Advertisement

Agroforestry Systems

, Volume 80, Issue 2, pp 191–201 | Cite as

Identifying the global potential for baobab tree cultivation using ecological niche modelling

  • Aida Cuni Sanchez
  • Patrick E. Osborne
  • Nazmul Haq
Article

Abstract

The benefits provided by underutilised fruit tree species such as baobab (Adansonia digitata L.) in combating increasing malnutrition and poverty become more apparent as awareness grows regarding concerns of climate change and food security. Due to its multiple uses, its high nutritional and medicinal value, drought tolerance and relatively easy cultivation, baobab has been identified as one of the most important edible forest trees to be conserved, domesticated and valued in Africa. In order to contribute towards the cultivation of the species, suitability of sites in Africa and worldwide was evaluated for potential cultivation using species’ locality data and spatial environmental data in MAXENT modelling framework. A total of 450 geo-referenced records of the baobab tree were assembled from herbarium records, commercial firm’s databases and fieldwork for modelling site suitability for global cultivation of the baobab tree. Climatic and topographic data were acquired from the Worldclim data while soil data was obtained from the Harmonized World Soil Database. MAXENT was found to be a successful modelling method for studying cultivation potential. The main variables that contributed towards predicting baobab’s global cultivation potential were annual precipitation and temperature seasonality. Results suggest that baobab tree could be widely cultivated in most countries in southern Africa and in the Sudano-Sahelian zone of West Africa from Senegal to Sudan. Angola and Somalia were found to be highly suitable for cultivating baobab in Africa. Model suggests, India, where the baobab tree already exists and is used, to be the most suitable country for baobab cultivation outside Africa. North-west Australia, Madagascar, north-east Brazil and Mexico resulted to be other suitable places for cultivating the tree species. Although it is recommended model results be validated with in situ seedling experiments, there seems to be a great potential for the cultivation of this species globally.

Keywords

Baobab tree Distribution Africa Global cultivation potential Modelling Maxent 

Introduction

The fruit pulp, leaves and seeds of the baobab tree (Adansonia digitata L., Family Malvaceae) are a good source of vitamins and minerals (Smith et al. 1996) and are common ingredients in traditional dishes in rural and urban Africa (Nordeide et al. 1996). Apart from food, baobabs supply medicine, livestock fodder, fibre, clothing, material for hunting and fishing, and shade to local people (Wickens 1982; Sidibé and Williams 2002). The tree also acts as an important source of food, water (during times of drought) and shelter for a wide range of animals (Fenner 1980). The species improves site conditions in the savannah by adding organic matter and nutrients through leaf-fall and by reducing soil temperatures and water loss due to evapotranspiration (Amundson et al. 1995). While trade in baobab products has long been an important source of income for people in Africa (Chicamai et al. 2004), baobab fruit pulp has recently been accepted as a novel food ingredient in the EU (CEC 2008) and in the US market (FDA 2009). Owing to the combination of health claims (e.g. pre-biotic and antioxidant functions, high calcium content) and food technological functions (high pectin and fiber content), baobab fruit pulp has been identified as a suitable candidate for a new generation of functional foods and drinks (Gruenwald and Galizia 2005). Baobab fruit pulp, which is traditionally used against diarrhoea, scurvy, cough, dysentery, small pox, measles and even malaria (Dweck 1997), has also been recognized as a botanical remedy and, because of its antioxidant effect, as a good raw material for natural cosmetic products (Gruenwald and Galizia 2005).

Considering the trend for consumers to prefer functional foods and drinks, exotic flavors and natural healthcare products, PhytoTrade Africa (the trade organization that submitted EU and USA proposal of baobab fruit pulp as food ingredient) believes that baobab fruit pulp has the potential to spawn a billion-dollar industry. Indeed, a 2007 report by the UK’s Natural Resources Institute estimated that baobab has the potential to be a billion dollar industry for Africa and could employ over 2.5 million households. In India, where the tree was introduced by Arab traders (Wickens and Lowe 2008), the potential of the baobab to be a billion dollar tree and to change the lives of rural people, has also been discussed (Sekhar 2008).

At present, there are mainly two baobab fruit pulp producers from two countries in Africa. In Senegal, baobab fruit pulp is processed by the Baobab Fruit Company Senegal, a subsidiary of the Italian Baobab Fruit Company (an Italian family firm). In Malawi, baobab fruit pulp is processed by Tree Crops Ltd., a member of PhytoTrade Africa (a trade association of the natural products derived from plants indigenous to Southern Africa). As the market is expected to grow, production increases have already been estimated. In Malawi, production could increase from 21,840 mt/year to 81,900 mt/year and in southern Africa in general from 190,104 mt/year to 712,890 mt/year (Regional Trade Facilitation Program 2009).

