Relationships between climate, soil moisture and phenology of the woody cover in two sites located along the West African latitudinal gradient

https://doi.org/10.1016/j.jhydrol.2009.01.023Get rights and content

Summary

The study quantifies the relationships at local scale between phenology and determinants of climate and soil water resources at two sites located along the latitudinal gradient of West Africa, one in the central Sahel (Mali), the other in the Sudanian bioclimatic zone (Benin). The aim is to improve our knowledge on possible vegetation response to possible climate change. Within the Sudanian site, average annual rainfall is 1200 mm, extending from April to October, while, in the Sahelian site, it is 370 mm, occurring from June to September. Physical data were collected from the African Monsoon Multidisciplinary Analysis research programme. The phenology of the dominant species was monitored in four types of vegetation cover at the wetter site, and in three types of vegetation cover at the drier site. For each sampled plant, leafing, flowering and fruiting were recorded as binary variables in terms of the presence/absence of phenophases. A small proportion of the variability of each phenophase occurrence is explained by the logit models. However, rainfall rise is significantly linked to leafing probability increase in the Sahelian site but not in the Sudanian site. Day length extension and temperature decrease are significantly correlated with an increase in leafing in the Sudanian site, but not in the Sahelian. On both sites, the increase in cumulative rainfall is not found to be linked to an increased probability of reproductive phenophases (negative or non-significant relationships). Air temperature is positively correlated with flowering rate in the Sudanian site, but, all other factors being constant, no climate factors are found to be highly significant of flowering occurrence in the Sahel. Fruiting probability is positively correlated mainly with temperature within the Sahelian site. Leafing occurrence is positively correlated with soil moisture in the 0–1 m layer for the Sudanian site, but not for the Sahelian site. Significant relationships between fruiting occurrence and soil moisture may reflect a prior selection of plants on fruiting period that maximizes seed dispersion and germination differently at the two sites. While vegetative and reproduction schedules may be determined by specific genetic factors, the physical environment controls the possibility of their expression. Reduction of the rainfall amount and intensity may increase reproduction rates in wet areas. Although this factor should decrease leafing rate, it does not influence reproduction at dry sites, except through the decrease in air humidity. In wetter areas, increasing temperature may reduce leafing, but may increase reproduction rates. Cover reduction may have an impact on local physical factors and, consequently, probably also affects vegetation phenology.

Introduction

Plant phenology is the study of the timing of the development of vegetative and reproductive plant organs, most notably leafing, flowering and fruiting (Haugaasen and Peres, 2005). It is concerned with understanding the variability of these events or phenophases, particularly in relation to their biotic and abiotic forcings (Lieth, 1974).

The timing of phenophases has major implications for plant community structure, function and regeneration, as well as the quantity and quality of resources available for consumers (Williams et al., 1999). Furthermore, timing and duration of leafing determine exchange periods of carbon dioxide, oxygen and water between the biosphere and the atmosphere. Temporal and spatial variability of the phenophase is not only an excellent indicator of short- and long-term climatic variability, but it also influences surface meteorology. However, we still lack the ability to predict the future trends of plant phenology in response to probable climate change (White and Nemani, 2003, Jolly et al., 2005). To fill this gap, a first step would be to document conditions of vegetation maintenance through quantification of the relationships between current climate and vegetation phenology. As a corollary, studying the response of vegetation phenology to the climate may help to assess how well the cover is adapted, but also is vulnerable, to climate variability or change. In addition, the importance of climatic factors in controlling plant phenology has been shown to vary with location.

The present study aims to investigate some climatic and hydric factors determining the phenophase timing of woody cover at one site located in the central Sahel (northern Mali) and another in the Sudanian bioclimate zone (northern Benin) (Aubreville, 1950). These two sites belong to the AMMA research programme (African Monsoon Multidisciplinary Analysis, Redelsperger et al., 2006). This comparative approach provides information on some selective physical factors that may operate in two contrasting environments positioned along the seasonal course of the monsoon, in the context of increasing climatic variability or warming (Petrů et al., 2006). Soil moisture dynamics is thought to reflect local environmental conditions, whereas time-evolution of climatic factors reflects regional conditions. However, high spatial variation of rainfall distribution is now recognised at the local scale, especially in Sahel (Le Barbé and Lebel, 1997). The study also provides knowledge about the adaptation of two cover types with markedly different floristic composition and disturbance regimes subjected to distinct climates and water resource dynamics.

