ArticlePDF Available

Abstract and Figures

Caesalpinia bonduc is overexploited and threatened due to its importance in medicine. This study aims at assessing on farm seedling productivity of C. bonduc by aid of simulation modelling in order to design its appropriate plantation techniques, harvesting intervals, and soil conditions. Data were collected from nursery and field experiments by measuring stem height, collar diameter, number of leaves and tap root length during 180 days. The simulation model was based on a metabolic pool type model calibrated first to simulate the observed growth data from the nursery (calibration). Following it was used to simulate the growth of plants from field experiments, first by an optimization of the utilization of leaves or roots only, and second by an optimization of the utilization of both leaves and roots at the same time at different plant densities and nitrogen levels. The models show that in order to optimize the utilization of C. bonduc it should be planted at high densities with high nitrogen levels. Leaves and roots harvesting should take place every 50-60 days, and maximum 15% of the biomass of roots and leaves should be harvested at each harvesting event.
Content may be subject to copyright.
Padonou et al.
Page
116
RESEARCH PAPER OPEN ACCESS
Investigations of on farm seedling productivity of the rare and
declining
Caesalpinia bonduc
in Benin (West Africa) by aid of
simulation modelling
EA. Padonou1*, AE. Assogbadjo1, R. Glèlè Kakaï2, AM. Lykke3, B. Sinsin1, J. Axelsen3
1University of Abomey, Calavi, Faculty of Agronomic Sciences,Cotonou, Benin
2
Laboratory of Biomathematics and Forest Estimations, University of Abomey-Calavi, Cotonou, Benin
3Aarhus University, Department of Bioscience, Silkeborg, Denmark
Article published on March 21, 2015
Key words: Caesalpinia bonduc, Simulation model, Harvesting intervals sustainable use, Plantations.
Abstract
Caesalpinia bonduc is overexploited and threatened due to its importance in medicine. This study aims at
assessing on farm seedling productivity of C. bonduc by aid of simulation modelling in order to design its
appropriate plantation techniques, harvesting intervals, and soil conditions. Data were collected from nursery
and field experiments by measuring stem height, collar diameter, number of leaves and tap root length during
180 days. The simulation model was based on a metabolic pool type model calibrated first to simulate the
observed growth data from the nursery (calibration). Following it was used to simulate the growth of plants from
field experiments, first by an optimization of the utilization of leaves or roots only, and second by an optimization
of the utilization of both leaves and roots at the same time at different plant densities and nitrogen levels. The
models show that in order to optimize the utilization of C. bonduc it should be planted at high densities with high
nitrogen levels. Leaves and roots harvesting should take place every 50-60 days, and maximum 15% of the
biomass of roots and leaves should be harvested at each harvesting event.
* Corresponding Author: EA. Padonou padonouelie@yahoo.fr
International Journal of Agronomy and Agricultural Research (IJAAR)
ISSN: 2223-7054 (Print) 2225-3610 (Online)
http://www.innspub.net
Vol. 6, No. 3, p. 116-131, 2015
International Journal of Agronomy and Agricultural Research (IJAAR)
ISSN: 2223-7054 (Print) 2225-3610 (Online)
http://www.innspub.net
Vol. 5, No. 1, p. 14-22, 2014
Padonou et al.
Page
117
Introduction
Information on optimal planting densities, harvesting
intervals and soil conditions are important in order to
design appropriate plantation techniques and
sustainable harvesting practices. As it can be very
laborious and time consuming to investigate a long
list of different growing and utilization strategies
(initiation time, frequency, intensity) under various
climatic conditions and soil nutrient levels, especially
when talking about woody species where growth to
maturity takes several years, simulation modelling is
an attractive alternative to long term experiments
(Emanuel et al., 2005; Pandey and Bhargava, 2014).
It will not be possible to use any kind of model for
plant growth, as the model must be detailed enough
to capture the biological reactions of the plants to
utilization of different organs such as leaves, branches
and roots. A simulation model with the appropriate
detail level has been described by Gutierrez (1996),
and a metabolic pool simulation model has been
developed further for instance by Sønderskov et al.
(2006) to include competition between several plant
species for light and nitrogen, and by Rodruguez et al.
(2011) and Ponti et al. (2013) to include the growth of
woody crops such as coffee and olive. Further
modifications of the metabolic pool simulation model
by Sønderskov et al. (2006) are, however, necessary
in order to make the models realistic and functional
for modelling growth of woody plants in African
natural environments.
Few long term data are available from Africa, but
suitable data were found to develop the models based
on growth data of one shrub species, Caesalpinia
bonduc L. Caesalpinia bonduc is a thorny shrub
widely distributed throughout the tropical and
subtropical regions of the world (Kapoor, 2005). It is
an important medicinal plant, where mainly roots,
but also leaves and seeds are exploited. It is
considered the most commercialized medicinal
species in the southern part of Benin (Vodouhê et al.,
2008). C. bonduc is classified as rare and endangered
(Harden, 2002), and in Benin it has been reported
extinct in the wild, although the species can be found
in home gardens from the Guinean to the Sudanian
zone (Adomou, 2005; Assogbadjo et al., 2012). In
Africa, the long-term viability of C. bonduc
subpopulations is threatened by overexploitation of
its roots (Hutton, 2001). Furthermore, there is a low
genetic diversity within the species, which may imply
high risks for future extinction (Assogbadjo et al.,
2012). Local people have traditionally used
unsustainable methods to extract roots, leaves and
seeds of C. bonduc; for example by harvesting the
branches in order to stimulate the development of
roots. Presently local people have developed different
strategies to domesticate the species on farms for
commercialization and in home gardens for local
medicinal use (Assogbadjo et al., 2011).