Despite these calculations, there is a growing concern in the popular press that the new interest in baobab fruit pulp might cause over-exploitation or misuse of baobab trees. For example, in eastern Zimbabwe, where baobab bark is harvested for craft purposes, Dovie (2003) stated that the baobabs are in danger of destruction in the short term as a result of harvesting and trade arrangements. Apart from over-exploitation by humans, a lack of natural regeneration of the baobab tree (Romero et al. 2001; Assogbadjo et al. 2005; Duvall 2007), elephant damage (Edkins et al. 2007), droughts and land clearance for mining, dams and construction (Wickens and Lowe 2008) are other threats to baobab tree populations, which stress the need for conservation strategies.

Another growing concern in the popular press is that baobab products might be sold at such a high price that local people in Africa might not be able to afford them. There might be fewer fruits available for harvesting, and local people could lose an important part of their diet and an essential pharmaceutical resource (e.g. Starin 2009).

In order to meet the demands of the new commercial markets without compromising the future of the existing trees and the availability of baobab products for local people in Africa, one solution is cultivation. Cultivation can increase the harvested volume, but also it can ensure reliability and quality of supply, aspects identified to be key factor determining the long-term viability of a given product in international trade (Chikamai et al. 2009). Domestication and cultivation also reduces the length of production while preserving the characteristics that are important to consumers (Chikamai et al. 2009). Gebauer et al. (2002) suggested that greater attention should be given to increased overall production of baobab via improvement through breeding and germplasm conservation.

Cultivation could also help raise awareness of the species’ value and promote conservation of the baobab tree and the ecosystems where it lives. The baobab itself represents an important tourist draw for several countries (Scales 2007), and it supports rare species, such as the seriously endangered black rhino (Diceros bicornis) that eats (among other things) baobab pods (Coe 2005, personal communication). Baobab is also the main host for Tapinanthus malacophyllus, an endemic mistletoe of the Luanda region in north-western Angola (Wickens and Lowe 2008).

Although baobabs are not commonly cultivated, they can be successfully propagated in a nursery from seeds and through grafting, stem cutting and air layering (Sidibé and Williams 2002; Assogbadjo et al. 2009). The baobab tree, like other parkland species, has a slow growth rate, a long juvenile phase before fructification, and variable productivity which discourages farmers from growing them (FAO 1999). However, the potential monetary value of baobab fruit pulp in the international market and the rise of awareness of its nutritional value could stimulate the willingness of locals for planting this species.

Investigating potential sites for cultivation is therefore an important first step towards the cultivation of the species and the sustainable commercialisation of the baobab tree. In this paper, geo-referenced baobab locality data and spatial environmental data were used in MAXENT, the species distribution modelling framework to predict potential sites for cultivation in Africa and globally.

Methodology

Species data

A total of 450 baobab growing localities (without duplicates) were assembled from diverse sources (Table 1). About 32% of the localities came from recent fieldwork while 68% were herbarium records. Some of the herbarium records contained geo-referenced coordinates representing presence locations but others had to be geo-referenced using the gazetteers of the Flora Zambesiaca (Pope and Pope 1998), the Flora of Tropical East Africa (Polhill 1988), the Geographic Names Data Base (GNS - National Geospatial-Intelligence Agency 2005) and Google Earth (Google 2008). In order to reduce the effect of temporal bias and to match the species data with the environmental climatic data (produced over the period 1950–2000), herbarium records collected before 1950 were eliminated. Herbarium records classified as “cultivated specimen” by the Herbarium or with controversial “cultivation” origin by Sidibé and Williams (2002) were eliminated from the species dataset.
Table 1

Source, number and geographic location of baobab presence records used in the study

Source

Number of records

Type of record

Geographical location

A. Cuni Sanchez

23

Field work

Benin, Malawi, Mozambique

A.S. Larsen

21

Field work

Several countries all over Africa

Arhus herbariuma

2

Herbarium record

Senegal, Tanzania

Botanic Garden and Botanical Museum Berlin-Dahlema

4

Herbarium record

Mali, Tanzania, Kenya

DADOBAT Project

20

Field work

Senegal, Mali

Database Schema for UC Davisa

1

Herbarium record

Niger

Dhillion and Gustad (2004), Duvall (2007)

