Article Versions
Export Article
Cite this article
  • Normal Style
  • MLA Style
  • APA Style
  • Chicago Style
Research Article
Open Access Peer-reviewed

Cloud Computing and Machine Learning for Analyzing Spatiotemporal Dynamics of Mangrove Ecosystems in the Grand Saloum (Senegal and Gambia)

Gayane Faye , Dome Tine, Charles Diédhiou, Claude Sene, Alioune Seydi, Mouhamadou Moustapha Mbacké Ndour
American Journal of Environmental Protection. 2021, 9(1), 29-42. DOI: 10.12691/env-9-1-4
Received September 17, 2021; Revised October 22, 2021; Accepted November 01, 2021

Abstract

The Grand Saloum is characterized by a vast coastal plain cut by a dense hydrographic network and populated by mangrove plant formations. It is an ecosystem of capital importance in view of its ecological, socio-economic and environmental role. However, the Saloum delta remains a complex and very sensitive environment, particularly in the context of climate change. It therefore deserves special attention for better conservation. The objective of this study is to analyze the spatiotemporal dynamics of its mangrove ecosystems in relation to the variability of rainfall. The methodology is based on the exploitation of Landsat satellite images time series using Machine Learning technic from the Google Earth Engine platform to make the diachronic maps of mangrove ecosystems and analyze its relationship with rainfall. The results showed an expansion of mangrove areas in the Gambian part where the surface increased from 9 381 ha in 1988 to 11611 ha in 2020 which represents an overall growth of 23,8%. In the Senegalese part, mangrove surface increased from 52 616 ha to 62 300 between 1988 and 2020 which is +18% growth. The detection of changes showed an important development of mangrove along the Saloum during the first decade and a strong growth in the Gambian part from the 2000s. The vegetation index showed a regeneration of the mangrove between 2000 and 2020. The temporal dynamics of the mangrove is strongly correlated with the rainfall variability.

1. Introduction

Senegal shares with Gambia an ecological complex that includes a vast mangrove area crossed by several waterway, temporarily flooded lagoons, marshes and mud flats 1. This marine estuary with coastal ecosystem contains a wide variety of aquatic and terrestrial landscapes and ecosystems, which support many plant and animal species. It also protects, sustains and enhances livelihoods and ultimately promotes the resilience of particularly vulnerable communities. Indeed, mangrove formations are the basis of multiple functions including coastal fixation by trapping sediments, reproduction of fauna, supply of various products (wood, oysters, fish, arches, etc.) and direct and/or indirect services to populations 2.

This strategic resource is subject to many pressures generated by a complex set of both natural and anthropogenic causes. The natural factors that affect the biodiversity of the Great Saloum are essentially climatic deterioration, water salinization, land acidification and coastal erosion. Population growth has led to overexploitation of resources, destruction and fragmentation of habitats 3. These combined effects make this ecosystem very vulnerable that deserves special conservation attention. Conservation initiatives have taken place on both sides of the border, including the Saloum Delta Biosphere Reserve (Senegal), classified as a Ramsar site and World Heritage, and the Niumi National Park (Gambia). The interest of this area has also attracted scientific curiosity, thus several research studies have been conducted in the area 4, 5, 6. However, the understanding of the spatiotemporal dynamics of the mangrove in the face of climatic variations remains poor.

This study aims to use the long time series of Landsat images to maps the spatiotemporal dynamics of the mangrove ecosystem of the Grand Saloum complex over the last 40 years (1988-2020) and analyze its relation with the spatiotemporal variability of rainfall. This by using Cloud computing (Google earth Engine-GEE) and Machine learning which are new methods of processing of large time series of images and extracting information’s with more accuracy than the classical methods of classification. Indeed, in recent years, several works have shown the interest of using remote sensing for monitoring wetlands. These new advanced techniques of geospatial data analysis are likely to provide new insights into a very complex ecosystem such as the Grand Saloum.

Normalized Difference Vegetation Index (NDVI) and Vegetation Condition Index (VCI) which are the most widely used vegetation indices 7, 8 is used the analyze the state of the mangrove. NDVI allows to inform on chlorophyll activity which provides information on the state of health on the mangrove and its density 9, 10, 11. The VCI compares the current NDVI to the range of values observed in the same period in previous years, it allows to evaluate mangrove condition in relation to its situation over the study period 10. Finally, CHIRPS (Climate Hazards Group Infrared Precipitation) rainfall is used to to understand the influence of rainfall on the dynamics of mangrove ecosystems.

2. Materials and Methods

2.1. Study Area

The Grand Saloum is located in the western part of Senegal and The Gambia between 13°23'20'' and 14°14'10'' North latitudes and 16°00'00'' and 15°51'40'' West longitudes. It is characterized by four main rivers, including the Saloum, Diombos, Bandiala and Gambia rivers, which flow directly into the Atlantic Ocean (Figure 1). This region is dominated by the Saloum estuary which shelters particular littoral formations of mangrove, drained by a multitude of bolongs with almost no freshwater input due to rainfall variability.

The Grand Saloum region belongs to the Senegalese-Mauritanian sedimentary basin which is made up of geological formations ranging from the Precambrian to the Quaternary. It is characterized by discontinuous Meso-Cenozoic formations over older geological formations. The sedimentary consist of sands, sandstones, limestones and whose evolution is dependent on marine dynamics.

The vegetation is essentially composed of two major formations: those occupying the submersible zones and their edges and those occupying the continental domain 2, 12, 13, 14. The vegetation formation of the submersible zones and their edges is entirely constituted of the mangrove, which is the basis of multiple functions that include coastal fixation by trapping sediments, fixation of carbon dioxide which is a greenhouse gas (hence its name of carbon sink), etc. This ecosystem also plays the role of reproduction site of the ichthyological fauna and the supply of various products (wood, oysters, fish, arches). There are six species belonging to three families:

- The Rhizophoraceae are composed of three species of the genus Rhizophora: Rhizophora racemosa of great size (up to 20 m) colonizes the edges of the channels. Behind, Rhizophora mangle and Rhizophora harrisonii, constitute a more extensive but lower formation.

- The Verbenaceae are represented only by Avicennia africana (or Avicenia nitida) and are located behind the Rhizophora, at the upper limit of the mudflats, in a dense and then discontinuous monospecific stand. The genus Avicenia is much better adapted to the over-salt than Rhizophora.

- The Combretaceous are much less abundant and are represented by Conocapus erectus and Languncularia racemosa occupying a domain that is only submerged at spring tides.

It should be noted that the spatial distribution of the different species is related to their capacity to adapt to salinity conditions 6.