In order to exploit C. bonduc optimally as a farming
crop, knowledge on appropriate utilization strategies
are needed, for instance planting and breeding
techniques. Seed germination of the species is
influenced by seed scarification (Hessou et al., 2009)
instead of seed morphology (under review). However
the seedling and plant growth are influence by seed
morphology and the large seeds are superior in terms
of seedling and plant growth (under review). The
vegetative propagation of the species has been found
to be efficient using hormonally treated stem cuttings
(Tiwari et al., 2010) and root explants (Santosh
Kumar et al., 2012).
Information on optimal planting densities, harvesting
intervals and soil conditions are also important in
order to design appropriate plantation techniques and
harvesting practices. In the present study a metabolic
pool simulation model developed by Sønderskov et al.
(2006) was modified to be able to simulate the
growth of trees and was used to come up with
suggestions on how to optimize the utilization of C.
bonduc on the short term, i.e. up to 3½ years. The
aim was to estimate both how frequently roots and
leaves can be harvested and which percentages can be
harvested each time to optimize the yield.
Materials and methods
Caesalpinia bonduc is a thorny shrub with large
bipinnate leaves. Its owers are yellow and fruits are
Padonou et al.
Page
118
inated pods with 1-2 seeds (Prajapati et al., 2006).
In traditional medicine different parts of the plant are
used to treat asthma, chronic fever, cough, headache
and stomach upset (Nandkarni, 1976; Satyavati et al.,
1956; Chopra et al., 1956). Twenty properties from
the leaves, roots and seeds of the species are used by
the local populations in Benin to ease childbirth, to
treat burns and for cultural practices like games,
weddings and the Fâ ritual (Assogbadjo et al., 2012).
Different parts of the plant have shown a variety of
pharmacological activities such as antimicrobial,
adaptogenic, contractile activity in smooth muscles
and skeletal muscles and antilarial activity (Simin et
al., 2001; Kannur et al., 2006; Datté et al., 2004;
Rastogi et al., 1996). The plant is proved
antiinflammatory (Jethmalani et al., 1966),
anthelminitic and antimalarial (Jain et al., 1992). It is
an aphrodisiac and a general tonic helping in the
rejuvenation of the body (Shrikantha Murthy, 2000).
The exploitation of the species leads to
overexploitation and threatens the species in Benin.
Planting experiments
A nursery experiment was carried out in January
2010 at the University of Abomey-Calavi, Benin
(6°45’N; 2°35’ E). The rainfall in this area is bimodal
with an annual mean of 1200 mm. The mean annual
temperature varies between 25 and 29 °C and the
relative humidity between 69 and 97%. Seeds were
sown at an equal depth; one in each of 120 pots (5.5
cm × 18 cm) made from a polythene bag and filled
with forest soil. The nursery period was 30 days. The
pots were watered twice a day (morning and evening)
throughout the duration of the experiment. The pots
were arranged in a randomized complete block design
with three replicates (or blocks). Each replicate was
composed of 40 pots. The pots were kept in a weaning
shed to reduce evaporation. At the end of this period,
growth characteristics (stem height, collar diameter,
biomass of leaves, number of leaves and root length)
were measured.
Forty eight seedlings were planted in new plastic pots
(36 cm × 32 cm) filled with forest soil. The pots were
arranged in a randomized complete block design with
6 replicates (or blocks). Each replicate was composed
of 8 pots. Every 30 days, one replicate was randomly
selected for measure of growth characteristics. Each
of the eight plants of the selected replicate was cut at
the collar with a secator. The roots were extracted
from the pot and the stem was isolated from the
leaves with secator. The different organs were
weighted with an electronic balance with precision of
0.0001g.
Sixty seedlings were used for a field experiment. The
experimental units were arranged in a randomized
complete block design with three replicates (or
blocks). Each replicate was composed of 20 seedlings.
The seedlings were planted at 1m × 1m spacing and
an equal depth. The experimental units were watered
twice a day (morning and evening) throughout the
duration of the experiment. Stem height, collar
diameter and number of leaves were measured on
selected seedlings with replication each 30 days
during 180 days (6 months).
Simulation model
The simulation model was based on the plant growth
model published by Sønderskov et al., 2006, which is
a metabolic pool type model (Gutierrez, 1996;
Gutierrez et al., 1999). For details and mathematical
description we refer to these publications and to
Appendix 1, where the special elements used in this
paper have been described. The input parameters
have been given in Appendix 2.
