Potential use of plant species for soil improvement in
Mali and Burkina Faso
“Best bet” Land-use Systems
Thematic reports
Impact of different land uses on biodiversity
Biodiversity and Productivity Assessment for Sustainable
Agroforest Ecosystems
Unique id: IDA1AQCE
Source file: D:\Projects\ASB\ASB Country and Thematic
reports\Above ground biodiversity assessmet WG\MaliRep.xml
Authors: Andrew N. Gillison
--------------------------------------------------------------------------
EXECUTIVE SUMMARY
The Sahelian countries of Mali
and Burkina Faso
are subject to decreasing soil fertility and biodiversity largely as a result
of rising population pressure and uncertain water supply. These are widely
accepted precursors to increased poverty and reduction in quality of life.
Traditional methods of agriculture and animal husbandry are becoming difficult
to sustain, requiring reassessment of methods of land use management in
particular the use of appropriate fallow systems. A brief reconnaissance was undertaken from 6
– 20 February 2000 to assess existing land use conditions and to consider
whether certain exotic fallow species could be screened for possible
introduction in addition to indigenous species. In several village centres the
team examined progress in the establishment of ‘live fences’ enclosing both agricultural subsistence and
animal fodder crops. A subsequent, intensive, four-day reconnaissance was
conducted along a rainfall gradient (500 to 950mm annual) from the Dogon
Plateau to the Neguela area north-west of Bamako.
The survey employed a recently developed rapid-vegetation survey method to
record data from a variety of land use types and natural resource
conditions. These included geocoded
(GPS) site physical features including land use type and site history,
vegetation structure, vascular plant species and plant functional types. Soil
samples (surface 15cm) were collected in each site for subsequent laboratory
analysis. A total of 16 (40x5m) transects were recorded under various land use
types ranging from semi-desertic, combretaceous shrub savanna on skeletal
sandstone in the Dogon plateau to mixed woodland savanna and cultivated Faidherbia and Vitellaria dominated ‘parkland’ on deeper soils. A cluster analysis
of mean canopy height (m), basal area (m2ha-1) total
species and plant functional types and spp:PFT ratios identified three key
groups: woodland savanna tending to be dominated by Combretaceae and Bombax costatum; more open and more
intensively used, sometimes seasonally flooded, woodland savannas with
Combretaceae and Lannea velutina
dominants, often with conspicuous termitaria, and ‘parkland’ dominated by Faidherbia, Vitellaria and Borassus
often with Adansonia (Baobab).
Outliers were depauperate, shrubby woodland savannas on skeletal soils (Dogon
Plateau) and Acacia nilotica dominant
shrublands on poorly drained soils. Apart from cultivation, grazing is
identifiably the most dominant ecological pressure. Reduction of this pressure
would result in a corresponding resurgence of growth in suppressed seedlings of
Acacia, Faidherbia and Vitellaria.
Recent research has shown that certain Asteraceae such as Tithonia diversifolia are capable of improving soil fertility when
used in managed ‘Daisy’ fallows. In several locations (notably peri-urban
Ougadougou and Bamako), Tithonia
diversifolia was observed as an adventive weed. This occurrence, together
with several other exotic Asteraceae,
suggests that with adequate water supply such taxa may be manipulated as
extended fallows to improve soil fertility.
Further research is indicated to ascertain which Asteraceae are best
suited to this region and to particular land uses. Certain Australian arid-land
Acacia and Casuarina species may help improve soil and serve as added fodder
crops. With this in mind, following the survey, seeds of Allocasuarina decasineana, Casuarina
obesa, Acacia cooleyi
var. cooleyi, A. cooleyivar. ileocarpa, A. epacanthaand A. torulosa from different Australian provenances were supplied by
the CSIRO Australian Tree Seed Centre, CSIRO Forestry and Forest Products.
These have been forwarded to the ICRAF office in Bamako
for screening. The DOMAIN spatial
mapping software package was used to generate an environmental ‘domain’ based
on rainfall precipitation and temperature for the sixteen transects. When compared with global data this domain
matched extensive areas in Northern Australia thus
indicating potential for germplasm exchange. Certain species and PFTs for
possible use in improving fodder production and soil conditions are considered
as an initial set for further investigation together with a suggested framework
for biodiversity-related research.
1. Introduction and terms
of reference
The purpose of my task was to help draft a concept paper related to
biodiversity conservation in the Sahel and to
help identify candidate plant species for soil fertility improvement. As background to this task ICRAF has been
instrumental in identifying certain species that have been shown to improve
soil fertility, in particular Tithonia
diversifolia from Mexico
that belongs to the Daisy family (Asteraceae).