The region has a tropical Sudan-Sahelian climate marked by alternating air flows depending on the time of year. It is characterized by rainfall and relative high humidity during the rainy season. Temperatures and insolation vary according to the season.

2.2. Data

Landsat images time series are used (Table 1) for studying mangrove ecosystem between 1988 to 2020 with a decennial time step: (i) in 1988, corresponding to the drought period of the 1980s and 1990s, which was marked by a strong degradation of ecosystems in the entire Sahelian strip 15, 16, 17; (ii) in 2000 coinciding with the beginning of the increase in rainfall in the Sahel; (iii) in 2020 to have the current situation and (iv) in 2010 which constitutes an intermediate year.

The images were obtained from the online image processing platform Google Earth Engine (GEE). This cloud computing platform has the advantage of containing images already corrected (geometric and atmospheric effects correction). Nevertheless, to avoid any noise on the images that could alter the quality of the classification, for each date, a synthetic image (median) was produced from all the dry season images (January to June) without cloud.

CHIRPS (Climate Hazards group Infrared Precipitation with Stations) rainfall data available on the Google Earth Engine platform are used to analyze the influence of rainfall variation on the dynamics of mangrove ecosystems. These data obtained from 0.05° x 0.05° spatial resolution satellite and validated with in situ station data allowed the creation of global precipitation time series over 40 years 18.

2.3. Methodology

The methodology can be summarized in three parts (Figure 2): the spatiotemporal analysis of land use; the analysis of mangrove dynamics and finally the analysis of the relationship between mangrove dynamics and rainfall variability.

A. Spatiotemporal analysis of land use

The Random Forest that is a Machine Learning algorithm 19, 20 is used to classify the different land use land cover classe. It is a collection n of random trees that are predictors based on a training database. Each tree produces a classification result, and the classification model result depends on the majority of votes 20.

The objective is to achieve the most accurate cartographic representation possible from the spectral and textural values of the various pixels of the image 21, 22. For this purpose, an analysis of the spectral signature of the different types of land use was carried out from the selected training areas based on existing maps and very high spatial resolution images from Google Earth.

Taking into account the objective (monitoring of mangrove ecosystems) the classification nomenclature is simplified into seven themes: (i) watercourses (river, ponds); (ii) mangrove; (iii) agricultural areas; (iv) savannah; (v) salted areas; (vi) mudflats and (vii) ocean.

To better understand the temporal dynamics of land use in the area, the Average Annual Spatial Expansion Rate (AASER) was calculated. The formula used is one of the formulas applied by 23 cited by 24 whose variable considered is the surface area (S). The formula for calculating this index is:

(1)

Where S1 and S2 corresponding respectively to the area of a land use category at date t1 and t2, t is the number of years of change between t1 and t2, e is the base of neperian logarithms (e =2.71828).

B. Analysis of mangrove dynamics

To characterize the state of the mangrove, two vegetation indices (NDVI and VCI) are used. NDVI is calculated from Near Infrared and Red bands.

(2)

- For Landsat 5 and Landsat 7: NIR (Near Infrared band) correspond to Band 4 and R (Red band) to Band 3

- For Landsat 8: NIR (Near Infrared band) correspond to Band 5 and R (Red band) to Band 4

NDVI value varies from -1 to 1: (i) -1 to 0 represent Water bodies; (ii) -0.1 to 0.1 represent snow or barre soil; (iii) 0.2 to 0.5 correspond to presence of vegetation such as mangrove which importance increases with NDVI (close to 0 indicated that less vegetation there is or its condition is strongly degraded) and (iv) 0.6 to 1.0 represent dense vegetation or tropical rainforest. For each date, the NDVI is calculated to evaluate the spatial distribution of the mangrove density.

The VCI was calculated using the following formula 25:

(3)

where NDVIi is the NDVI value of a specific date (1988 or 2000 or2010 or 2020 in this case), and NDVImax and NDVImin are the maximum and minimum NDVI values for over a specific time period (between 1988 and 2020 in this case). VCI is an index that expresses the current state of the mangrove in relation to its minimum level reached over the study period which allows to analyses the temporal dynamics of the mangrove 25, 26. A VCI value near 0% reflects a highly degraded mangrove 27 while a value around 50% corresponds to an average growth situation and values between 50 and 100% indicate optimal or above normal conditions.

C. Analysis of mangrove density in relationship with rainfall variability

Correlation between mangrove NDVI and annual average of rainfall for each decade is done. This in order to understand if the resumption of rainfall noted these last decade could impact the evolution of the mangrove. In addition, the Standardized Precipitation Index (SPI; 28) is compared with the ICV to better explain the deficit pockets observed on the mangrove maps.

(4)

Where Xi is the cumulative rainfall for year i considered (here 2020); Xm and Si are respectively the mean and standard deviation of annual rainfall for a given period (1998 to 2020 in this case).

3. Results and Discussions

3.1. Land Use Lande Cover Dynamic

A. Land cover mapping

Figure 2 shows that in Senegal, the mangrove is mainly located in the south of the Saloum River in the wettest part of the estuary with a few pockets in the north, notably in Joal-Fadiouth. The northern part of the estuary is dominated by mudflats associated with large bare and over-salted areas called tannas, which in some places border agricultural areas.

In the Gambian part, the mangrove is located along the Gambia River and is distributed mainly in four sites: (i) the National Niumu Parck, (ii) along the belongs west of Bakindick, (iii) the Sika Bolong Estuary and (iv) the Kumadi Estuary which extends to the border with Senegal (Figure 3).

Table 2 presents statistics of the evolution of the selected thematic classes. Overall, the study area is dominated by rainfed crops with an average of more than 20% in Senegal and 30% in The Gambia.

Situation in 1988: Mangroves cover an area of 52,616 hectares, i.e., 13.4% of the total area of the Senegalese study area, compared to 9,381 hectares in The Gambia (7% of the area). In the Senegalese part, the mangrove is located on south of the Saloum River. In The Gambia, the mangrove is distributed in four pockets along the Gambia River (Figure 3).

Situation in 2000: The 2000 map shows an overall increase in mangrove area of about 12%, although there is a regression in some areas such as along the belongs west of Bakindick in The Gambia (Figure 3 and Table 2). The area of salted areas has declined sharply, particularly in The Gambia, where there has been a 50% decline.

Situation in 2010: a positive evolution of mangrove areas is noted with an increase of nearly 6% over the entire study area. An extension of mangrove areas is noted on the Gambia side, especially along the belongs west of Bakindick where mangrove had disappeared in places between 1988 and 2000. The decline of tans noted in 2000 has accelerated sharply in 2010 when the mudflat is expanding.