2

Field work

Mali

Frankfurt Herbarium

23

Herbarium record

Burkina Faso, Benin, Nigeria

KEW Herbarium

48

Herbarium record

Several countries all over Africa

Marine Science Institute, UCSBa

1

Herbarium record

Tanzania

Missouri Herbariuma

7

Herbarium record

Tanzania

Paris Herbarium

20

Herbarium record

Several countries all over Africa

Phytotrade Africa database

58

Herbarium record and field work

Malawi, Zambia, Mozambique, Zimbabwe

Pock Tsy et al. (2009)

51

Field work

Several countries all over Africa

PRECIS database

40

Herbarium record

South Africa, Namibia, Botswana

Uppsala Herbariuma

2

Herbarium record

Kenya, Eritrea

Wagningen Herbariuma

1

Herbarium record

Cameroun

Wickens and Lowe (2008)

126

Herbarium record

Several countries all over Africa

Total (without duplicates)

450

  

aIndicates that the occurrence data was accessed through GBIF Data Portal (www.gbif.net)

Although there is only one species of baobab tree growing in Africa (A. digitata) and there are no subspecies or varieties officially accepted, a recent study has shown that there are genetic differences between populations from West and East Africa (Pock Tsy et al. 2009). Thus, in this study, potential sites for baobab cultivation were modelled using all 450 presence records, and using the East (307) and West African (143) records separately based on Pock Tsy et al. (2009) (Fig.1). These are referred to as the “All Records model”, the “East African model” and the “West African model”, respectively.
Fig. 1

Geographical distribution of the baobab presence records used in the study. Triangles: West Africa records, squares: East Africa records

Environmental data

Growing potential was assessed by characterising the climate and soils at the baobab presence locations. Twenty spatial datasets from the WorldClim database (at 5′ resolution) (Hijmans et al. 2004, 2005) and one from the Harmonized World Soil Database (30″ resolution) (FAO et al. 2008) were used. The Worldclim dataset included altitude and 19 bioclimatic variables derived from temperature and rainfall. Although 21 spatial datasets were used in the beginning, some were eliminated during modelling due to their low contribution to the final model. The objective of the study was to determine potential sites for future cultivation of the baobab tree. In agricultural systems, abiotic factors are more limiting than biotic factors (such as competition or dispersal); therefore no biotic factors were included in this study. Although several studies have correlated baobab distribution with human settlements (Assogbadjo et al. 2005; Duvall 2007), it was assumed that at a continental scale, climatic variables were much more limiting than human population.

Species distribution modelling

The baobab’s global cultivation potential was modelled using Maxent (version 3.2), a general-purpose algorithm that uses presence-only data to model species’ distributions (Phillips et al. 2006). Maxent predicts the potential species distribution by estimating the probability distribution of maximum entropy across a specified region, subject to a set of constraints that represent the missing information (lack of absence data) about the target distribution (Phillips et al. 2006). The Maxent method is currently considered to be the most accurate approach to modelling presence-only data (Elith et al. 2006; Pearson et al. 2007).

Model validation

Model performance was evaluated using several methods. First, model performance was determined by assigning a subset of the presence records for training and using the remaining records to test the resulting model. A good model should predict correctly the presence of the baobab tree in the test locations. As performance can vary depending upon the particular set of data selected for building the model and for testing it, 10 random partitions of the presence records were made to assess the average behaviour of Maxent, following Phillips et al. (2006). Each partition was created by randomly selecting 75% of the total presence records to build the model and the remaining 25% of presence records were used for testing. However, the full set of presence records were used to build the final model to obtain the best estimate of the species distribution (for all records model, West Africa model and East Africa model). Second, receiver operating characteristic (ROC) analysis was used to evaluate how well the Maxent model compared to a random prediction. The area under the ROC curve (AUC) serves as a measure of model performance in terms of sensitivity versus specificity. The sensitivity for a particular threshold is the fraction of all positive instances that are classified as present and specificity is the fraction of all negative instances that are classified as not present. The value of the AUC is typically between 0.5 (random) and 1.0. The closer the AUC value to 1, the better the model performance. Moreover, the success of the model was also evaluated by visually examining how well the mapped probability values matched the presence records. A good model should produce regions of high probability that cover the majority of presence records and areas of low probability should contain few or no presence points. Additionally, literature was used to determine if high probability areas identified in the model corresponded with areas known to have baobabs, even though precise locations were not available for model building.