Parallel studies in Mexico, Cameroon
and South East Asia (Gillison, 1999; Gillison and
Liswanti, 1999) also indicated that under late fallows dominated by other
Asteraceae (e.g. Chromolaena odorata,
Senecio and Vernonia spp.) soil
fertility also appeared to improve, in some cases with increased
biodiversity. With this in mind a
reconnaissance of land use types and landscapes in Mali
and Burkina Faso
sought to identify conditions where the introduction of certain herbaceous and
woody species might prove beneficial. As well as seeking species to help
improve soil fertility, of equal concern are species that can be used as fodder
for domestic grazing animals. Although not part of the present TOR, this was also kept in mind during the field
visit. An itinerary is listed in Annex II.
Methods
I accompanied the ICRAF team on
a reconnaissance of a range of field conditions where various trials were in progress (Annex I). In
order to gain some preliminary understanding of vegetation response to
different land management practices a series of
40x5 m strip transects were established along gradients of soil types
and rainfall precipitation. Because landscapes in Mali
and Burkina Faso
have relatively little relief overall, minor changes in drainage pattern
translate into significant differences in soil depth and fertility for
indigenous people. A total of 16 such
transects were established ranging from the Dogon plateau in the North East to
the Foret Classeé des Monts Maniague towards Siby west of Bamako and the
Neguela region to the North West of Bamako.
Fourteen of these were recorded in 3.5 days following departure of the
main team. The general survey design followed the gradsect method (Gillison and
Brewer, 1985; Wessels et al., 1998)
that is designed to improve chances of detecting changes in biodiversity
compared with more traditional designs based on probability theory. In each
transect data were recorded according to the proforma technique devised by
Gillison (1988, 1999a,b). Data included
location (Latitude, Longitude in deg. min. sec.) recorded by GPS, slope%,
aspect (deg.), elevation (m) recorded using clinometer, compass and digital
altimeter respectively; soil depth, soil type (where information available),
litter depth 9cm), and terrain position. Surface soil samples (top 15cm) were
taken for each site and forwarded to ICRAF Nairobi for analyses. Vegetation
structure (mean canopy height, crown cover percent, basal area m2ha-1 estimated using the Bitterlich technique,
furcation index (and architectural index based on forking in the main stem of a
woody plant), cover-abundance of woody
plants <2m tall and cover-abundance of bryophytes (both Domin scale
estimates).
Where identifications were
possible, all vascular plant species were recorded together with local names of
key species and Plant Functional Types (PFTs – after the method of Gillison and
Carpenter, 1997). All data were
subsequently entered into a computer using the recently developed Windows-based
‘VegClass’ software (Gillison, unpubl.)
that enables summaries and graphic output of all data including derived
diversity indices for PFTs (Shannon-Wiener, Simpson’s and Fisher’s a). Because the data are summarised for each 5x5m
quadrat within the 40x5m transect it is possible to use VegClass to generate
species:area and PFT:area curves to assess sampling efficiency. Raw data and metadata summaries can be output
to Excel and MS Access. Because the survey method is generic, comparative analyses are possible between
geographically remote locations where, for example, environments and vegetation response may be
similar but where the taxa may differ. In this way the method provides a basis
for identifying species with similar adaptive response characteristics in
different countries and ecoregions. In the present Sahelian study this is
potentially useful in identifying exotic species from other countries that show
parallel adaptive features and vice versa.
Observers were A N. Gillison (CBM) and D-Y. Alexandré (IRD/ICRAF).
The data were analysed using the
CSIRO PATN pattern analysis software (Belbin 1992). Five variables that are known from previous
studies (Gillison and Liswanti, 1999a,b) to be key indicators of animal habitat
and soil nutrient availability were used in the analysis. These were vegetation
structure (mean canopy height (m), basal area (m2ha-1 ),
total vascular plant species, total Plant Functional Types (PFTs) and the ratio
of Species:PFTs. The data were
classified using a Gower metric similarity measure and an unweighted pair group
average fusion strategy. A multi-dimensional scaling (MDS) analysis was applied
to produce a two-vector solution and the scores used to construct a
2-dimensional graph. Diversity indices
based on PFTs rather than taxa (Gillison et
al., 2000) were generated by VegClass. These include Shannon-Winer,
Simpson’s and Fisher’s a diversity indices. An additional measure of PFT
complexity within each plot is indicated by the Plant Functional Complexity
(PFC) measure (Gillison et al.,
2000).