Situation in 2020: The trend observed since 2010 has generally continued until 2020. However, there are pockets where a regression of the mangrove is important. It is especially the coastal fringe marked by a strong erosion with as a consequence a degradation or even a disappearance of the mangrove. The oceanic dynamics more and more sustained is also a factor of degradation of the mangrove especially the small plantlets. This often compromises the reforestation efforts made by the populations.

Figure 4 shows the distribution of the area of the different land use units of the protected areas. The Saloum Delta National Park (SDNP) and the Marine Protected Areas (MPAs) of Sangomar, Bamboung and Gandoul concentrate the most mangroves with 5,407, 4,626 and 4,236 hectares respectively. However, Bamboung MPA with more than half (62.6%) of its area occupied by mangrove has the highest rate of mangrove coverage followed by Gandoul with 26.9%, the other protected areas having rates below 10%. These low rates are explained by the fact that more than 50% of the reserves are located in the Atlantic Ocean.

B. Land use land cover change analysis

Many changes have been noted over the last forty years in the Grand Saloum complex, particularly in the periphery of the mangrove areas (Figure 5).

Change between 1988 and 2000: The dynamics of land use between 1988 and 2000 is marked by a notable change in the mangrove. An appearance of mangrove in the chainons (between the two rivers of the Saloum) and a degradation towards the peripheries at the lower limit of the tans. In the Senegalese part, 8.6% of the total mangrove area disappeared during this period against 22% of new mangroves representing a gain of +13.4% against +6.9% in The Gambia where 12% of mangrove surfaces are lost for 18.9% of new spaces. The transformation from mangroves to savannahs and vice versa is very important in both parts, due to the fact that grasses and other halophytic trees cohabit with the mangrove with a swing according to the climatic dynamics and anthropic actions. We also note a colonization of the mudflats by the mangrove resulting most often from reforestation campaigns. Over the same period, in The Gambia, more than 50% of the tans have disappeared to the benefit of mudflats and savannah.

Change between 2000 and 2010: The period 2000-2010 is marked by a strong appearance of new mangrove areas and a low disappearance in both countries. For example, in the Senegalese area, only 9% of the area was lost against an appearance of 13.7%, which is a jump of +4.4%. In The Gambia, a total of +13% increase was noted during this period (-3% versus +16%). As in the period 1988-2000, mangrove-savanna and savanna-mangrove transformations remained significant. The disappearance of salted areas has also accelerated, with a significant part of them being converted into savannah. This shows once again that the resumption of rainfall since the 2000s has washed out the soils leading to the development of vegetation on previously very salty land.

Change between 2010 and 2020: In the last ten years, mangrove has remained almost stable despite some losses noted on the peripheries, contrary to previous decades when mangrove losses were observed throughout the area. At the same time, the savannah areas have strongly progressed in Senegal, particularly in the northern part where the erection of micro-dams has favored land desalination. In The Gambia, the mangrove in the Kumadi estuary has recorded some pockets of degradation. However, the total area of mangroves has increased overall in The Gambia by more than 2%.

The Average Annual Spatial Expansion Rate (AASER) calculated from the above matrices allow to understand the temporal interannual dynamics of land use, the Average Annual Spatial Expansion Rate (AASER) was calculated and the results are illustrated in Figure 6.

This curve shows a continuous decline in the expansion of the mangrove from +0.98 between 1998 and 2000, to +0.55 during the decade 2000-2010 and +0.03 between 2010 and 2020. The agricultural areas, after a slight expansion (+0.44 between 1988 and 2000) have continuously regressed since 2000 with -1.55 between 2000 and 2010 and then -1.24 in the last decade). Unlike the savannah which has continued to gain ground. Mudflats and tans have also lost space.

3.2. Analysis of Spatialtemporal Magrove Danamic

Over the study period, the area of mangrove increased globally. From 52,616 ha in 1988 to 59,699 ha in 2000 before increasing to 62,317 in 2010 and 62,300 ha in 2020 in the Senegalese part of the study area. In the Gambian part, they successively increased from 9,381 in 1988, to 10,025 in 2000, 11,343 in 2010 to reach 11,611 ha in 2020. Overall, the mangrove areas have continuously increased from 1988 to 2020, following the regressions noted in the early 1980s due to the severe droughts recorded in the Sahel during this period. The evolution of mangrove areas thus seems to be a direct response to rainfall fluctuations which play an important role in water and land desalination. However, these evolutions are not homogeneous in space as illustrated by the change maps in the following paragraphs.

Figure 7 below shows the spatiotemporal evolution of the mangrove since 1988, highlighting the stable areas but also the parts where the mangrove is retreating and those where it has appeared during each decade.

Table 4 summarizes the statistics of mangrove evolution in the two sites (Saloum and Niumi) over the periods (1988-2000, 2000-2010 and 2010-2020)

a) 1988-2000 decade: a significant development of mangroves was noted along the Sine-Saloum River at the same time as the eastern periphery experienced fairly significant mangrove disappearances. A total of 4,509 hectares of mangroves were lost between 1988 and 2000, i.e., 8.6% of the total mangrove area available in 1988, compared to an appearance of 11,590 ha (+22.0%), which represents a gain of 7,081 ha (+13.6%). In the Gambia, mangrove regressed in the southwestern part towards Bakindick and Tubo Kolong. However, the progress noted in several places of the zone allowed a total overall increase of 664 ha (+6.9%) of mangrove between 1988 and 2000.

b) Decade 2000-2010: this decade is marked by a slowdown in the increase of mangrove areas in Senegal and a strong growth in the Gambian part. Indeed, in the Saloum, 9.3% of the mangrove was lost against 13.7% of appearance, that is to say an overall progression of +4.4%. The development of the mangrove observed along the Saloum River in 2000 has been maintained with an extension to the eastern periphery in areas where degradation was noted the previous decade. The most significant degradation is visible in the central part of the Saloum delta and around Joal-Fadiouth. In the Gambia, mangrove areas increased by +13.2%. This strong growth is the result of a small disappearance (-3%) combined with a significant appearance (+16.2%).

c) Décennie 2010-2020 : Dans le Saloum, les superficies de mangrove sont restées globalement constantes avec 4 990 ha apparue contre 4 976 ha de perte, représentant une perte nette de 14 hectares durant cette décennie. Les pertes les plus importantes se situent au nord-ouest, vers Foundiougne. L’évolution de la mangrove dans le Niumi entre 2010 et 2020 est restée positive mais de faible ampleur (+2,4%). Cependant, quelques poches de pertes sont à déplorer.

d) Decade 2010-2020: In the Saloum, mangrove areas remained globally constant with 4,990 ha appeared against 4,976 ha lost, representing a net loss of 14 hectares during this decade. The most important losses are located in the northwest, towards Foundiougne. The evolution of the mangrove in the Gambia between 2010 and 2020 remained positive but of low magnitude (+2.4%). However, some pockets of loss are to be deplored.