Results

Potential distribution of the baobab tree

For the All Records model, the resulting Maxent cultivation potential map for Africa predicted the baobab tree to occur in most of the Sahel and in much of the mopane savannah in southern Africa, in south-east Somalia and eastern Angola (Fig. 2). Outside Africa, the model showed a strong prediction through most of southern India (except the west coast), in south-west Madagascar, in northern Australia, in north-east Brazil and on the east coast of Mexico (Fig. 3).
Fig. 2

Cultivation potential of the baobab tree in Africa. Top: all records model, bottom left: West Africa model, bottom right: East Africa model. Black: high suitability (>70% probability), grey: medium suitability (between 40 and 70% probability), white: low suitability (<40% probability)

Fig. 3

Global cultivation potential of the baobab tree. From top to bottom, all records model, West Africa model, East Africa model. Black: high suitability (>70% probability), grey: medium suitability (between 40 and 70% probability), white: low suitability (<40% probability)

For the West Africa model, the cultivation potential map predicted the baobab tree to grow in most of the Sahel from Senegal to south-west Chad, and in southern Ghana-Nigeria (Fig. 2). For the East Africa model, the baobab was predicted to grow in the mopane savannah in southern Africa from Kenya to South-Africa/Namibia, in south Somalia, Ethiopia, Eritrea, Soudan and north-east Nigeria (Fig. 2). Outside Africa, the West Africa model showed a strong prediction only in Madagascar and Australia, while East Africa model showed a strong prediction in most of the places predicted by the All Records model and some additional sites in south-east Asia, around Paraguay and the Gulf of Mexico (Fig. 3).

Model performance

All the ten generated training/testing models (for All records model, West Africa model and East Africa model) showed a high level of performance when compared to random (where the AUC would be 0.5). Training AUC values ranged from 0.87 to 0.91, from 0.952 to 0.966, and from 0.928 to 0.933 for All records model, West Africa model and East Africa model respectively. Test AUC values were lower but close to training AUC. The training/testing models correctly predicted most of the test locations in all models.

The final All Records model built using all the data had an AUC of 0.879 while the West African and East African model had AUC values of 0.963 and 0.933 respectively. The predicted distributions showed good agreement with qualitative data from the literature. For example, in Angola, the baobab is present in Luanda, Cuanza Norte, Malange and Namibe districts (Figueiredo and Smith 2008) and the All Records model predicted occurrence in these areas. The All Records model also correctly predicted presence in India (where the baobab tree was introduced in the past), Madagascar and Australia (where other species of Adansonia naturally occur).

Variable contribution

The All Records model indicated that the presence of the baobab was mainly associated with temperature seasonality (29.1%), annual precipitation (20.9%) and precipitation of the wettest four months (15.5%). The West African model indicated that baobab presence was mainly correlated with mean temperature of the warmest four months (37.4%), altitude (19%), and precipitation of the wettest four months (11.9%), while the East African model indicated mean temperature of the coldest four months (17.5%), temperature seasonality (15.2%), annual precipitation (13.4%) and precipitation of the wettest four months (9%) to be the most important predictors. Maxent’s jackknife test of variable importance also suggested that the variables which contributed the most to the model (for example, temperature seasonality and annual precipitation for all records model) were good predictors as they had the most information not contained in other variables and they could describe the baobab tree’s distribution on their own (without the other variables).

Maxent’s response curves for the All Records model indicated that the baobab tree occurs in areas of low annual precipitation (between 200 and 1,400 mm), low altitude (below 1,200 m), low temperature seasonality (between 25 and 35°C) but on a broad range of soil types. When comparing the West African and East African response curves, although curves for annual precipitation, temperature seasonality and soil type were similar, large differences were observed for other variables. For example, West African baobabs occur in areas with 0–400 m, 26–34°C of mean temperature of the warmest four months and 22–26°C of mean temperature of the coldest four months, while East Africa baobabs occur in areas with 500–100 m, 200 m, 20–31°C and 14–22°C, respectively.

Geographic origin of planting material

Large differences in potential cultivation sites outside Africa were predicted when the West and East African populations of baobab were modelled separately (Figs. 3, 4). On the basis of this analysis, it is suggested that West African planting material should be used in north-west Australia and the west coast of Madagascar, while East African planting material should be used in India, America, southern Madagascar and north-east Australia.
Fig. 4

From left to right: Cultivation potential of the West African baobab tree population (Australia and Madagascar) and cultivation potential of the East African baobab tree population (Australia and Madagascar). Black: high suitability (>70% probability), grey: medium suitability (between 40 and 70% probability), white: low suitability (<40% probability)

Discussion

Model performance was good, for predicting suitable conditions in test location, for AUC values and in relation to predictions in areas where no records were used to build the model but are known to have the baobab tree (such as Angola). Outside Africa, the fact that the potential cultivation sites include the known occurrence of baobab in India and the known occurrence of closely related species in Australia (Adansonia gregorii) and Madagascar (Adansonia sp.) positively validates the model results.