A climatic domain was generated
for the sixteen sites using the DOMAIN potential mapping software package
(Carpenter et al., 1993) based on
temperature and rainfall precipitation minimal and maximal acquired from a
public domain data set. This domain was then compared with the global data set
and a thematic map produced showing differing similarity matches.
Results
Site location and physical data are listed in Table 1, with
land use management regime and vegetation structure in Table 2 and species, PFT
and PFT diversity indices in Table 3. Detailed taxonomic and PFT data are
listed in Annex II.
Classification
The dendrogram in Fig.1
indicates a primary division between ‘parkland’ (variously dominated by Faidherbia albida (Gau), Vitellaria paradoxa (Karité) and Borassus aethiopum with occasional Adansonia digitata (Baobab) (Sites
3,5,6,7,15,16) including two outlier sites (sparse Combretum shrub savanna on skeletal sandstone soils (site 4) and an
Acacia nilotica shrub savanna on
poorly drained soils (site 8). The other side of the dendrogram contains two
main groups: one composed of woodland savannas dominated by Combretaceae and Bombax costatum commonly with Cochlosperum tinctorium and C. planchoni (sites 2,12,13,14) and the
other with more open woodland savanna with mixed species (Acacia macrostachya, Combretaceae, Entada africana, Lannea velutina and Sclerocarya birrea)
commonly with termitaria (sites 9,10,11) and an outlying site (1) that is a Faidherbia /A. nilotica parkland on a cultivated, stony terrrace.
Ordination
The MDS (Fig2) reflects the
classification to a large extent, with the above groups clearly identifiable.
The first vector (X axis) indicates increasing plant biodiversity towards the
more complex woodland savannas in the positive direction while vector 2 (Y
axis) separates tall ‘parkland’ (negative) from the shorter and more sparse
shrub savannas (positive).
Diversity indices
To the extent that vegetation complexity (richness and
composition of taxa and PFTs) can be said to reflectsoverall biodiversity, this
is least in the highly modified ‘parkland’ land use types. A measure of this is
indicated in Table 3 that includes diversity indices based on PFTs rather than
species. The Plant Functional Complexity and Shannon-Wiener values correspond
most closely with obvious ecological interpretations of adaptive complexity and
available ecological niches.
DOMAIN mapping
The map (Fig. 3) shows
remarkable similarities with extensive areas of northern Australia,
especially those areas in North-West Queensland with
similar vegetation formations and with North-West Australia
also with similar vegetation structure and many families including Bombacaceae
(Adansonia, Bombax) and closely related Cochlospermaceae (Cochlospermum). It is of interest that Mimosa pigra of the Sahel and indigenous
to Central and South America has become an aggressive, invasive exotic weed in
the Australian North-West as have Acacia
nilotica and Parkinsonia aculeata
(also tropical America) in the highly
seasonal, outwash plains of the Gulf of Carpentaria in
NW Queensland. Another widespread weed common to both areas is Calotropis gigantea. More restricted
areas with very similar climate domains can be seen in many other countries
including some sub-tropical areas in Africa and in Central and North
America.
4. Discussion
Aside from the influence of rainfall precipitation and
distribution, the major ecological determinants of vegetation are cultivation
and grazing by domestic animals. In all
‘parklands’ first inspection suggests the dominant species are not being
replaced by any juveniles. Closer investigation however, indicates the presence
of many tree seedlings that are suppressed by grazing animals (cattle, goats, sheep).
The removal of grazing pressure would therefore result in the regeneration of
such species The presence of large evergreen trees (e.g. Karité, Gau, Mango) in
an otherwise arid environment indicates root access to artesian water over much
of the area surveyed. Many villagers commented that the water table can be
highly variable within and between years, causing local hardship and a
corresponding changes in plant species composition.
Pattern analysis showed a clear
separation of vegetation types that reflect mainly variation in soil fertility,
in particular run-on versus run-off
sites. With some exceptions (such as swamps, swamp margins and seasonal
floodplains) the former support the productive parklands whereas the latter
support the more depauperate woodland savannas especially on skeletal soils or
landscapes with lateritic crusts. It is expected that results from soil
analyses will contribute further to an understanding of the vegetation
dynamics. The Sahelian landscapes of Mali and Burkina Faso show striking
parallels with the savannas on Proterozoic, Jurassic and Cretaceous sandstones
and Late Miocene, lateritic duricrusts of Northern Australia the main exception
being the lack of naturally occurring Eucalyptus
and Melaleuca species in the otherwise
Combretaceae-dominant Sahel. Many plant families (e.g. Anacardiaceae,
Bombacaceae, Capparidaceae, Cochlospermaceae, Combretaceae, Fabaceae and
Rubiaceae) are found in both regions, often with the same genera (Acacia, Adansonia, Bombax, Cochlospermum, Erythrina,
Gardenia, Grewia, Strychnos, Terminalia …). It is therefore hardly
surprising that the DOMAIN map (Fig. 3) highlights similar climate matches
based on temperature and rainfall precipitation. Other regional matches are also evident in
the seasonal, dry landscapes of Eastern and Southern Africa, including Madagascar,
Saudi Arabia, the Indian sub-continent
and parts of mainland South East Asia and Central and South America. A conspicuous feature of the Mali
woodland savannas is the dominance of Combretaceae and the development of some
extreme life forms that possess woody upper parts but well-developed,
below-ground storage organs (Cochlospermum
spp., Entada africana). (Fig. 4).
