (2015) in the Santchou forest reserve, Temgoua et al. (2010) in the Mbalmayo forest reserve and Fokeng et al. (2020) in Metchie-Ngoum forest reserve. Indeed most forest reserves of Cameroon are being transformed by rising population growth from within and without alongside settlement related activ- ities like firewood harvesting, logging, farming and grazing (Fokeng et al., 2020). In fact in the Congo basin in general, deforestation is linked to increasing slash-and-burn agricultural activities, artisanal timber logging, firewood collection and charcoal production (Declée et al., 2014; Megevand, 2013). Kissinger et al. (2012) showed that agriculture is the main cause of Plate 1. Logged pine for cambium extraction. deforestation in tropical areas, contributing to 35% of forest destruction in Africa. In MFR, the extraction of wood is the second most important activity practiced within the forest reserve. forest in Chad by Temgoua et al. (2018b); in the The wood is intended for personal household use but Yangambi Biosphere Reserve in the Democratic also for sale. This sale is facilitated by the national road Republic of Congo by Kyale et al. (2019) and in number 6 which crosses the reserve in its western part, Katimok Forest Reserve in Kenya by Jebiwott et al. but also by the proximity of Foumban city. Villages (2020). surrounding the classified forest largely depend on Increasing anthropisation is the cause of the firewood as their main source of energy. Eucalyptus changes in land cover and land use that have occurred and Pine trees are the most exploited species. These in the Melap Forest Reserve over the last 30 years. We species are used for firewood, but also for the con- can note that in 1988, there were no croplands in the struction of houses (poles and slats). Pine is also used reserve. However, this class is present in the 2000 and to make indoor furniture and panelling. In recent 2018 images. The practice of agriculture in the reserve years, logging of pine tree for cambium extraction can be explained by the relaxation of controls in the has intensified in the reserve, creating a little more early 1990s, which facilitated its encroachment by damage to the forest vegetation. The rehabilitation of local people. This situation was also exacerbated by the national road number 6 which passes inside the the economic crisis and the decline in civil servants’ reserve would have also had an impact on the degra- salaries in the 1990s. In Cameroon, with population dation of the vegetation cover. According to Gillet growth and in the absence of a real strategy to control et al. (2016), the development of road networks is and secure reserves and wooded areas, the phenom- most often accompanied by negative impacts on forest enon of illegal occupation has developed throughout cover. the decades. Despite the fact that access to the forest GEOLOGY, ECOLOGY, AND LANDSCAPES 313 An analysis of respondents’ perceptions of changes degradation of the reserve, it is important that they in the MFR reveals that 78.58% don’t perceive the should be involved in the management and restoration degradation. This can be explained by the fact that, through participatory management. local populations don’t yet feel any concrete impact in their activities of agriculture and wood extraction that Acknowledgments they carry out in the reserve. This may also reflect an implicit refusal by local people to acknowledge that We wish to thank the populations of surveyed villages for their illegal activities have a negative impact on the their collaboration. forest reserve. Since 2015, the management of the reserve has been Disclosure statement transferred to ANAFOR with the objective of restoring degraded forest cover. However, surveys show that No potential conflict of interest was reported by the only 59.52% of people are aware of this management author(s). entity. This