The modelled response of the baobab to different environmental variables agrees with the ecological requirements suggested in the literature by Sidibé and Williams (2002) and Wickens and Lowe (2008). Although precipitation of the driest month or maximum temperature of the hottest month could seem to be the limiting factors for the baobab tree as it is found in the driest parts of the savannah, this was not the case. Instead, modelling indicated that the presence of baobab tree is mainly related to annual precipitation and temperature seasonality. As the baobab’s distribution was found to be mainly correlated with annual precipitation, it is possible that it could also be cultivated in areas where there is a little annual precipitation but water for irrigation is available. However, further research is needed to confirm this hypothesis. The fact that the baobab tree was found on a broad range of soils [agreeing with Sidibé and Williams (2002) and Wickens and Lowe (2008)] also has implications for cultivation: the broader the range of soil types the baobab tree tolerates, the more the possibilities for cultivation.

Differences between the West African and East African models in terms of the main variables and baobab response curves probably reflect differences in the environment where the baobab tree lives. For example, in the Sahel (West Africa) high temperatures (reflected in the variable mean temperature of the warmest 4 months) and little rain during the only rainy season (reflected in precipitation of the wettest 4 months) might be more limiting here than in East Africa. Differences in response curves, such as altitude, might also reflect differences in the environment, with East Africa having more areas with high altitudes than West Africa. However, it should be noted that although differences in response curves might only be reflecting differences in the environment, they could also reflect different ecological requirements of the West African and the East African baobab populations, which have genetic differences (Pock Tsy et al. 2009). Although further studies are required to confirm differences in ecological requirements between West and East African baobab populations, we consider that these potential differences should be taken into account, especially when choosing planting material.

In general, Maxent modelling suggests that the baobab tree has great potential for cultivation in Africa and in other countries of the world. In West Africa, it could be grown throughout most of the Sahel, but also further south in the Sudanian zone from northern Ghana to Northern Cameroon. In these areas, where locals already know and use the species, cultivation might be easier than elsewhere due to cultural acceptance. The cultivation of this species could also reduce the pressure on existing baobab trees which are threatened by overexploitation, bush fires and grazing (Sidibé and Williams 2002; Assogbadjo et al. 2005; Wickens and Lowe 2008). In East and Southern Africa, most countries that already export baobab fruit products (Tanzania, Zimbabwe, Mozambique, Malawi and South Africa) unsurprisingly have highly suitable conditions for baobab cultivation. Although baobabs are not as widely used there as in West Africa (Jama et al. 2008), the fact that baobab products can be commercialized internationally might motivate farmers to increase production.

Outside Africa, based on current climate, India appears to have the greatest potential for growing the baobab tree. The fact that the tree is already fairly widely distributed throughout parts of India (Wickens and Lowe 2008) and is already used for local consumption and medicine (Vaid and Vaid 1978) might encourage local farmers. The baobab tree, which has been identified as amongst the traditional African fruits whose cultivation and use may help malnutrition problems in Africa (NRC 2006), could help in combating malnutrition in India. Apart from local consumption, the cultivation of the baobab tree could also be aimed at international commercialisation, which could help reduce poverty. Baobab cultivation could also render fertility to the existing baobab trees in India which are thought to be self-incompatible (Wickens and Lowe 2008). Moreover, the cultivation and promotion of the baobab tree could help raise awareness of the multiple uses of these trees and their historical and cultural significance, and perhaps avoid the destruction of the remaining old specimens found in the country. For example, in 2004 a huge baobab found in Mombai was destroyed because it was disturbing a public development project (TNN 2004). It has been reported in the literature that the baobabs already existing in India were brought from East Africa by Arab traders (Wickens and Lowe 2008). Modelling results from this study suggest that planting material for India should be taken from East Africa.