When running a metabolic pool type simulation
model, the first step is to calculate the amount of
sugar being produced by photosynthesis, which is
dependent on solar radiation, temperature, and the
leaf area. The second step is to distribute the
produced sugar to the different organs of the plant
according to where it is demanded, and according to a
row of priority. The row of priority in woody plants is
fruits, flowers and buds (fruits, flowers and buds are
not important when simulating young trees only) and
then roots, leaves and aerial wooden parts (trunk and
branches) according to their demands. All plant
organs in the model have minimum required nitrogen
Padonou et al.
Page
119
contents, and the photosynthesis efficiency of the
leaves is dependent on the nitrogen contentment.
Therefore, the uptake of nitrogen is important and is
simulated as being dependent on the root biomass.
Consequently, removing roots will reduce the
nitrogen uptake efficiency and result in a nitrogen
shortage. On the other hand removing leaves reduces
the photosynthetic active leaf area and thereby affects
the production of photosynthetic material resulting in
a shortage of available carbohydrates for growth. In
order to make the model capable of responding to
utilization of leaves and roots, nitrogen deficiency
changed the row of priority when distributing the
photosynthetic material to give roots higher priority
than trunk, branches and leaves. Similarly a shortage
of photosynthetic material will cause the model to
give priority to producing more leaves. This means
that harvesting roots will cause the simulated tree to
prioritize replenishing these roots, and harvesting
leaves will cause it to give priority to replenishing the
leaves. These priorities are in line with optimal
partitioning theory (Bloom et al., 1985; Gedroc et al.,
1996).
In this paper we describe different uses of the model.
First, the model was calibrated to simulate the
observed growth data from the nursery (calibration).
Second, the model was used to simulate the growth of
the plants from the field experiment (initial
validation). Third, it was used to simulate how to
optimize the utilization of leaves only, fourth, roots
only, and fifth, both leaves and roots at the same time.
The simulations were carried out using different
density and nitrogen level.
The models were run for 3½ year only, because it was
assumed to be the longest acceptable period based on
the available growth data on C. bonduc. The model
was parameterized by aid of nursery data from a
period of about 9 months (Fig. 1) and validated
against independent field data from the same period
of time (Fig. 2). The lack of validation over the
year period means that the results must be regarded
hypotheses that should be confirmed by field data
before being regarded as facts.
Results
Simulation of nursery and field experiments
The model was calibrated to simulate the
development of the data from the nursery (Fig. 1) and
was able to simulate the general trend, but was not
able to simulate all data points precisely. The
simulations of the biomass of stems, height and collar
diameter coincided well with observations, but the
data on biomass of leaves, number of leaves and roots
were so irregular that the model was not able to
capture the irregularity (Fig. 1). Especially the very
irregular data points from the third and fourth
sampling event were impossible to simulate precisely.
With the simulation of field experiment (Fig. 2), the
model was found to simulate the development in
number of leaves and height very well as the
simulations coincided very well with the observations.
Utilization of leaves or roots only
The optimal picking strategy depends on the time
scale of the model. If the time scale is only 200 days
after establishment, the optimal strategy was to pick a
large fraction of the leaves as late as possible (Fig.
3A), but when increasing the time scale to years,
the optimal strategy shifted to picking only between
10 and 20% of the leaves at a high frequency, such as
about every 20 days (Fig. 3D). The yield was almost
independent on when the utilization was initiated
(Fig. 4), i.e the model suggests that farmers can start
getting an income quickly without reducing the total
yield over a 3½ year priod. However, the yield was
lower the last year if picking was initiated early.
The results concerning the utilization of roots were
different from the results of the leaves as the optimal
strategy under all timescales was to utilize (dig up) a
fraction of the roots every 60 days, the longest
possible interval. With the short time range of 200
days the optimal utilization level was 60% (Fig. 5A)
and this shifted down to the lowest simulated
utilization of only 10% (the lowest value simulated).
Like for the leaves, the yield was almost independent
on the time of initiating the utilization (not shown).
Padonou et al.
Page
120
Utilization of leaves and roots at the same time
The simulations above were made using 1 plant m-2
but in order to investigate the importance of plant
density on the yield, the following simulations were
made using densities of 1, 5 and 25 plants per m2. The
highest yields of both roots and leaves were obtained
in the scenarios with 25 plants per m2 (Fig. 6, 7 and 8)
and high nitrogen levels (Fig. 8). At this plant density,
the utilization intervals, every 40, 50 and 60 days,
showed high yields for roots and best yields for leaves
at the medium utilization frequencies of 10-20%. At
high plant density, the highest utilization frequency,
every 10 days, utilizing only 5%, showed good yields
for roots at low nitrogen levels, but rather low yields
for leaves at the same utilization frequencies (Fig. 6E,
F). High utilization levels showed generally low yields
although there is a tendency to relatively high yields
at high utilization levels at the highest nitrogen level
(Fig. 6, 7 and 8). At high plant density and high
nitrogen level the highest utilization frequency, every
10 days, showed high yields at large utilization levels
for leaves, but low for roots (Fig. 8E, F).
Fig. 1. Comparison between observations from nursery experiment and simulations of the development of
biomass of leaves (A), number of leaves per tree (B), biomass of roots (C), biomass of stems (D), height in cm. (E),
Padonou et al.
Page
121
and collar diameter in cm. (F). Full lines show simulations and dots observations. Biomasses are in units of g per
tree. The vertical lines are standard errors (SE).