In tropical areas, information remains scarce and the debate continues on the factors determining phenophase timing, especially for woody cover (Bordiert, 1994, Breman and Kessler, 1995, Bie (de) et al., 1998, Myers et al., 1998, Sarmiento and Monasterio, 1983, Seghieri et al., 1995, Eamus and Prior, 2001). We expect different levels of significance of correlation between phenophases and abiotic variables in two disparate environments and vegetation. At the drier site, rainfall distribution is expected to determine leafing, with a high significance level because the rainy season shows a strong periodicity. The rain distribution, especially at the beginning of the season, is highly fluctuating and poorly predictable from 1 year to the next. Within the Sudanian site, rainfall distribution appears to be less limiting on the woody vegetation activity. Consequently, woody vegetation phenophases should be more strongly controlled by factors other than those responsible for water resource distribution (rainfall or soil moisture dynamics). Since no seasonal constraint occurs within the Sudanian site, the photoperiod has been suggested as being the only abiotic cue regulating canopy leafing at the global scale (Jolly et al., 2005), although day length is more stable over the year than in the Sahel. At this relatively wet site, not only climatic factors but also high intensity of land use has selected non-cultivated vegetation. Maintenance of most of the present vegetation cover is determined by its resistance to human pressure in addition to its adaptation to physical environment. Thus, it is interesting to analyse how far the phenology of this secondary vegetation is still controlled by climate, since the dominant species regenerated after cropping belong to the original flora, although occurring in lower abundance.

Section snippets

Bioclimatic sites

The Sudanian site is located in the Donga watershed that extends over 586 km2 in northern Benin (9°40′–9°54′N and 1°34′–1°58′E), 450 km North-West of Cotonou. Average annual rainfall is 1200 mm (1950–2005, Lawin, 2007). Sixty percent of the annual rainfall is concentrated between July and September, with a wet period sensu Gaussen (Charre, 1997) extending from April to October. However, rainfall can occur throughout the year with the lowest probability during December and January. Vegetation is

Annual woody cover phenology and annual cycle of physical environmental parameters

At scale of bioclimatic site, accounting for all the woody cover, local sites and species mixed, differences in patterns of annual phenology occured between Sudanian and Sahelian sites, in relation to differences in course of climatic parameters and water resources distribution (Fig. 1, Fig. 2).

There was a shift of 3–4 months of the phenological calendars between the two sites. The ombrothermic diagrams (Figs. 1b and 2b) indicated yet that, over the studied period, the wettest part of the rainy

Overall features

Dynamics of physical parameters described here for the Sudanian site (Fig. 1) agree with the description for the end of the rainy season of Guyot et al. (2009). Guichard et al. (2009) go deeper than we do in this study into analysis of the relationships between seasonal variations in thermodynamics parameters, surface radiative budget and monsoon dynamics at the Sahelian site.

The results obtained here clearly indicate that the phenology of the dominant woody cover on the two investigated sites

Conclusions

It is difficult to predict trends in the response of the woody cover to probable tropical climate change because of the complexity of relationships between phenology and physical factors. Furthermore, the challenge is all the greater due to significant partial correlations (p < 0.001) between predictor variables, which differ in relation to the dataset or the site considered (Table 3, Table 4, Table 5). These collinear relationships between predictor variables imply prudence in the interpretation

Acknowledgements

Based on a French initiative, AMMA was set up by an international scientific group and is currently funded by a large number of agencies, especially from France, the UK, the USA and African countries. It has received a major financial contribution from the European Community’s Sixth Framework Research Programme. Detailed information on scientific coordination and funding is available on the AMMA International website http://www.ammainternational.org.

We thank especially Pierre Onodjè Agbani

References (50)

  • R. Borchert

    Induction of rehydration and budbreak by irrigation or rain in deciduous trees of a tropical dry forest in Costa Rica

    Trees

    (1994)
  • R. Borchert

    Water status and development of tropical trees during seasonal drought

    Trees

    (1994)
  • R. Borchert et al.

    Environmental control of flowering periodicity in Costa Rican and Mexican tropical dry forests

    Global Ecology and Biogeography

    (2004)
  • R. Bordiert

    Water status and development of tropical trees during seasonal drought

    Trees

    (1994)
  • H. Breman et al.

    Woody Plants in Agro-Ecosystems of Semi-Arid Regions

    (1995)
  • J. Charre

    Dessine moi un climat. Que penser du diagramme ombrothermique?

    MappeMonde

    (1997)
  • C. Flahault

    La nomenclature en géographie botanique

    Annales de Géographie X

    (1901)
  • T. Foucart

    Colinéarité et régression linéaire

    Mathematics and Social Sciences, 44e année

    (2006)
  • Fournier, C., 1995. Fonctionnement hydrique de six espèces ligneuses coexistant dans une savane sahélienne (région du...
  • Fournier, A., 199l. Phénologie, croissance et production végétale dans quelques savanes d’Afrique de l’Ouest. ORSTOM...
  • Harris, J., 2004. Contribution à l’analyse du fonctionnement d’une végétation Soudano-Guinéenne en réponse au climat et...
  • T. Haugaasen et al.

    Tree phenology in adjacent Amazonian flooded and unflooded forests

    Biotropica

    (2005)
  • C. Hély et al.

    Sensitivity of African biomes to changes in the precipitation regime

    Global Ecology and Biogeography

    (2006)
  • P. Hiernaux et al.

    Fluctuations saisonnières de la feuillaison des arbres et des buissons sahéliens. Conséquences pour la quantification des ressources fourragères

    Revue d’Elevage et de Médecine Vétérinaire des Pays tropicaux

    (1994)
  • D.W. Hosmer et al.

    Applied Logistic Regression

    (2000)
  • Cited by (0)

    View full text