The positive developments noted during the first decade (1988-2000) indicate that mangroves have begun to recover in many areas since 1988. This suggests that the gradual recovery of rainfall in the Sahel in the 2000s 15, 29, 30 led to desalination of the land, thus favoring the development of mangroves. The positive dynamics that have been maintained until 2010 support this statement.

The differences in evolution rates between the two countries are certainly the result of a more rapid recovery of the mangrove in the Saloum, where it rapidly recolonized its environment at the end of the drought years (decade 1988-2000). It can also be attributed to conservation efforts in both countries.

The periphery of the Saloum delta is experiencing a strong dynamic marked by a succession of disappearance and appearance of mangroves. This phenomenon can be the result of exposure to anthropic effects (abusive cutting) but also to climatic hazards that induce a balancing of soil salinity levels.

The constancy of mangrove areas in the Saloum during the last decade (-4,990 ha versus +4,976) shows that the reforestation efforts of the last few years are often wiped out by losses.

3.3. Analysis of Mangrove Density in Relationship with Rainfall Variability

The 1988 mangrove NDVI index is overall below 0.4 indicating a sparse and heavily stressed mangrove. This result is consistent with what has been observed in other studies 2000s 5, 31 where it has been shown that the mangrove was severely stressed between 1972 and 1986 as a result of the drought years that hit the Sahel as shown on the map of average rainfall between 1981 and 1987 (Figure 8). Indeed, the decrease in freshwater inflow, combined with strong evaporation and penetration of marine waters, is at the origin of an increase in salinity that has greatly contributed to the degradation of the mangrove. The latter being particularly reactive to variations in water salinity 2000s 32.

In 2000, the mangrove index showed a clear improvement. The revitalization of the mangrove is particularly clear and strong in three (3) sectors: the area between the two rivers in Senegal, in the south of the Senegalese part and finally in the eastern estuary in the Gambia. The increase in rainfall after 1988, as indicated by the average rainfall map between 1990-1999 in Figure 8, has favored the leaching of halomorphic lands, thus leading to a decrease in saline lands. Indeed, the period 1990-1999 is characterized by the sporadic return of rainfall. Several studies have shown that the 1990s marked the beginning of mangrove regeneration in Senegambia 31.

The regeneration of the mangrove intensified in 2010 with, however, pockets where the mangrove remains very weak, particularly along the Saloum River in the northeast and east. This positive trend is the result of more favorable climatic conditions but also the development of mangrove reforestation activities initiated in recent years by the various projects and programs on the management and development of mangroves 33. The construction of micro dikes during this period has also played a key role in the recovery of saline lands.

The dynamics of mangrove regeneration observed in 2000 and 2010 are confirmed in 2020 with a vegetation index globally above 0.4 over almost the entire area except for a few places like southwest of Bakindick in The Gambia where the index remains low.

After the period of drought in the 1980s which negatively impacted the mangrove, there has been a continuous regeneration of the latter since 1988, with an acceleration after 2000. This state of the mangrove has multiple origins. The freshwater inflow following the rainfall recovery has greatly reduced the salinity and acidity of the soil 34. It is also important to note the implementation of several projects and programs to restore and safeguard the mangrove. We can cite for example the 1996-2001 Orientation Plan for Economic and Social Development, the Forestry Action Plan which highlights the need for mangrove management as one of the priority actions in the action program at the regional level.

Figure 9 shows the current level of degradation/regeneration compared to its average level over the study period (1988-2020). On the one hand, the peripheries of the reserves constitute the areas where the level of mangrove is low with an VCI below 0.5 (50%). On the other hand, there are pockets where the mangrove remains weak, notably south of Bétenti up to the PNDS at the Gambian border and in the Kumadi estuary in Gambia. The mangrove along the banks of the channels of the Saloum Delta is in a stable state even if weaknesses are visible in some places.

We note that the overall average level of the mangrove in 2020 is good in comparison to its average level between 1988 and 2020 with, however, parts where the degradation is very marked. This analysis has made it possible to delimit a certain number of buffer sites around the conservation areas for restoration actions.

3.4. Analysis of Mangrove Degradation in Priority Areas

Figure 10 below gives the distribution of the percentages of level (very low, low, medium, high and very high) of the mangrove, calculated from the VCI for the protected and unprotected sites delimited from Figure 9. In the protected areas, more than 80% of the mangrove is in a strong to very strong (good to very good) situation except for Sangomar MPA and PNDS which are at about 50%. This shows that the mangrove in the protected sites is globally in a good state contrary to the peripheral areas that we have delimited where the weak to average levels reach most often 40%.

A comparison of the temporal evolution of mangrove areas between protected and unprotected areas does not show much difference, although it should be noted that for unprotected sites the areas have remained globally constant since 2000, or even decreased (Figure 11). Only site 3, which is located to the north of the Sangomar MPA, has experienced a positive evolution, going from 1,561 to 1,971 ha between 2000 and 2010, an increase of more than 26%.

These different results show once again that the situation in the protected areas is better than in the peripheries. However, if actions are not taken to preserve or even restore the mangrove in these peripheral sites, the pressure on the protected sites may increase, thus threatening all the conservation and sustainable management efforts underway. It is therefore important to carry out strong actions to better protect the buffer zones for a more effective conservation of the protected areas.

3.4. General Discussion

The results of the diachronic mapping show an overall positive evolution of +19.2% of the mangrove of the Grand Saloum 1988 and 2020. This variation is marked by a strong growth of mangrove surfaces (+12%) during the first decade. From 2000, the increase in area slowed to +6% between 2000 and 2010 and then to less than +1% in the last decade. The extension of the mangrove is more important in the Gambian part with +23% (from 9,381 ha in 1988 to 11,611 ha in 2020) against +18% in the Saloum (52,616 ha in 1988 and 62,300 in 2020). However, these overall trends conceal some disparities, including (i) a slight decrease in mangrove areas between 2010 and 2020 in Saloum; (ii) significant disappearances in some places against progress in others; (iii) areas that lose their mangrove over a period before it reappears the following decade and vice versa; (iv) degradations are more marked in unprotected areas and especially the peripheries.