Fig. 2. Comparison between observations from field experiment and simulations of the development of number
of leaves per tree (A), and height in cm. (B). Biomasses were not measured in the field experiment. Full lines show
simulations and dots observations. The vertical lines are standard errors (SE).
Fig. 3. The yield of leaves (dry weigh) vs. the utilization level for different utilization frequencies (every 10 days
to every 60 days) and time frame of the model. A: 200 days from the 1 June, B: 565 days, C: 930 days, and D:
1295 days. Mind the different scales on the y-axes.
Padonou et al.
Page
122
Fig. 4. The yield of leaves after 1295 days at different utilization strategies in relation to utilization frequency
(every 10 60 days) and utilization level where the utilization was initiated after 2½ months (A), 1 year and 2½
months (B) and 2 years and 2½ months (C).
Fig. 5. The yield of roots vs. the utilization level depending on the utilization frequency (every 10 days to every 60
days). A: During 200 days from the 1 June (almost to the end of the year), B: During 565 days, C: during 930
days, and D: 1295 days. Mind the different scales on the y-axes.
Padonou et al.
Page
123
Fig. 6. The yield per m2 of roots (left column) and leaves (right column) after 1295 days at a low nitrogen level at
1 plant per m2 (upper row), 5 plants per m2 (mid row) and 25 plants per m2 (bottom row).
Padonou et al.
Page
124
Fig. 7. The yield per m2 of roots (left column) and leaves (right column) after 1295 days at a medium nitrogen
level at 1 plant per m2 (upper row), 5 plants per m2 (mid row) and 25 plants per m2 (bottom row).
Padonou et al.
Page
125
Fig. 8. The yield per m2 of roots (left column) and leaves (right column) after 1295 days at a high nitrogen level at
1 plant per m2 (upper row), 5 plants per m2 (mid row) and 25 plants per m2 (bottom row).
At low plant density (1 pr m2) the best yields of roots
were obtained at low utilization levels and low
utilization frequency (every 30 60 days), while the
best yields of leaves were shown at medium
utilization frequencies and medium utilization levels
(Fig. 6A, B, Figure 7A, B, and Fig. 8A, B). At lower
nitrogen levels, the combination of low utilization
level and high utilization frequency (every 10 days)
showed the best yield of leaves but lowest yield of
roots at the same combination (Fig. 6A, B, and Fig.
7A, B). In all combinations of plant densities and
nitrogen levels, the yield of roots was simulated to be
clearly higher than the yield of leaves (Fig. 6, 7 and 8).
Discussion
The simulations of the growth of C. bonduc in this
study indicated that the productivity and dynamics of
Padonou et al.
Page
126
the leaves and roots of the species are critical for the
sustainable use of the species.
An efficient conservation and domestication
programme of C. bonduc should be set up based on
the present study and the existing results on its
distribution (Assogbadjo et al., 2012), germination
and propagation (Hessou et al., 2009; Tiwari et al.,
2010; Santosh Kumar et al., 2012) and growth (under
review). Such program will specify the rate of
harvesting of the root and the leaves and prevent the
overexploitation with negative impact on the
sustainability of the species, especially in long term.
Indeed, the species is characterized to date by a very
low density and limited distribution of its populations
in the wild. Participatory domestication of indigenous
trees has been proposed as an appropriate means to
alleviate poverty (Poulton and Poole, 2001), and
could also have positive benefits on the environment
since new plantings of C. bonduc would help to
restore the declining resources of this important
species.
The simulation models show that in order to optimize
the utilization of C. bonduc it should be planted at
high densities, grown at high nitrogen levels.
Harvesting leaves and roots should take place every
50-60 days, and maximum 15% of the present
biomass of roots and leaves should be harvested at
each harvesting event. However, this conclusion
depends on economic calculations which must be
added to the simulation of yield presented here.
Economic calculations must take the sales price of
roots and leave, and the price of fertilizing the stand
of tree into account, as well as the salaries to
employees in case the owner is not the one doing the
work of harvesting the roots and leaves.
Therefore, the results presented here are useful in the
hands of growers or extension officers who have the
necessary input to turn the results of our simulations
into grower’s revenues. In the present context where
the species is threatened due to the overexploitation
of its organs (mainly the roots), this conclusion is very
important to design the appropriate condition for
plantation and harvesting period of leaves and roots
for the different purpose. This is an important step for
the sustainable use of the species.
The result that C. bonduc should be grown at high
densities may be surprising but can be explained by
the relatively short time horizon used in these
simulations. Over a short period of time, it is
important to reach an optimal utilization of the
available resources of light and nutrients as quickly as
possible, and this is done by planting high densities of
trees. By planting, for instance, only one tree per 10
m2 it will take several years to achieve a complete
canopy cover to utilize the available solar radiation,
but if 10 trees are planted per m2 a complete canopy
cover can be achieved rather quickly.
Our results suggest different utilization strategies
depending on whether the grower wants to utilize
roots or leaves. This suggests that growers may have
to decide on focusing on either roots or leaves as it
can be difficult to optimize both at the same time.