In Australia, the cultivation of the baobab tree could be aimed at international commercialisation. Although one species of baobab already exists in Australia, the African baobab has a higher nutritional value than the former (Miller et al. 1993) and its fruits are generally bigger. A few African baobabs have already been planted in Australia in botanical gardens and in an Aboriginal settlement in north-western Australia (Wickens and Lowe 2008). Modelling results suggest that planting material from West Africa and East Africa should be used for cultivation in north-western Australia and north-eastern Australia respectively.

In Madagascar, the cultivation of the African baobab is recommended only for commercial purposes. The two reasons in support of this recommendation are: one, in Madagascar, locals do not use the baobab tree as much as in mainland Africa (Wickens and Lowe 2008); and second, the local Malagasy species (especially A. rubrostipa) has been reported to have higher nutritional value than the African baobab in terms of leaf vitamin and crude protein contents (Maranz et al. 2007). As there are no chemical studies on Malagasy baobab fruit nutritional value, it could be possible that the fruits of the local species are also more nutritious. The model suggests if the cultivation of the African baobab is to be considered in the dry deciduous forests of western Madagascar, planting material should be taken from West Africa, but if the potential cultivation site is the southern part of Madagascar, East African are recommended.

In the Americas, the Maxent model suggests highly suitable environment in Mexico and in north-east Brazil for cultivation of the baobab (East African planting material). In fact, Dutch or Portuguese travellers are believed to have introduced the tree to northern Brazil, where a huge specimen is still growing in Recife (Wickens and Lowe 2008). Although the baobab tree could be cultivated for commercial purposes in Brazil and Mexico (for example, they could supply the potential US market), it is not commonly found in these two countries, which implies that the local population does not know the species’ nutritional, medicinal and economic values. A multi-stakeholder approach would be needed in these areas to establish baobab as a commercial crop. Another problem for baobab tree cultivation in these two countries is the fact that no mature trees are available for grafting. Grafting significantly reduces the fruiting period of the young baobabs from 8–23 years to 3 years (Sidibé et al.1996; Sidibé and Williams 2002).

In spite of its immense global potential cultivation possibilities, the difficulty of persuading local communities that baobabs can be successfully propagated in a nursery and that its long maturation period can be reduced remains to be the main issue of its cultivation (Wickens and Lowe 2008). Regardless of the country, in order to cultivate the baobab tree, training workshops for imparting knowledge transfer of seed propagation and grafting techniques would be needed.

In situ seedling experiments, which could validate the potential cultivation of baobab tree in new areas, will also be required. Although trees can be assumed to grow well in a determined area with favourable environmental conditions (considering highly suitable habitat as potential areas for cultivation from our model), it can not be assumed that the trees will produce fruits and/or a high yield. For example, it has been noted for other under-utilised species such as tamarind (Tamarindus indica L.) that if the dry season is not long enough, the quality of the fruits is not good.

Conclusion

MAXENT was found to be a successful predictive modelling tool for studying the potential global cultivation of baobab. There is great potential for baobab cultivation in Africa and outside. Although more research is needed to confirm some of the potential cultivation sites and the preference of East or West Africa planting material, policy makers of several countries could include this species as a “recommended” crop due to its high nutritious and medicinal value, its multiple uses and relatively easy cultivation. The maps produced here based on passport data are expected to be useful tools for planning cultivation activities.

Notes

Acknowledgments

This research was supported by DADOBAT (Domestication and Development of Baobab and Tamarind)- EU funded project. We express our gratitude to the support given by A. Overgaard and all baobab locality data providers. We are grateful to G. E. Wickens for his advice and to anonymous reviewers for useful comments on the manuscript.