The spatial expansion rate of mangrove has been continuously decreasing since 1988. This indicates that the strong progression of the mangrove during the first decade (1988-2000) at the end of the drought of the 1980s has slowed down in the following years. Indeed, after the return of favorable conditions (increase of rain, decrease of salinity and acidity of soils after 1990) favored a rapid development of the mangrove which was able to recolonize a large part of the favorable pockets for the development of mangrove. This explains that the following decades (2000-2010 and 2010-2020) the expansion of the mangrove has slowed down even if it still remains positive.

These different observations show that conservation and protection efforts are often counterbalanced by the sometimes-significant losses in uncontrolled pockets. The resumption of rainfall has favored the development of mangroves contrary to climate change and its consequences such as salinization of land, coastal erosion and the development of increasingly large swells with their share of consequences on the mangrove.

Today, the major challenges for the conservation of this highly strategic but also fragile ecosystem are, among others (i) a good policy of integrated and sustainable management of mangrove ecosystems by strengthening the achievements in the Protected Areas (MPAs and PAs); (ii) a participatory management of buffer zones around the protected sites to maintain this natural barrier without which the anthropic pressure on these sites becomes a real threat; (iii) strategies or economic models that allow the populations to live from the ecosystem services of the mangrove without destroying them. This last recommendation requires the involvement of local populations in the management and restoration of the mangrove.

The negative changes of the mangrove noted are the result of several natural and anthropogenic factors.

1) Natural factors: The climatic pejoration’s, which the region experienced in the 1970s, have considerably decreased the freshwater inflow, accompanied by a global warming of the land 35, which leads to a hyper salinity in certain areas, thus causing degradation or even loss of mangroves. Coastal erosion is also a natural phenomenon that negatively impacts the mangrove. Indeed, the rupture of the Sangomar spit has led to a degradation or even a disappearance of the mangrove in front of the new breach because of the strong marine currents and waves that hit this part of the estuary. The strong swells noted in recent years and which are a consequence of climate change are often a hindrance to the development of small seedlings, often compromising the reforestation efforts initiated by the populations.

2) Anthropogenic threats: The many functions that the mangrove provides for the populations make it a particularly prized environment, thus leading to an increasing pressure 36. These growing needs for resources, driven by a growing demography in coastal areas, also contribute to the process of environmental degradation in the estuary 3. The overexploitation of mangrove wood used by populations as firewood and construction wood is also a factor in the destruction of the mangrove in the Grand Saloum. To this must be added pollution, especially industrial pollution, which will increase with the exploitation of offshore oil in the Sangomar Field.

4. Conclusion

This study analyzes spatiotemporal dynamics of mangrove ecosystems from Landsat images time series and its relationship with rainfall. The use of the Google Earth Engine platform made it possible to process this mass of data without much difficulty with machine learning algorithms that improve the accuracy of the mapping of the different land use classes.

After having been strongly tested by the drought that hit the Sahel in the 80’s decade, the mangrove has experienced a positive evolution in the last 30 years. Indeed, this study shows that from 1988 to 2020, mangrove areas have increased slightly in the Grand Saloum, with +23.8% of growth in Gambian part and +18.4% in Senegalese side. Similarly, the density of mangrove has undergone a positive evolution as shown by the NDVI which passes from an average value lower than 0.4 in 1988 to about 0.6 in 2020. In the end, the situation of the mangrove in 2020 is largely above its average situation over the last 30 years. However, there are pockets where a regression or even degradation of the mangrove is noted. These pockets are often located outside of conserved perimeters (Marine Protected Areas, natural reserves among others) where conservation seems to be successful.

The positive correlation between the evolution of mangrove ecosystems and the interannual dynamics of rainfall shows that the latter plays an important role in the regeneration of the mangrove. This can be explained by the fact that the resumption of rainfall after years of drought has favored the decrease of water salinity and also soil acidity, two key parameters in the regulation of the mangrove.

References

[1]  Republic of The Gambia., (1997). Ramsar Wetland Study. Report produced by The Department of Parks and Wildlife Management under The Ministry of Fisheries and Natural Resources with The Ramsar Bureau.
In article      
 
[2]  Ngom F., (2005). Les fonctions de la mangrove dans la structuration et la biologie des peuplements de poissons de l’estuaire du Sine Saloum, thèse de doctorat de troisième cycle, mention biologie animale, Université Cheikh Anta Diop, Dakar, 141 p.
In article      
 
[3]  Ndour, N., (2005). Caractérisation et étude de la dynamique des peuplements de mangrove de la Réserve de Biosphère du Delta du Saloum (Sénégal), Dakar, UCAD, 180 p.
In article      
 
[4]  Diop, E.S., (1990). La côte ouest-africaine. Du Saloum (Sénégal) à la Mellacorée (Rép. Guinée). ORSTOM, Coll. Etudes et Thèses, 380 p.
In article      
 
[5]  Cormier-Salem, M.C., (1994). Dynamique et usages de la mangrove dans les pays des rivières du Sud. IRD, ORSTOM Editions, collection colloques et séminaires, Paris, 357p.
In article      
 
[6]  Marius, C., (1985). Mangrove du Sénégal et de la Gambie. Ecologie-Pédologie-Géochimie : mise en valeur et aménagement. Thèse de doctorat en Sciences Naturelles, Université Louis Pasteur, Editions de l’ORSTOM, Collection Travaux et Documents, n° 193, 357 p.
In article      
 
[7]  Bappel, A.E., (2005). Apport de la télédétection Aerospatiale pour l’aide à la gestion de la sole canniere réuninnaise. Biophysique. Thèse de doctorat de l’Université de la Réunion, 2005. Français.
In article      
 
[8]  Rakotoniaina, S., Rakotomandrindra, P., Ranaivoarimanana, S., Rakotondraompiana, S., (2014). La cartographie et la télédétection comme système d’évaluation des TGRNR. Exemple de site d’application : La commune de Didy, région d’Aloatra Mangoro, Madagascar.
In article      
 
[9]  Hmimina, G., Dufrêne, E., Pontailler, J.Y., Delpierre, N., Aubinet, M., Caquet, B., de Grandcourt, A., Burban, B., Flechard, C., Granier, A., Gross, P., Heinesch, B., Longdoz, B., Moureaux, C., Ourcival, J. M., Rambal, S., Saint André, L., Soudani, K., (2013). Evaluation of the potential of MODIS satellite data to predict vegetation phenology in different biomes: An investigation using ground-based NDVI measurements, Remote Sensing of Environment, Volume 132, 2013, Pages 145-158, ISSN 0034-4257.
In article      View Article
 