reflects the lack of awareness, but also the fact that these restoration actions are not yet percep- tible on the field and does not yet have an impact on ORCID forest cover, as the project only started in 2017. In the Lucie Félicité Temgoua http://orcid.org/0000-0001- next few years, this project, if well carried out, could 9708-0411 help restore the forest cover. As an example, Gbedahi et al. (2019) reports on the reconstitution of vegetation References cover in the commune of Bassila in northern Benin through the implementation of a forest resource Arun, M., Sananda, K., Surendra, K. C., Rituraj, S., & restoration project. Mishra, P. K. (2012). Comparison of support vector Since its creation, the management methods of the machine and maximum likelihood classification techni- que using satellite Imagery. International Journal of Melap Forest Reserve were based on repression char- Remote Sensing and GIS, 1(2), 116–123. acterized by strict control and exclusion of local Benoudjita, N., & Djinet Ignassou, A. (2017). Comment les communities that were not involved in the manage- aires protégées structurent les écosystèmes des ment (Njoukam et al., 1996). This management périphéries. International Journal of Biological Chemical approach focussed on the central administration has Sciences, 11(5), 2225–2242. https://doi.org/10.4314/ijbcs. shown its limits and local communities continue to v11i5.22 Boser, B. E., Guyon, I., & Vapnik, V. (1992). A training invade the reserve as the State no longer has the algorithm for optimal margin classiers. In Proceedings of means to ensure control. It would be unrealistic the Fifth Annual Workshop on Computational Learning today to envisage sustainable management of the Theory, 144–152. Pittsburgh: ACM Press. MFR without taking into account the presence and Carnol, M., & Verheyen, K. (2010). Les services interests of the various stakeholders involved in the écosystémiques dans les forêts mélangées et pures. Forêt Wallonne, 106, :49–59. forest, including local communities. It is essential to Chako, S. M. (2016). Evaluation des opportunités de restau- involve them in the planning and the management ration des paysages forestiers : Cas de la Réserve Forestière processes of the reserve as they can play an important de Melap à Foumban. Mémoire d’Ingénieur des Eaux, role in land use/land cover change. Then, we strongly Forêts et Chasses, FASA. recommend that local communities of the MFR Chevassus-au-louis, B., & Pirad, R. (2011). Les services should be associated in the restoration project, and écosystémiques des forêts et leur rémunération éventuelle (Vol. LXIII). Revue Forestière Française. encouraged to set up individual or community affor - Côte, S. (1993). Plan de zonage du Cameroun estation of the reserve. forestier méridional : Objectifs, méthodologie, plan de zonage préliminaire. MINEF-ACDI-PTI. Cortes, C., & Vapnik, V. (1995). Support-vector networks. Conclusion Machine Learning, 20(3), 273–297. https://doi.org/10. 1007/BF00994018 Land cover and land use dynamics between 1988 and De Wasseige, C., Devers, D., Marcken, P., Eba’a Atyi, R., 2018 in the Melap Forest Reserve is characterized by Nasi, R., & Mayaux, P. (2009). Les forêts du Bassin du a significant and continuous regression of the forest Congo – État des forêts 2008. Office des publications de area in favour of other types of land use, particularly l’Union Européenne. cropland and savannah. The change is mainly due to De Wasseige, C., Marcken, P., Bayol, N., Hiol Hiol, F., Mayaux, P., Desclée, B., Nasi, R., Billand, A., the weakening of state control of the reserve, which Defourny, P., & Eba’a Atyi, R. (2012). Les forêts du has led to its anarchic invasion by local communities. Bassin du Congo – État des forêts 2010. Office des pub- Agriculture and wood extraction are the main activ- lications de l’Union européenne. ities at the base of this change. Livestock raising, lead- Declée, B., Mayaux, P., Hansen, M., Amani, P. L., ing to bush fires, is also carried out in the reserve. Sannier, C., Mertens, B., Hӓusler, T., & Lubamba, J.