References

  1. Amundson RG, All AR, Belsky AJ (1995) Stomatal responsiveness to changing light intensity increases rain use efficiency of below-tree-crown vegetation in tropical savannas. J Arid Environ 29(2):139–153CrossRefGoogle Scholar
  2. Assogbadjo AE, Sinsin B, Codjia JTC, Van Damme P (2005) Ecological diversity and pulp, seed and kernel production of the baobab (Adansonia digitata) in Benin. Belg J Bot 138:47–56Google Scholar
  3. Assogbadjo AE, Sinsin B, De Caluwe E, Van Damme P (2009) Développement et domestication du baobab au Bénin. LEA-FSA-UAC/DADOBAT, CotonouGoogle Scholar
  4. CEC (2008) Commission decision 27/June/2008. Off J Eur Union. 11/7/2008, http://www.food.gov.uk/multimedia/pdfs/commdec2008575ec.pdf [last accessed on 21/12/09]
  5. Chicamai B, Eyog-Matig O, Mbogga M (2004) Review and appraisal on the status of indigenous fruits in Eastern Africa. IPGRI-SAFORGEN reportGoogle Scholar
  6. Chikamai B, Tchatat M, Tieguhong JC, Ousseynou N (2009) Forest management for non-wood forest products and services in Sub-Saharan Africa, lessons learnt on sustainable forest management in Africa, discovery and innovation, SFM special edition no. 1Google Scholar
  7. Dhillion SS, Gustad G (2004) Local management practices influence the viability of the baobab (Adansonia digitata Linn.) in different land use types, Cinzana, Mali. J Agric Ecosyst Environ 101:85–103CrossRefGoogle Scholar
  8. Dovie DBK (2003) Rural economy and livelihoods from the non-timber forest products trade. Compromising sustainability in southern Africa? Int J Sustain Dev World Ecol 10(3):247–262CrossRefGoogle Scholar
  9. Duvall CS (2007) Human settlement and baobab distribution in south-western Mali. J Biogeogr 34:1947–1961CrossRefGoogle Scholar
  10. Dweck AC (1997) Ethnobotanical use of plants, part 2: African plants. Cosmet Toilet 112:47–54Google Scholar
  11. Edkins MT, Kruger LM, Harris K, Midgley JJ (2007) Baobabs and elephants in Kruger National Park: nowhere to hide. Afr J Ecol 46(2):119–125CrossRefGoogle Scholar
  12. Elith J, Graham CH, Anderson RP, Dudík M, Ferrier S, Guisan A, Hijmans RJ, Huettmann F, Leathwick JR, Lehmann A, Li J, Lohmann LG, Loiselle BA, Manion G, Moritz C, Nakamura M, Nakazawa Y, Overton JMM, Peterson AT, Phillips SJ, Richardson K, Scachetti-Pereira R, Schapire RE, Soberón J, Williams S, Wisz MS, Zimmermann NE (2006) Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29(2):129–151CrossRefGoogle Scholar
  13. FAO (1999) Agroforestry parklands in sub-Saharan Africa, FAO conservation guide 34. Rome, ItalyGoogle Scholar
  14. FAO/IIASA/ISRIC/ISSCAS/JRC (2008) Harmonized world soil database (version 1.0). FAO, RomeGoogle Scholar
  15. FDA (2009) Agency response letter GRAS notice no. GRN 000273, 25/July/2009. US Food and Drug Administration. http://www.fda.gov/Food/FoodIngredientsPackaging/GenerallyRecognizedasSafeGRAS/GRASListings/ucm174945.htm [last accessed on 21/12/09]
  16. Fenner M (1980) Some measurements on the water relations of baobab trees. Biotropica 12(3):205–209CrossRefGoogle Scholar
  17. Figueiredo E, Smith GF (2008) Plants of Angola/Plantas de Angola. Strelitzia 22, the South African National Biodiversity Institute, PretoriaGoogle Scholar
  18. Gebauer J, El-Sidding K, Ebert G (2002) Baobab (Adansonia digitata L.): a review on a multipurpose tree with promising future in the Sudan. Gartenbauwissenschaft 67(4):155–160Google Scholar
  19. GNS - National Geospatial-inteligence agency (2005) U.S. National Imagery and Mapping Agency’s (NIMA) database of foreign geographic feature names—country filesGoogle Scholar
  20. Gruenwald J, Galizia M (2005) Market brief in the European Union for selected natural ingredients derived from native species, Adansonia digitata L. United Nations Conference on Trade and Development, GenevaGoogle Scholar
  21. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2004) The WorldClim interpolated global terrestrial climate surfaces, vo1 3. (http://biogeo.berkeley.edu/)
  22. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas. Int J Climatol 25:1965–1978CrossRefGoogle Scholar
  23. Jama BA, Mohamed AM, Mulatya J, Njui AN (2008) Comparing the “Big Five”: a framework for the sustainable management of indigenous fruit trees in the drylands of East and Central Africa. Ecol Indic 8(2):170–179CrossRefGoogle Scholar