[10]  Fensholt, R., Rasmussen, K., Kaspersen, P., Huber, S., Horion, S., Swinnen, E., (2013). Assessing Land Degradation/Recovery in the African Sahel from Long-Term Earth Observation Based Primary Productivity and Precipitation Relationships. Remote Sens. 2013, 5, 664-686.
In article      View Article
 
[11]  Diouf, A.A., Hiernaux, P., Brandt, M., Faye, G., Djaby, B., Diop, M.B., Ndione, J.A. and Tychon, B., (2016). Do Agrometeorological Data Improve Optical Satellite-Based Estimations of Herbaceous Yield in Sahelian Semi-Arid Ecosystems? Remote Sensing, 8, 668. http://www.mdpi.com/journal/remotesensing.
In article      View Article
 
[12]  Dia, I.M.M., (2003). Elaboration et mise en œuvre d'un plan de gestion intégrée - La Réserve de biosphère du delta du Saloum, Sénégal. UICN, Gland, Suisse et Cambridge, Royaume-Uni. xiv + 130 pp.
In article      
 
[13]  Diouf, P.S., (1996). Les peuplements de poissons des milieux estuariens de l’Afrique de l’Ouest: l’exemple de l’estuaire hyperhalin du Sine-Saloum. Thèse de doctorat, Univ. Montpellier II, 267 p.
In article      
 
[14]  Cormier-Salem, M.C., (1999). Les Rivières du Sud. Sociétés et mangroves ouest-africaines. Paris, Orstom, 2 volumes (Volume 1: 416 pp.; Volume 2: 288 pp).
In article      View Article
 
[15]  Le Barbé, L., Lebel, T., Tapsoba, D., (2002). Rainfall variability in West Africa during the years 1950-1990, J. Climate, 15(2), 187-202.
In article      View Article
 
[16]  GIEC., (2007). Rapport final du Groupe Intergouvernemental sur l’Evolution du Climat 2007.
In article      
 
[17]  IPCC., (2007). Intergovernmental Panel on Climate Change. Bilan 2007 des changements climatiques : Rapport de synthèse, 113 pages.
In article      View Article
 
[18]  Funk, C., Peterson, P., Landsfeld, M., Pedreros, D., Verdin, J., Shukla, S., Husak, G., Rowland, J., Harrison, L., Hoell, A., (2015). The Climate Hazards Infrared Precipitation with Stations -a New Environmental Record for Monitoring Extremes. Sci. Data 2015, 2, 150066.
In article      View Article  PubMed
 
[19]  Masoud, A.A., Koike, K., (2006). Arid land salinization detected by remotely-sensed landcover changes: A case study in the Siwa region, NW Egypt. Journal of Arid Environments, Volume 66, Issue 1, July 2006, Pages 151-167 2005.
In article      View Article
 
[20]  Liu, Y., Wang, Y.,and Zhang, J., (2012). “New machine learning algorithm: Random Forest” in International Conference on Information Computing and Applications, 2012, pp. 246-252.
In article      View Article
 
[21]  De Wispeleare, G., (1990). Dynamique de la désertification au Sahel du Burkina Faso: Cartographie de l’évolution et de recherches méthodologiques sur les applications de la télédétection, Thèse d’ingénieur, CIRAD-EMVT, France, 546 p.
In article      
 
[22]  Dièye, E.B., Diaw, A.T., Diatta, C.S., De Wispealare, G., (2008). Evolution spatiale de la mangrove de l’estuaire du Saloum (Sénégal) entre 1972 et 1999 : approche méthodologique par Télédétection», Journal des Sciences et Technologies 2008, vol. 6 No. 1, Faculté des Sciences et Techniques, UCAD, Dakar, 36-48.
In article      
 
[23]  Bernier, B., (1992). Introduction à la macroéconomie. Dunod, Paris, 217p.
In article      
 
[24]  Ndour, M.M.M., Ndonky, A., Sarr, C.A.T., Ndiaye, M., (2020). Dynamique de l’occupation du sol 1986-2016 et géprospective de l’évolution urbaine de la région de Dakar (Sénégal) à l’horizon 2035. Revue de géographie du laboratoire Leïdi_ISSN 0851-2515, N°24 Décembre 2020.
In article      
 
[25]  Kogan, F. N. 1990. Remote sensing of weather impacts on vegetation in non-homogeneous areas. International Journal of Remote Sensing 11: 1405-1419.
In article      View Article
 
[26]  Kogan, F.N., (1997). Global drought watch from space. Bulletin of the American Meteorological Society, vol. 78, n°4, p.621-636.
In article      View Article
 
[27]  Owrangi, M.A., Adamowski, J., Rahnemaei, M., Mohammadzadeh, A., Sharifan, R.A. (2011). Drought Monitoring Methodology Based on AVHRR Images and SPOT Vegetation Maps. Journal of Water Resource and Protection, 3, p. 325-334.
In article      View Article
 
[28]  Mickee, T.B., Doesken N.J., et Kleist J. (1993) The relationship of drought frequency and duration to time scale. Actes de la 8th Conference on Applied Climatology (Anaheim, Californie), p. 179-184.
In article      
 
[29]  Dai, A., Lamb, P.J., Trenberth, K.E., Hulme, M., Jones, P.D., Xie, P., (2004). The recent Sahel drought is real. Int. J. Climatol., 24, 1323-1331.
In article      View Article
 
[30]  Nicholson, S., (2005). On the question of the “recovery” of the rains in the West African Sahel. J. Arid Environ., 63: 615-641.
In article      View Article
 
[31]  Dièye, E.B., Diaw, A.T., Sané, T., Ndour, N., (2013). Dynamique de la mangrove de l’estuaire du Saloum (Sénégal) entre 1972 et 2010», Cybergeo: European Journal of Geography, Environnement, Nature, Paysage, document 629, mis en ligne le 09 janvier 2013, consulté le 07 octobre 2021. URL: http://journals.openedition.org/cybergeo/25671.
In article      
 
[32]  Marius, C., (1984). Contribution à l’étude des mangroves du Sénégal et de la Gambie-Ecologie-Pédologie-Géochimie. Mise en valeur et aménagement, thèse de doctorat de troisième cycle, ORSTOM, Paris, 309p.
In article      
 