However, our results also give indications on how to
optimize yield of both roots and leaves at the same
time. As the roots are more commercialized than the
other parts (Vodouhê et al., 2008) we may suggest
the optimization of the use of the roots. However, this
optimization may induce a death to the plant and
limit its productivity. Considering the fact that the
leaves and the roots are used sometimes for the same
purposes (Assogbadjo et al., 2011; Simin et al., 2001;
Kannur et al., 2006; Datté et al., 2004; Rastogi et al.,
1996), the optimization of the yield of both roots and
leaves at the same time may be the appropriate
strategy for the sustainable use of the plant organs.
The result that the total yield is not strongly
dependent on the time when utilization is initiated is
probably surprising, but it is important to keep in
mind that an early utilization will cause a smaller
harvest after 3 years, but the yield during the first 2
years compensates to some extend for this. However,
a delayed start of utilization pays off in year 3 but
growers may not be able to wait this long to get an
Padonou et al.
Page
127
income, and the results here give some indications of
the long term effects of starting the utilization early.
The model presented here relies on many of the same
principles as the modelling of coffee trees (Rodríguez
et al., 2011) and olive trees (Ponti et al., 2013) by
having: 1) detailed model units on the different
organs of the tree, 2) being driven by the innate rate
of increase, and 3) by being based on the
physiological needs and climatic input. Unlike many
other tree growth models that are using fixed
distribution rations between root and shoot (e.g.
Simoni et al., 2000), our model has a dynamic
allocation between the different organs, that responds
to the factor limiting the growth of the plant, i.e. the
root growth gets high priority in case of nitrogen
shortage, and leaves growth in case of carbon
shortage. This means that the model tree will respond
to root utilization that limits the nitrogen uptake by
allocating resources to root growth which is in
accordance with optimal partitioning theory (Bloom
et al., 1985; Gedroc et al., 1996). This makes it very
well suited to simulate the impact of utilization of
roots and leaves on the continued growth.
It might be interpreted as a weakness of the model
that it was not possible to calibrate it to simulate the
growth in biomass of leaves and roots precisely (Fig.
1), but the data showed so large fluctuations between
the third and fourth sampling dates that they must be
ascribed to random sampling errors. Under relative
stable weather conditions it is difficult to explain the
fluctuations by other factors. Instead it supports the
model that it is capable of well simulating the field
data (Fig. 2), although it cannot be regarded a
validation of the model. The model output has to be
validated yet against field data covering a period of
time similar to the period of the simulations
presented in order to make the present results
strongly trustworthy. Therefore the results can rather
be regarded hypotheses, and used in designing the
optimal field experiments.
Acknowledgements
This work was supported by the UNDESERT
"Understanding and combating desertification to
mitigate its impact on ecosystem services" funded by
the European Commission, Directorate General for
Research and Innovation, Environment Programme
for financial support under Grant number EU FP7
243906.
References
Adomou CA. 2005. Vegetation patterns and
environmental gradients in Benin. Implications for
biogeography and conservation. PhD dissertation,
Wageningen University, Wageningen, p. 150.
Assogbadjo AE, Fandohan B, Glèlè Kakaï R,
Kyndt T, Hardy OJ, Gheysen G, Sinsin B. 2012.
Genetic evidence of the contribution of ethnic
migrations to the propagation and persistence of the
rare and declining scrambling shrub Caesalpinia
bonduc L. Human Ecology 40, 117-128.
Assogbadjo AE, Glèlè Kakaï R, Adjallala FH,
Azihou AF, Vodouhê GF, Kyndt T, Codjia JTC.
2011. Ethnic differences in use value and use patterns
of the threatened multipurpose scrambling shrub
(Caesalpinia bonduc L.) in Benin. Journal of
Medicinal Plants Research 5, 1549-1557.
Bloom AJ, Stuart Chapin F, Mooney HA. 1985.
Resource limitation in plants an economic analogy.
Annual Review of Ecology, Evolution, and
Systematics16, 363-392.
Chopra RN, Nayer SL, Chopra IC. 1956. Glossary
of Indian Medicinal Plants. Council of Scientic and
Industrial Research, New Delhi, p. 244.
Datté JY, Yapo PA, Kouame-Kof GG, Kati-
Coulibaly S, Amoikon KE, Offoumou AM. 2004.
Leaf extract of Caesalpinia bonduc Roxb.
(Caesalpiniaceae) induces an increase of contractile
force in rat skeletal muscle in situ. Phytomedicine 11,
235-241.
Emanuel PL, Shackleton CM, Baxter JS. 2005.
Modeling the sustainable harvest of Sclerocarya
birrea sussp. Caffra fruits in the South African
Padonou et al.
Page
128
lowveld. Forest Ecology and Management 214, 91-
103.
Gedroc JJ, McConnaughay KDM, Coleman JS.
1996. Plasticity in root/shoot partitioning: optimal,
ontogenetic, or both? Functional Ecology 10, 44-50.
Graf B, Dingkuhn M, Schnier F, Coronel V,
Akita S. 1991. A simulation model for the dynamics
of rice growth and development. III. Validation of the
model with high-yielding varieties.
Agricultural Systems 36, 329-349.