- Since local communities are the main actors in the P. K. (2014). Evolution of forest cover at a national and 314 L. F. TEMGOUA ET AL. regional scale and drivers of change. In C. De Wasseige, Biotechnologie Agronomie Société Environnement, 20(2), J. Flynn, D. Louppe, F. Hiol Hiol, & P. Mayaux (Eds.), 183–194. https://popups.uliege.be/1780-4507/index.php? Forest of the Congo Basin - State of the Forest 2013 (pp. id=13022 21–46). Weyrich, Belgium. Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Djiongo, J. E. B., Desrochers, A., Avana, M. L. T., Khasa, D., Turubanova, S. A., Tyukavina, A., Thau, D., Zapfack, L., & Fotsing, E. (2020). Analysis of Stehman, S. V., Goetz, S. J., Loveland, T. R., spatio-temporal dynamics of land use in the Bouba Kommareddy, A., Egorov, A., Chini, L., Justice, C. O., & Ndjidda National Park and its adjacent zone (North Townshend, J. R. G. (2013). High-resolution global maps Cameroun). Open Journal of Forestry, 10(1), 39–57. of 21st-century forest cover change. Science, 342(6160), https://doi.org/10.4236/ojf.2020.101004 850–853. https://doi.org/10.1126/science.1244693 Donnay, J. P. (2000). Les spartiocartes en composition Harris, N. L., Brown, S., Hagen, S. C., Saatchi, S. S., colorée. Bulletin de la Société Géographique de Liège, 38 Petrova, S., Salas, W., Hansen, M. C., Potapov, P. V., & (1), 43–61. https://popups.uliege.be/0770-7576/index. Lotsch, A. (2012). Baseline map of carbon emissions from php?id=2506&file=1 deforestation in tropical regions. Science, 336(6088), Ernst, C., Mayaux, P., Verhegghen, A., Bodart, C., 1573–1576. https://doi.org/10.1126/science.1217962 Christophe, M., & Defourny, P. (2013). National forest IPCC. (2014). Summary for policymakers. In C. B. Field, cover change in Congo Basin: Deforestation, reforesta- V. R. Barros, D. J. Dokken, K. J. Mach, tion, degradation and regeneration for the years 1990, M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, K. L. Ebi, 2000 and 2005. Global Change Biology, 19(4), Y. O. Estrada, & R. C. Genova (Eds.), Climate change 1173–1189. https://doi.org/10.1111/gcb.12092 2014: Impacts, adaptation, and vulnerability (pp. 1-32). FAO. (2011). La situation des forêts dans le bassin amazo- Cambridge University Press. nien, le bassin du Congo et l’Asie du Sud-Est. Rapport Jebiwott, A., Ogendi, G. M., Makindi, S. M., & Esilaba, M. O. préparé pour le Sommet des trois bassins forestiers tro- (2020). Forest cover change and its impacts on ecosystem picaux Brazzaville. 31 mai-3 juin, 2011. services in Katimok forest reserve, Baringo County, FAO. (2016). Global forest resource assessment 2015. How Kenya. In W. Leal Filho, (Ed.), Handbook of Climate are the world’s forests changing? (Second ed.). Food and Change Resilience (pp. 515–536). Springer Nature. Agricultural Organization of the United Nations. https://doi.org/10.1007/978-3-319-93336-8_155 FAO. (2020). Global forest resources assessment 2020. Key Kissinger, G., Herold, M., & De Sy, V. (2012). Drivers of findings . Food and Agriculture Organization of United Deforestation and Forest Degradation: A Synthesis Report Nation. http://www.fao.org/3/CA8753EN/CA8753EN. for REDD + Policymakers. Lexeme Consulting. pdf Kyale, J. K., Wardell, D. A., Mikwa, J.