[33]  FAYE, S., (2017). Les enjeux d’une gouvernance de l’estuaire du Saloum dans la perspective d’une préservation durable des patrimoines de la Réserve de la Biosphère du Delta du Saloum (Sénégal), zone d’interface homme-nature en dégradation, dans un contexte de réchauffement climatique », These de doctorat de l’Université de Jean Moulin de Saint-Etienne. https://tel.archives-ouvertes.fr/tel-02174828.
In article      
 
[34]  Dièye, E.B., Diaw, A.T., Sané, T., Sy, O., Dioh, P., (2011). Changement climatique et évolution de la mangrove dans la lagune de Joal-Fadiouth (Sénégal). In Actes du 24ème Colloque de l’AIC, Rovereto (Italie), pp183-188.
In article      
 
[35]  Niang, I., (1998). Etude de vulnérabilité des zones côtières sénégalaises aux changements climatiques: le cas des pays africains côtiers, Bull. Africains, No.10, Dakar, 25-37.
In article      
 
[36]  Diop, E.S., (1998). Contribution à l’élaboration du plan de gestion intégrée de la Réserve de la Biosphère du Delta de Saloum (Sénégal)», Dakar, UCAD-UNESCO-MAB, 86 p.
In article      
 

Published with license by Science and Education Publishing, Copyright © 2021 Gayane Faye, Dome Tine, Charles Diédhiou, Claude Sene, Alioune Seydi and Mouhamadou Moustapha Mbacké Ndour

Creative CommonsThis work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Cite this article:

Normal Style
Gayane Faye, Dome Tine, Charles Diédhiou, Claude Sene, Alioune Seydi, Mouhamadou Moustapha Mbacké Ndour. Cloud Computing and Machine Learning for Analyzing Spatiotemporal Dynamics of Mangrove Ecosystems in the Grand Saloum (Senegal and Gambia). American Journal of Environmental Protection. Vol. 9, No. 1, 2021, pp 29-42. http://pubs.sciepub.com/env/9/1/4
MLA Style
Faye, Gayane, et al. "Cloud Computing and Machine Learning for Analyzing Spatiotemporal Dynamics of Mangrove Ecosystems in the Grand Saloum (Senegal and Gambia)." American Journal of Environmental Protection 9.1 (2021): 29-42.
APA Style
Faye, G. , Tine, D. , Diédhiou, C. , Sene, C. , Seydi, A. , & Ndour, M. M. M. (2021). Cloud Computing and Machine Learning for Analyzing Spatiotemporal Dynamics of Mangrove Ecosystems in the Grand Saloum (Senegal and Gambia). American Journal of Environmental Protection, 9(1), 29-42.
Chicago Style
Faye, Gayane, Dome Tine, Charles Diédhiou, Claude Sene, Alioune Seydi, and Mouhamadou Moustapha Mbacké Ndour. "Cloud Computing and Machine Learning for Analyzing Spatiotemporal Dynamics of Mangrove Ecosystems in the Grand Saloum (Senegal and Gambia)." American Journal of Environmental Protection 9, no. 1 (2021): 29-42.
Share
[1]  Republic of The Gambia., (1997). Ramsar Wetland Study. Report produced by The Department of Parks and Wildlife Management under The Ministry of Fisheries and Natural Resources with The Ramsar Bureau.
In article      
 
[2]  Ngom F., (2005). Les fonctions de la mangrove dans la structuration et la biologie des peuplements de poissons de l’estuaire du Sine Saloum, thèse de doctorat de troisième cycle, mention biologie animale, Université Cheikh Anta Diop, Dakar, 141 p.
In article      
 
[3]  Ndour, N., (2005). Caractérisation et étude de la dynamique des peuplements de mangrove de la Réserve de Biosphère du Delta du Saloum (Sénégal), Dakar, UCAD, 180 p.
In article      
 
[4]  Diop, E.S., (1990). La côte ouest-africaine. Du Saloum (Sénégal) à la Mellacorée (Rép. Guinée). ORSTOM, Coll. Etudes et Thèses, 380 p.
In article      
 
[5]  Cormier-Salem, M.C., (1994). Dynamique et usages de la mangrove dans les pays des rivières du Sud. IRD, ORSTOM Editions, collection colloques et séminaires, Paris, 357p.
In article      
 
[6]  Marius, C., (1985). Mangrove du Sénégal et de la Gambie. Ecologie-Pédologie-Géochimie : mise en valeur et aménagement. Thèse de doctorat en Sciences Naturelles, Université Louis Pasteur, Editions de l’ORSTOM, Collection Travaux et Documents, n° 193, 357 p.
In article      
 
[7]  Bappel, A.E., (2005). Apport de la télédétection Aerospatiale pour l’aide à la gestion de la sole canniere réuninnaise. Biophysique. Thèse de doctorat de l’Université de la Réunion, 2005. Français.
In article      
 
[8]  Rakotoniaina, S., Rakotomandrindra, P., Ranaivoarimanana, S., Rakotondraompiana, S., (2014). La cartographie et la télédétection comme système d’évaluation des TGRNR. Exemple de site d’application : La commune de Didy, région d’Aloatra Mangoro, Madagascar.
In article      
 
[9]  Hmimina, G., Dufrêne, E., Pontailler, J.Y., Delpierre, N., Aubinet, M., Caquet, B., de Grandcourt, A., Burban, B., Flechard, C., Granier, A., Gross, P., Heinesch, B., Longdoz, B., Moureaux, C., Ourcival, J. M., Rambal, S., Saint André, L., Soudani, K., (2013). Evaluation of the potential of MODIS satellite data to predict vegetation phenology in different biomes: An investigation using ground-based NDVI measurements, Remote Sensing of Environment, Volume 132, 2013, Pages 145-158, ISSN 0034-4257.
In article      View Article
 
[10]  Fensholt, R., Rasmussen, K., Kaspersen, P., Huber, S., Horion, S., Swinnen, E., (2013). Assessing Land Degradation/Recovery in the African Sahel from Long-Term Earth Observation Based Primary Productivity and Precipitation Relationships. Remote Sens. 2013, 5, 664-686.
In article      View Article
 
[11]  Diouf, A.A., Hiernaux, P., Brandt, M., Faye, G., Djaby, B., Diop, M.B., Ndione, J.A. and Tychon, B., (2016). Do Agrometeorological Data Improve Optical Satellite-Based Estimations of Herbaceous Yield in Sahelian Semi-Arid Ecosystems? Remote Sensing, 8, 668. http://www.mdpi.com/journal/remotesensing.
In article      View Article
 
[12]  Dia, I.M.M., (2003). Elaboration et mise en œuvre d'un plan de gestion intégrée - La Réserve de biosphère du delta du Saloum, Sénégal. UICN, Gland, Suisse et Cambridge, Royaume-Uni. xiv + 130 pp.
In article      
 