Graf B, Hill JE. 1992. Modelling the competition for
light and nitrogen between rice and Echinochloa
crus-galli. Agricultural Systems 40, 345-359.
Graf B, Rakatombe O, Zahner P, Delucchi V,
Gutierrez AP. 1990. A simulation model for the
dynamics of rice growth and development: Part I
the carbon balance. Agricultural Systems 32, 341-
365.
Guo D, Xia M, Wei X, Chang W, Liu Y, Wang Z.
2008. Anatomical traits associated with absorption
and mycorrhizal colonization are linked to root
branch order in twenty-three Chinese temperate tree
species. New Phytologist 180, 673-683.
Gutierrez AP. 1996. Applied Population Ecology: A
supplydemand approach. Wiley, New York.
Gutierrez AP, Yaninek JS, Neuenschwander P,
Ellis CK. 1999. A physiologically-based tritrophic
metapopulation model of the African cassava food
web. Ecological Modelling 123, 225-242.
Harden GJ. 2002. Flora of New South Wales. Vol. 2,
2nd eds. UNSW Press, Sydney.
Hessou C, Glèlè Kakaï R, Assogbadjo AE, Odjo
T, Sinsin B. 2009. Test de germination des graines
de Caesalpinia bonduc (L.) Roxb au Bénin.
International Journal of Biological
and Chemical Sciences 3, 310-317.
Jain S, Saraf S, Kharya MD, Renapurkar DM,
Dixit VK. 1992. Antimalarial activity of Caesalpinia
crista nuts. Indian Journal of Natural Products 8, 13-
15.
Jethmalani M, Sabnis PB, Gaitonde BB. 1966.
Anti-inflammatory activity of Caesalpinia bonducella.
Indian Journal of Pharmaceutical Sciences 28, 341.
Kannur DM, Hukkeri VI, Akki KS. 2006.
Adaptogenic activity of Caesalpinia bonduc seed
extracts in rats. Journal of Ethnopharmacology 108,
327-331.
Kapoor LD. 2005. Handbook of Ayurvedic
Medicinal Plants. Boca Raton, Florida.
Makita N, Kosugi Y, Dannoura M, Takanashi
S, Niiyama K, Kassim AR, Nik AR. 2012.
Patterns of root respiration rates and morphological
traits in 13 tree species in a tropical forest. Tree
Physiology 32, 303-312.
Nandkarni AK. 1976. Indian Materia Medica.
Popular Prakashan, Bombay, p. 1.
Pandey AK, Bhargava P. 2014. Effects of
harvesting intensities and techniques on re-growth
dynamics and quality of Terminalia bellerica fruits in
central India. Journal of Forestry Research 25,
177−184.
Ponti L, Gutierrez AP, Basso B, Neteler M,
Ruti PM, Dell’Aquilla A, Ianetta M. 2013. Olive
agroecosystems in the Mediterranean basin:
Multitrophic Analysis of climate effects with process-
based representation of soil water balance. Procedia
Environmental Sciences 19, 122-131.
Poulton C, Pool N. 2001. Poverty and fruit tree
research: Issues and options Paper. DFID Forestry
Research Programme, Wye College, Ashford.
Prajapati ND, Purohit SS, Sharma AK, Kumar
T. 2006. A Handbook of Medicinal Plants a Complete
Source Book, Reprint. Agrobios, Jodhpur.
Padonou et al.
Page
129
Rastogi S, Shaw AK, Kulshreshtha DK. 1996.
Characterisation of fatty acids of antilarial
triclyceride fraction from Caesalpinia bonduc.
Fitoterapia 67, 63-64.
Rodruguez D, Cure JR, Cotes JM, Gutierrez
AP. 2011. A coffee agroecosystem model: I. Growth
and development of the coffee plant. Ecological
Modelling 222, 3626-3639.
Satyavati GV, Raina MK, Sharma M. 1956.
Medicinal Plants of India, vol. 1. Indian Council of
Medical Research, New Delhi.
Severini M, Baumgaertner J, Ricci M. 1998.
Theory and practice of parameter estimation of
distributed delay models for insect and plant
phenologies. In: Guzzi R, Navarra A, (eds)
Meteorology and Environmental Sciences. World
Scientific and International Publishers, Singapore
(1990), pp. 674-719.
Simoni G, Le Roux X, Gignoux J, Sinoquet H.
2000. Tree grass : a 3D. process-based model for
simulating plant interactions in tree-grass
ecosystems. Ecological Modelling 131, 47-63.
Shrikantha Murthy KR. 2000. Vagabhata’s
Ashtang Hridayam, vol. II. Krishnadas academy,
Varanasi.
Simin K, Khaliq-uz-Zaman SM, Ahmad VU.
2001. Antimicrobial activity of seed extracts and
bondenolide from Caesalpinia bonduc (L.) Roxb.
Phytotherapy Research 15, 437-440.
Sønderskov M, Axelsen JA, Bruus-Pedersen M,
Tybirk K. 2006. Assessment of the effects of reduced
herbicide applications on selected arable weeds by a
simulation model.
Agriculture, Ecosystems & Environment 116, 216-
224.
Ter-Mikaelian MT, Korzukhin MD. 1997.