  24. Maranz S, Niang A, Kalinganire A, Konate D, Kaya B (2007) Potential to harness superior nutritional qualities of exotic baobabs if local adaptation can be conferred through grafting. Agrofor Syst 72:231–239CrossRefGoogle Scholar
  25. Miller BJ, James KW, Maggiore P (1993) Tables of composition of Australian aboriginal foods. Aboriginal Studies Press, CanberraGoogle Scholar
  26. National Research Council (2006) Lost Crops of Africa: volume II: vegetables. Development, security, cooperation policy and global affairs. National Research Council of the National Academies, The national Academies press, Washington, DC, 378 pGoogle Scholar
  27. Nordeide MB, Hatløy A, Følling M, Lied E, Oshaug A (1996) Nutrient composition and nutritional importance of green leaves and wild food resources in an agricultural district, Koutiala, in southern Mali. Intern J Food Sci Nutr 47:455–468CrossRefGoogle Scholar
  28. Pearson RG, Raxworthy CJ, Nakamura M, Peterson AT (2007) Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. J Biogeogr 34:102–117CrossRefGoogle Scholar
  29. Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modelling of species geographic distributions. Ecol Model 190:231–259CrossRefGoogle Scholar
  30. Pock Tsy J-ML, Lumaret R, Mayne D, Vall AOM, Abutaba YIM, Sagna M, Raoseta AOR, Danthu P (2009) Chloroplast DNA phylogeography suggests a West African centre of origin for the baobab, Adansonia digitata L. (Bombacoideae, Malvaceae). Mol Ecol 18:1707–1715CrossRefGoogle Scholar
  31. Polhill D (1988) Flora of Tropical East Africa; index of collecting localities. Royal Botanical Gardens, KewGoogle Scholar
  32. Pope GV, Pope DG (1998) Flora Zambesiaca, collecting localities in the Flora Zambesiaca area. Royal Botanic Gardens, KewGoogle Scholar
  33. Regional Trade Facilitation Program, Trade in natural products: Product information: Baobab (2009) http://www.rtfp.org/media/natural_products_baobab_product_info.pdf [last accessed on 21/12/09]
  34. Romero C, Dovie D, Gambiza J, Luoga E, Schmitt S, Grundy I (2001) Effects of commercial bark harvesting on Adansonia digitata (baobab) in the Save-Odzi valley, Zimbabwe, with considerations for its management. Hotspring Working Group. IES working paper no. 18. Institute of Environmental Studies, Harare, pp 28–50Google Scholar
  35. Scales H (2007) Land of the giants, Geographical March 2007, http://findarticles.com/p/articles/mi_hb3120/is_3_79/ai_n29331217/[last accessed on 21/12/09]
  36. Sekhar MR (2008) Baobab: a billion dollar tree to change the lives of rural India. Rural development of India 29/7/08, http://ruralindia.blogspot.com/2008/07/baobab-tree-to-change-lives-of-rural.html [last accessed on 21/12/09]
  37. Sidibé M, Williams JT (2002) Fruits for the future. Baobab Adansonia digitata. International Centre for Underutilised Crops, University of Southampton, SouthamptonGoogle Scholar
  38. Sidibé M, Scheuring JF, Tembely D, Sidibé MM, Hofman P, Frigg M (1996) Baobab—homegrown vitamin C for Africa. Agrofor Today 8(2):13–15Google Scholar
  39. Smith GC, Clegg MS, Keen CL, Grivett LE (1996) Mineral values of selected plant foods common to Burkina Faso and to Niamey, Niger, West Africa. Int J Food Sci Nutr 47:41–53CrossRefPubMedGoogle Scholar
  40. Starin D (2009) What will happen when the Baobab goes global? The New York Times, 25/5/2009, http://www.nytimes.com/2009/05/26/opinion/26starin.html?_r=2 [last accessed on 21/12/09]
  41. TNN (2004) Urbania eats up ancient baobab, The Times of India, 26/6/2004, http://timesofindia.indiatimes.com/city/mumbai/Urbania-eats-up-ancient-baobab/articleshow/754395.cms [last accessed on 21/12/09]
  42. Vaid KM, Vaid R (1978) Currency paper from Adansonia. Indian J For 1:53–55Google Scholar
  43. Wickens GE (1982) The baobab—Africa’s upside-down tree. Kew Bull 37:173–209CrossRefGoogle Scholar
  44. Wickens GE, Lowe P (2008) The Baobabs, Pachycauls of Africa, Madagascar and Australia. Kluwer, DordrechtCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2010

Authors and Affiliations

  • Aida Cuni Sanchez
    • 1
  • Patrick E. Osborne
    • 2
  • Nazmul Haq
    • 1
  1. 1.Centre for Underutilised CropsUniversity of SouthamptonHighfield, SouthamptonUK
  2. 2.Centre for Environmental Sciences, School of Civil Engineering and the EnvironmentUniversity of SouthamptonHighfield, SouthamptonUK

Personalised recommendations