[13]  Diouf, P.S., (1996). Les peuplements de poissons des milieux estuariens de l’Afrique de l’Ouest: l’exemple de l’estuaire hyperhalin du Sine-Saloum. Thèse de doctorat, Univ. Montpellier II, 267 p.
In article      
 
[14]  Cormier-Salem, M.C., (1999). Les Rivières du Sud. Sociétés et mangroves ouest-africaines. Paris, Orstom, 2 volumes (Volume 1: 416 pp.; Volume 2: 288 pp).
In article      View Article
 
[15]  Le Barbé, L., Lebel, T., Tapsoba, D., (2002). Rainfall variability in West Africa during the years 1950-1990, J. Climate, 15(2), 187-202.
In article      View Article
 
[16]  GIEC., (2007). Rapport final du Groupe Intergouvernemental sur l’Evolution du Climat 2007.
In article      
 
[17]  IPCC., (2007). Intergovernmental Panel on Climate Change. Bilan 2007 des changements climatiques : Rapport de synthèse, 113 pages.
In article      View Article
 
[18]  Funk, C., Peterson, P., Landsfeld, M., Pedreros, D., Verdin, J., Shukla, S., Husak, G., Rowland, J., Harrison, L., Hoell, A., (2015). The Climate Hazards Infrared Precipitation with Stations -a New Environmental Record for Monitoring Extremes. Sci. Data 2015, 2, 150066.
In article      View Article  PubMed
 
[19]  Masoud, A.A., Koike, K., (2006). Arid land salinization detected by remotely-sensed landcover changes: A case study in the Siwa region, NW Egypt. Journal of Arid Environments, Volume 66, Issue 1, July 2006, Pages 151-167 2005.
In article      View Article
 
[20]  Liu, Y., Wang, Y.,and Zhang, J., (2012). “New machine learning algorithm: Random Forest” in International Conference on Information Computing and Applications, 2012, pp. 246-252.
In article      View Article
 
[21]  De Wispeleare, G., (1990). Dynamique de la désertification au Sahel du Burkina Faso: Cartographie de l’évolution et de recherches méthodologiques sur les applications de la télédétection, Thèse d’ingénieur, CIRAD-EMVT, France, 546 p.
In article      
 
[22]  Dièye, E.B., Diaw, A.T., Diatta, C.S., De Wispealare, G., (2008). Evolution spatiale de la mangrove de l’estuaire du Saloum (Sénégal) entre 1972 et 1999 : approche méthodologique par Télédétection», Journal des Sciences et Technologies 2008, vol. 6 No. 1, Faculté des Sciences et Techniques, UCAD, Dakar, 36-48.
In article      
 
[23]  Bernier, B., (1992). Introduction à la macroéconomie. Dunod, Paris, 217p.
In article      
 
[24]  Ndour, M.M.M., Ndonky, A., Sarr, C.A.T., Ndiaye, M., (2020). Dynamique de l’occupation du sol 1986-2016 et géprospective de l’évolution urbaine de la région de Dakar (Sénégal) à l’horizon 2035. Revue de géographie du laboratoire Leïdi_ISSN 0851-2515, N°24 Décembre 2020.
In article      
 
[25]  Kogan, F. N. 1990. Remote sensing of weather impacts on vegetation in non-homogeneous areas. International Journal of Remote Sensing 11: 1405-1419.
In article      View Article
 
[26]  Kogan, F.N., (1997). Global drought watch from space. Bulletin of the American Meteorological Society, vol. 78, n°4, p.621-636.
In article      View Article
 
[27]  Owrangi, M.A., Adamowski, J., Rahnemaei, M., Mohammadzadeh, A., Sharifan, R.A. (2011). Drought Monitoring Methodology Based on AVHRR Images and SPOT Vegetation Maps. Journal of Water Resource and Protection, 3, p. 325-334.
In article      View Article
 
[28]  Mickee, T.B., Doesken N.J., et Kleist J. (1993) The relationship of drought frequency and duration to time scale. Actes de la 8th Conference on Applied Climatology (Anaheim, Californie), p. 179-184.
In article      
 
[29]  Dai, A., Lamb, P.J., Trenberth, K.E., Hulme, M., Jones, P.D., Xie, P., (2004). The recent Sahel drought is real. Int. J. Climatol., 24, 1323-1331.
In article      View Article
 
[30]  Nicholson, S., (2005). On the question of the “recovery” of the rains in the West African Sahel. J. Arid Environ., 63: 615-641.
In article      View Article
 
[31]  Dièye, E.B., Diaw, A.T., Sané, T., Ndour, N., (2013). Dynamique de la mangrove de l’estuaire du Saloum (Sénégal) entre 1972 et 2010», Cybergeo: European Journal of Geography, Environnement, Nature, Paysage, document 629, mis en ligne le 09 janvier 2013, consulté le 07 octobre 2021. URL: http://journals.openedition.org/cybergeo/25671.
In article      
 
[32]  Marius, C., (1984). Contribution à l’étude des mangroves du Sénégal et de la Gambie-Ecologie-Pédologie-Géochimie. Mise en valeur et aménagement, thèse de doctorat de troisième cycle, ORSTOM, Paris, 309p.
In article      
 
[33]  FAYE, S., (2017). Les enjeux d’une gouvernance de l’estuaire du Saloum dans la perspective d’une préservation durable des patrimoines de la Réserve de la Biosphère du Delta du Saloum (Sénégal), zone d’interface homme-nature en dégradation, dans un contexte de réchauffement climatique », These de doctorat de l’Université de Jean Moulin de Saint-Etienne. https://tel.archives-ouvertes.fr/tel-02174828.
In article      
 
[34]  Dièye, E.B., Diaw, A.T., Sané, T., Sy, O., Dioh, P., (2011). Changement climatique et évolution de la mangrove dans la lagune de Joal-Fadiouth (Sénégal). In Actes du 24ème Colloque de l’AIC, Rovereto (Italie), pp183-188.
In article      
 
[35]  Niang, I., (1998). Etude de vulnérabilité des zones côtières sénégalaises aux changements climatiques: le cas des pays africains côtiers, Bull. Africains, No.10, Dakar, 25-37.
In article      
 
[36]  Diop, E.S., (1998). Contribution à l’élaboration du plan de gestion intégrée de la Réserve de la Biosphère du Delta de Saloum (Sénégal)», Dakar, UCAD-UNESCO-MAB, 86 p.
In article