Biomass equations for sixty-five North American tree
species. Forest Ecology and Management 97, 1-24.
Vodouhê FG, Coulibaly O, Assogbadjo AE,
Sinsin B. 2008. Medicinal plant commercialization
in Benin: An analysis of profit distribution equity
across supply chain actors and its effect on the
sustainable use of harvested species.
Journal of Medicinal Plants Research 2, 331-340.
Appendix 1.
Here we give a summary of the plant growth model
published by Sønderskov et al., 2006 and a
description of the modifications made to make the
model capable of simulating the growth of tropical
woody plants. The plant growth model published by
Sønderskov et al., 2006 is a metabolic pool type
model (Gutierrez, 1996). For basic details and
mathematical description we refer to these
publications. The basic element of the model is a
plant population (may consist of only one plant), and
the growth of the population is simulated as the
growth of coupled populations of plant organs (roots,
leaves, stems, buds, flowers and fruits). The driving
force of the growth of a population of plant organs is
the innate demand for growth and reproduction and
therefore each plant organ (except flowers and fruits)
has a reproduction rate. This is the rate of production
of new roots, new leaves, new stems and new buds,
while there is no reproduction rate for flowers and
fruits. Flowers are produced from buds that have
finalized their development, and fruits are produced
from flowers that have finished their bloom. Plant
organs are in principle simulated as having a growing
stage and a mature stage where the organ has reached
the final size, which means that growing organs and
mature organs are controlled separately in the model.
However, buds and flowers are special in this respect
because they only have one stage. Buds have no
mature stage as mature buds burst into flowers, and
flowers have no growing stage. In the model of
Sønderskov et al. (2006) roots and stems had a
growing stage and a mature stage, because the model
was constructed to simulate annual weeds and
grasses. In the present model it was necessary to
reduce the number of stages of roots and stems
Padonou et al.
Page
130
(trunks + branches) to only one, namely a developing
stage, and this developing stage was given indefinite
life time, because these two organs will persist during
the entire life time of the tree, which was not defined
in the model. In the model of Sønderskov et al.
(2006) and the large number of models by Gutierrez
and co-workers (e.g. Rodriguez et al., 2011; Gutierrez
et al., 1999; Gutierrez, 1996), the population
dynamics is controlled by distributed delays, which
splits each stage into a number of sub-stages. A
distributed delay can be regarded a kind of book
keeping device, keeping track of the ageing and
growth by aid of transferring matter from one sub-
stage to the other, and while leaving the last sub-stage
the matter is being transferred to the following stage,
e.g. from growing stage to mature stage. As the life
time of stems and roots is not defined in this model,
the population dynamics of these organs was
controlled by simple variables describing the numbers
and biomass instead of using distributed delays
(Severini et al., 1998) that require a finite
developmental time.
The model makes daily calculations of demand for
photosyntheates and nitrogen (split on demand for
respiration, reproduction, growth and reserves) of
each plant organ and this demand is the driving force
of the simulation model. Following the calculation of
the demand, the model calculates the supply of
photosyntheates and nitrogen for the plant growth
based on the available light (available from climate
files) and simulated amount of nitrogen in the soil.
The distribution of photosyntheates and nitrogen
between plant organs depends on the priority
between the organs and the row of priority within the
organs (respiration, reproduction, growth and
reserves). The priority between organs was made
dependent on the nitrogen supply/demand ratio
relative to the photosyntheate supply/demand ratio.
If the nitrogen supply/demand rate was lowest the
roots were favored in the allocation of both nitrogen
and carbon the following day making it possible for
the plant/tree to acquire more nitrogen from the soil
the following simulation days, and if the carbon
supply/demand ratio was lowest this was interpreted
as a carbon shortage, and therefore the obvious
reaction of the plant/tree was to increase the
photosynthetic apparatus, e.g. leaves. This possibility
of switching priorities between roots and aerial parts
(all organs except roots) made it possible for the
plant/tree to react to human utilization of either roots
or aerial parts, i.e. after removal of either roots
and/or leaves. Utilization of fruits, flowers or buds
did not affect the allocation priorities. The calculation
of the supply of photosyntheates for the plant
population was based on the photosynthesis of the
leaves only and the uptake of nitrogen was based on
the roots only, although the demands of the plant
population was the sum of demands from all plant
organs.
The demand for respiration was dependent on
temperature and biomass of leaves, buds, flowers and
fruits, but for the roots and stems the respiration was
only made dependent on the mass of the actively
metabolizing cambium layer between the wooden
part of the trunk and the bark. For the rooting system
this is supported by Pretziger et al. (2002), who found
a higher content of nitrogen in the finest roots of all
investigated trees species from North America
indicating a high metabolic activity in the finer roots,
and by Makita et al. (2012), who found an increase in
respiration with decreasing root diameter in 13
tropical tree species. Combining this with the fact that
the finest roots also have the lowest “stele: root
diameter ratio” (Guo et al., 2008), and thereby the
highest “cortex: root diameter ratio”, it seems to be
reasonable to let the growth and respiration depend
on an estimate of the biomass of the cambium layer
and not on the biomass of the entire rooting system.