-F., Kabuanga, J. M., Faure, J. J. (1986). Réserve forestière de Melap. Plan Monga, N. A. M., Oszwald, J., & Doumenge, C. (2019). d’aménagement (1987-1996).MINAGRI/MESIRES. Dynamique de la déforestation dans la Réserve de Fokeng, M. R., Forje, W. G., Meli, M. V., & Bodzemo, B. N. biosphère de Yangambi (République démocratique du (2020). Multi-temporal forest cover change detection in Congo) : Variabilité spatiale et temporelle au cours des the Metchie-Ngoum protection forest reserve, West 30 dernières années. BOIS & FORETS DES TROPIQUES, Region of Cameroon. The Egyptian Journal of Remote 341(3), 15–28. https://doi.org/10.19182/bft2019.341. Sensing and Space Sciences, 23(1), 113–124. https://doi. a31752 org/10.1016/j.ejrs.2018.12.002 Lambin, E. F., Geist, H. J., & Lepers, E. (2003). Dynamics of Fokeng, M. R., & Meli, M. V. (2015). Modelling drivers of land-use and land –cover change in tropical regions. forest cover change in the Santchou Wildlife Reserve, Annual Review of Environment and Resources, 28, West Cameroon using remote sensing and land use 205–241. http://dx.doi.org/10.1146/annurev.energy.28. dynamic degree indexes. Canadian Journal of Tropical 050302.1054591 Geography, 2(2), 29–42. Landis, J. R., & Koch, G. G. (1977). The measurement of Foley, J. A., De Fries, R., Asner, G. P., Barford, C., Bonan, G., observer agreement for categorical data. Biometrics, 33 Carpenter, S. R., Stuart, C., & Snyder, P. K. (2005). Global (1), 159–174. https://doi.org/10.2307/2529310 consequences of land use. Science, 309(5734), 570–574. Le Quéré, C., Andrew, R. M., Canadell, J. G., Sitch, S., https://doi.org/10.1126/science.1111772 Korsbakken, J. I., Peters, G. P., Manning, A. C., Gbedahi, O. L. C., Biaou, S. S. H., Mama, A., Zaehle, S., Tans, P. P., Houghton, R. A., Keeling, R. F., Gouwakinnou, G. N., & Yorou, N. S. (2019). Alin, S., Andrews, O. D., Anthoni, P., Barbero, L., Dynamique du couvert végétal à Bassila au nord Bénin Bopp, L., Chevallier, F., Chini, L. P., Ciais, P., Currie, K., pendant et après la mise en œuvre d’un projet & Zaehle, S. (2016). Global carbon budget 2016. Earth d’aménagement forestier. International Journal of System Science Data, 8(2), 605–649. https://doi.org/10. Biological and Chemical Sciences, 13(1), 311–324. 5194/essd-8-605-2016 https://doi.org/10.4314/ijbcs.v13i1.25 Loveland, T. R., & Dwyer, J. L. (2012). Landsat: Building Geist, H. J., & Lambin, E. F. (2002). Proximate causes and a strong future. Remote Sensing of Environment, 122, underlying driving forces of tropical deforestation. 22–29. https://doi.org/10.1016/j.rse.2011.09.022 BioScience, 52(2), 143–150. https://doi.org/10.1641/ Lucas, R., Blonda, P., Bunting, P., Jones, G., Inglada, J., 0006-3568 (2002)052[0143:PCAUDF]2.0.CO;2 Arias, M., Kosmidou, V., Petrou, Z. I., Manakos, I., Giam, X. (2017). Global biodiversity loss from tropical Adamo, M., Charnock, R., Tarantino, C., Mücher, C. A., deforestation. Proceedings of the National Academy of Jongman, R. H. G., Kramer, H., Arvor, D., Honrado, J. P., Sciences USA, 114(23), 5775–5777. https://doi.org/10. & Mairota, P. (2015). The earth observation data for 1073/pnas.1706264114 habitat monitoring (EODHaM) system. International Gillet, P., Vermeulen, C., Feintrenie, L., Dessard, H., & Journal of Applied Earth Observation and Garcia, C. (2016). What are the causes of deforestation Geoinformation, 37, 17–28. http://dx.doi.org/10.1016/j. in the Congo Basin? Literature review and case studies. jag.2014.10.011 GEOLOGY, ECOLOGY, AND LANDSCAPES 315 Malhi, Y., Gardner, T. A., Goldsmith, G. R., Silman, M. R., & Temgoua, L. F. (2011). Déterminants socio-économiques et Zelazowski, P. (2014). Tropical forests in the anthropocene. écologiques de la plantation d’arbres producteurs de bois Annual Review of Environment and Resources, 39(1), d’œuvre et d’artisanat dans l’Ouest Cameroun. Thèse de 125–159. https://doi.org/10.1146/annurev-environ-030713- Doctorat. Université de Bordeaux 3. Temgoua, L. F., Allaissem, B., Tchamba, M., Saradoum, G., Mama, A., Sinsin, B., De Canniere, C., & Bogaert, J. (2013). Osée, M. M., & Momo Solefack, M. C. (2018b). Spatio- Anthropisation et dynamisation des paysages en zone temporal dynamic of land use and land cover in the soudanienne au nord du Bénin. Tropicultura, 31(1), classified forest of Djoli-Kera, South-Eastern, Chad. 78–88. http://www.tropicultura.org/text/v31n1/78.pdf Open Journal of Forestry, 8(3), 283–296. https://doi.org/ Megevand, C. (2013). Deforestation trend in the Congo bassin. 