These features provide insurance against extreme drought and fire but create
difficulties for classification into specific PFTs as they defy Raunkiaerean
life form classification (Raunkiaer, 1934).
Other than this, the occurrence of ‘parklands’ typifies this part of the
Sahel (Figs, 5,6) as do Baobabs (Fig. 7). Functional
diversity and, possibly biodiversity, are clearly highest in the least modified
vegetation types. The evident resilience of life forms and highly efficient
dispersal mechanisms for seed and fruit are likely to support rehabilitation of
degraded landscapes provided these are within range of source material.
5. Screening trials of
Australian arid zone species
On request from CBM, the
Australian Tree Seed Centre[1], kindly provided seed
from a range of species considered to be potentially useful as either fuel or
fodder crops or soil ameliorants (Casuarinaceae). These are listed in Table 4.
Information supplied by the ATSC includes provenance location and elevation,
phytosanitary certificate, procedures for pre-treatment of seed and seed
viability. Both the seeds and the information have been forwarded by CBM to the
ICRAF office in Mali
for screening trials.
6. Conclusions and
Recommendations
Soil fertility and crop
production seem certain to decline under the currently increasing population
and associated land use pressures. Within the framework of traditional land use
practices that appear to be well entrenched, significant improvements in fallow
management may be possible with introduction of Tithonia and Tithonia-like
species where adequate water harvesting and water management are feasible. Recent developments in small, low-cost,
solar-powered, medium-lift, impeller pumps could be explored for trialling in
several representative sites where artesian water is available. These would
make water more readily available during periods of reduced water levels and
assist in nursery production of species for transplanting into selected trial
cultivation areas. Further research is needed to identify and review likely
species for screening trials given the climate and soil domains in the Sahel. Atriplex
species and similar arid-land ‘salt-bush’ Chenopodiaceae should be investigated
as potential fodder species. Other
potential indigenous and exotic ‘Live-fence’ species should also be explored,
one such indigenous species being Pterocarpus
erinaceus that is common in woodland savannas in Mali.
The climate and soil
similarities between the Sahelian lands of Mali
and Burkina Faso and Northern
Australia point to likely sources of germplasm for species that may be
introduced to the Sahel to assist with enhancing fuel,
fodder and soil improvement. The types of plant species most likely to succeed
should match the more arid climates in order to cope with unforeseen climate
fluctuations such as El Niño events. To this extent species of Acacia, Allocasuarina, Casuarina and Melaleuca might be profitably screened
for introduction. For fodder plants, arid Australian and Asian Chenopodiaceae
should be carefully examined, especially Atriplex
spp. as these are well established
fodder species in arid and semi-arid environments and their management is
relatively well known. Further research is needed to explore likely sources of
Asteraceae in particular Olearia,
Senecio, Tithonia and Vernonia
species.
To better understand grazing effects, exclosures are one
obvious experimental design. Previous experience with exclosures in the Sahel
suggest these should be approached with caution: One early experiment resulted
in sand build up against the exclosure fence that subsequently prevented
surface flow of water into the exclosure creating a localised droughting
effect. Another resulted in a heavy legume seed crop that attracted rodents.
When the seed supply was exhausted the rodents attacked other plants inside and
outside the exclosure resulting in severe damage (D-Y Alexandre, pers. com.).
Human impact in the Sahel
has resulted in highly manipulated ecosystems with mosaics of locally intensive
cultivation and loosely controlled grazing and indeterminate fire control.