It has, however, not been possible to find literature on
the biomass of the cambium layer in relation to the
biomass of the shoot, and therefore an estimate of the
cambium layer was assumed to be the same fraction
of the bark layer, as the bark layer was of the total
biomass of the stem/trunk. The simulation of the
biomass of bark was done through calculations of the
bark fraction of the total biomass of the stem/trunk,
which varies with species and the size of the tree.
Here an allometric relationship between diameter at
Padonou et al.
Page
131
breast height DBH and the percentage of bark was
used. Ter-Mkkaelin and Korzukhin (1997) have made
a review of allometric relationships for the correlation
between DBH and of different components of the
trees for a long list of North American species, and
their data on deciduous species were used to calculate
average values of a and b for the allometric
relationship (1). These average values were used in
this model, because the species specific relationship
was not known
(1)
where B is the biomass fraction of bark, DBH is the
diameter at breast height, and a = 0.02687 and b =
1.86913 are constants. Due to shortage of knowledge
the same proportion of cambium layer was used for
the roots as for the stems/trunks.
The vertical distribution of the plant organs was
simulated by equation (2), which was modified from
Graf et al. (1992)
(2)
where p(x) is the biomass of a plant organ at the
height x, h is the actual plant height, and k1and k2 are
constants. If k1 = 4 and k2 = 5 the equation will
distributed the biomass as having the largest part
close to the height h, and when using lower values of
k1 and k2 such as k1 = k2 = 1 the distribution is
almost even up through the vegetation layer. The
assimilation of photosynthetic material follows the
description of Graf et al., 1990, Graf et al., 1991,
Gutierrez, 1996, but the canopy is split into a number
of vertical zones, where the amount of solar radiation
available for the second layer, is the global solar
radiation minus what was assimilated in the
uppermost layer. The solar radiation available for the
third layer is the global solar radiation minus what
was assimilated in the two uppermost layers, etc.
Appendix 2.
The parameters of C. bonduc in the simulation model and their origin. The allometric parameters were from the
equation Mass = a×DBHb where DBH is given in mm. and mass in g.
Roots
Trunk
Growing leaves
Mature leaves
T0 (°C)
10 a)
10 a)
10 a)
10 a)
Developmental time/life time (D°)
No limit
Nop limit
250 b)
2500 b)
Initial mass (g)
0,0022 b)
0,0061 b)
0,0061 b)
Creation rate (g/g /D°)
3.5×10-4 b)
5×10-4 b)
3.5×10-4 b)
Respiration rate (g/g /D°)
0.00883 b)
0.00883 b)
0.00883 b)
0.00883 b)
Growth rate (g/g /D°)
0.004 b)
0.020 b)
0.020 b)
Allometric parameter a
0.131 c)
0.163 c)
0.11 c)
Allometric parameter b
1.764 c)
1.565 c)
1.767 c)
Max height (cm)
1000 a)
Start heright (cm)
23 b)
Elongation factor
6×10-4 b)
a) Educated guess
b) Parameter fitted to simulate data from pot experiments
c) Parameter calculated based on data from pot experiments
... L'extinction de C. bonduc au Bénin a déclenché une série de recherche visant sa conservation. Les données existantes sur l'espèce au Bénin ont documenté son champ de distribution et sa caractérisation génétique selon les zones climatiques du pays (Assogbadjo et al., 2012), l'impact du climat sur la morphologie de ses graines (Padonou et al., 2015b), les types d'habitats qui favorisent le développement des jeunes tiges de l'espèce (Padonou et al., 2015a), les différentes formes d'usage de l'espèce, les connaissances ethnobotaniques, les perceptions locales et les stratégies locales pour sa conservation (Assogbadjo et al., 2011;Lokonon et al., 2021). Les activités de recherches devront dans le futur viser à approfondir les travaux précédents. ...
... On la retrouve uniquement dans les systèmes agroforestiers (champs, jardin de case, jachère etc). Les études conduites sur l'espèce au Bénin ont toutes annoncé sa présence dans plusieurs jardins de case dans le pays(Padonou et al., 2015a;Padonou et al., 2015b;Assogbadjo et al., 2011), et autour des champs et des maisons(Assogbadjo et al., 2012), illustrant sa domestication pour la survie de sa population. ...
Technical Report
Full-text available
Caesalpinia bonduc (L.) Roxb est la plante médicinale la plus commercialisée du sud Bénin. Elle est éteinte à l'état sauvage. Cela est dû à la surexploitation de ses racines qui sont essentiellement utilisées dans le traitement de la prostate et de la faiblesse sexuelle. De plus, la dureté des graines ne favorise pas la régénération spontanée de l'espèce à l'état sauvage. L'espèce est classée comme rare et menacée de disparition dans le monde. Des pieds isolés de l'espèce, plantés de main d'homme sont souvent rencontré dans les agglomérations et quelques fois dans les jardins de cases et systèmes agroforestiers. Pour inverser la tendance actuelle et réduire les menaces qui pèsent sur cette espèce, des stratégies de conservation in situ et ex situ sont nécessaires de toute urgence. C'est dans cette optique que la présente fiche technique se propose de donner l'état des connaissances techniques sur l'espèce au Bénin.
ResearchGate has not been able to resolve any references for this publication.