10.4236/ojf.2018.83019 Reconciling economic growth and forest protection. World Bank. Temgoua, L. F., Momo Solefack, M. C., Nguimdo Voufo, V., Momo Solefack, M. C., Njouonkouo, A. L., Temgoua, L. F., Tagne Belibi, C., & Tanougong, A. (2018a). Spatial and Djouda Zangmene, R., Wouokoue Taffo, J. B., & temporal dynamic of land-cover/land-use and carbon Ntoukpa, M. (2018). Land use/land cover change and anthropogenic causes around Koupa-Matapit gallery for- stocks in Eastern Cameroon: A case study of the teaching est, West-Cameroon. Journal of Geography and Geology, and research forest of the University of Dschang. Forest 10(2), 56–65. https://doi.org/10.5539/jgg.v10n2p56 Science and Technology, 14(4), 181–191. https://doi.org/ Naguib, A. M., Farag, M. A., Yahia, M. A., Ramadan, H. H., & 10.1080/21580103.2018.1520743 Abd Elwahab, M. S. (2009). Comparative study between Tsayem, D. M. (2010). La télédétection en tant qu’outil de support vector machines and neural networks for lithological mesure de la réduction de la déforestation (REDD). Revue discrimination using hyperspectral data. Egypt Journal of Télédétection, 9(3–4), 245–257. https://www.archivescon Remote Sensing and Space Science, 12(1), 27–42. temporaines.com/articles/391 Njoukam, R., Tchouamo, I. R., & Peltier, R. (1996). SOS : La Usman, M., Liedl, R., Shahid, M. A., & Abbas, A. (2015). réserve forestière de Melap (Foumban). Le Flamboyant, Land use/land cover classification and its change detec- 38, 12–16. tion using multi-temporal MODIS NDVI data. Journal of Pan, Y., Birdsey, R. A., Fang, J., Houghton, R., Kauppi, P. E., Geographical Sciences, 25(12), 1479–1506. https://doi.org/ Kurz, W. A., Phillips, O. L., Shvidenko, A., Lewis, S. L., 10.1007/s11442-015-1247-y Canadell, J. G., Ciais, P., Jackson, R. B., Pacala, S. W., Verburg, P. H., Neumann, K., & Nol, L. (2011). Challenges McGuire, A. D., Piao, S., Rautiainen, A., Sitch, S., & Hayes, D. (2011). A Large and Persistent Carbon Sink in using land use and land cover data for global change in the World’s Forests. Sciences, 333(6045), 988–993. studies. Global Change Biology, 17(2), 974–989. https:// https://doi.org/10.1126/science.1201609 doi.org/10.1111/j.1365-2486.2010.02307.x Peng, J., Wu, J., Yin, H., Li, Z., Chang, Q., & Mu, T. (2008). Wafo, T. G., Fotsing, J. M., & Huaman, D. (2005). Evolution Rural land use change during 1986–2002 in Lijiang, du couvert végétal dans la réserve forestière de Laf- China, based on remote sensing and GIS data. Sensors, 8 Madjam (Extrême nord du Cameroun) de 1976 à 2003. (12), 8201–8223. https://doi.org/10.3390/s8128201 Anais XII Simposio Brasileiroo de Sensoriamento Tankoano, B., Hien, M., N’Da, D. H., Sanon, Z., Remoto, INPE: 3379–3386. Yameogo, J. T., & Somda, I. (2016). Anthropogenic pres- Zekeng, J. C., Sebego, R., Mphinyane, W. N., Mpalo, M., sure and vegetation dynamics in the classified forest of Nayak, D., Fobane, J. L., Onana, J. M., Funwi, F. P., & Tiogo in Burkina Faso: Contribution of remote sensing. Abada Mbolo, M. M. (2019). Land use and land cover Tropicultura, 34(2), 193–207. http://www.tropicultura. changes in Doume Communal Forest in eastern org/text/v34n2/193.pdf Cameroon: Implications for conservation and sustainable Temgoua, L., Peltier, R., & Owona Ndongo, P. A. (2010). La management. Modeling Earth Systems and Environment, réserve forestière de Mbalmayo: Pratiques et mode 5(4), 1801–1814. https://doi.org/10.1007/s40808-019- d’accès des populations locales. Le Flamboyant, 66, 25–29. 00637-4
Journal
Geology Ecology and Landscapes – Taylor & Francis
Published: Oct 2, 2022
Keywords: Spatio-temporal dynamic; forest cover lost; land use/land cover; sustainable management; melap forest reserve