These conditions create difficulties in assessing the actual and potential role
of biodiversity in sustaining agricultural productivity and environmental
services. To better understand these
interactions will require a series of intensive baseline studies to be
establshed along a series of predefined biophysical environments under known
land management systems. Results from recent developments in gradient-based,
rapid biodiversity assessment methods in the ICRAF ASB project indicate these
methods would be well suited to designing and implementing a baseline study for
selected areas of the Sahel. Baseline data of this kind are essential in
identifying range distributions of key plant and animals taxa and in
establishing and simulating dynamic linkages between soil nutrient availabilty,
biodiversity and profitability. Provided efficient correlations can be
detected, these can be used to help forecast land use impacts on biodiversity
and profitability (cf. Tomich et al., 2000) and thereby assist in
generating options for adaptive and sustainable management.
7. References
Belbin, L.
(1992). PATN Pattern Analysis Package: Technical Reference. CSIRO Div. Wildlife
and Ecology, Canberra.
Carpenter, G.,
Gillison, A.N. and Winter, J. (1993). DOMAIN: A flexible modelling procedure
for mapping potential distributions of plants and animals. Biodiv. Cons. 2, 667-680.
Gillison, A.N.
(1988). 'A Plant Functional Proforma for
Dynamic Vegetation Studies and Natural Resource Surveys' Tech Rep. 88/3.
CSIRO Division of Water and Land Resources, Canberra.
Gillison, A.N.
(1999a) (coord.), Above-ground biodiversity assessment working group summary
report 1996-99. Impact on biodiversity of different land uses. Alternatives to
slash and burn project. ICRAF. Part A:
Executive summary., pp.2-3 Part B: Above-ground, ecoregional benchmark surveys.
Pp. 4-14.
Gillison, A.N.
and Brewer, K.R.W. (1985). The use of gradient directed transects or gradsects
in natural resource surveys. J. Environ.
Manag. 20: 103-127
Gillison, A.N. and
Carpenter, G. (1997). A generic plant functional attribute set and
grammar for dynamic vegetation
description and analysis. Funct. Ecol. 11, 775-
783.
Gillison, A.N.,
Liswanti, N.L., (Eds.), (1999a). An
intensive biodiversity baseline study in Jambi province, Central Sumatra, Indonesia. Part C In: Gillison, A.N.
(coordinator), Above-ground biodiversity assessment working group summary
report 1996-99. Impact on biodiversity of different land uses. Alternatives to
slash and burn project. ICRAF, Nairobi.
Incl. Maps and Annexes.
Gillison, A.N. and
Liswanti, N. (1999b). Biodiversity and productivity assessment for sustainable
agroforest ecosystems. Mae Chaem, Northern Thailand:
Preliminary Report. Part D in: Gillison, A.N. (1999) (coordinator),
Above-ground biodiversity assessment working group summary report 1996-99.
Impact on biodiversity of different land uses. Alternatives to slash and burn
project. ICRAF, Nairobi. Pp. 38,
plus maps and Annexes.
Gillison, A.N.
(1999b). Biodiversity and productivity assessment for sustainable agroforest
ecosystems. Mbalmayo, Cameroon.
Part E. In Gillison, A.N. (1999) (coordinator), Above-ground biodiversity
assessment working group summary report 1996-99. Impact on biodiversity of
different land uses. Alternatives to slash and burn project. ICRAF, Nairobi.
Pp. 43, plus Annexes.
Gillison, A.N., Liswanti, N., van Noordwijk, M., Suseno, and
Tomich, T. (2000). . Biodiversity and productivity assessment for sustainable
agroforest ecosystems. Lampung, S. Sumatra. Part F. In Gillison, A.N. (1999)
(coordinator), Above-ground biodiversity assessment working group summary
report 1996-99. Impact on biodiversity of different land uses. Alternatives to
slash and burn project. ICRAF, Nairobi.pp ##.
Raunkiaer, C. (1934). The
Life Forms of Plants and Statistical Plant Geography Being the collected papers
of C. Raunkiaer. Oxford
at the Clarendon Press. 632 pp.
Tomich, T.P., van Noordwijk, M., Budidarsono, S., Gillison,
A.N., Kusumanto, T., Mudiyarso, D., Stolle, F.and Fagi, A.M. Kusumanto, T.,
(2000). "Agricultural Intensification,
Deforestation, and the Environment: Assessing Tradeoffs in Sumatra,
Indonesia" In: DR
Lee and CB Barrett, eds, Tradeoffs or
Synergies? Agricultural Intensification, Economic Development and the
Environment.Wallingford, UK:
CAB International, 2000. (in press).
Wessels, K.J., Van Jaarsveld, A.S., Grimbeek, J.D. and Van
der Linde, M.J. (1998). An evaluation of the gradsect biological survey method.
Biodiversity and Conservation 7:
1093-1121.
_____________________________________________________
