Biomass and litterfall in a native lowland rainforest: Marelongue Reserve, La Réunion Island, Indian Ocean

doi:10.1016/j.foreco.2007.06.041

Forest Ecology and Management

Volume 252, Issues 1–3, 30 November 2007, Pages 257-266

https://doi.org/10.1016/j.foreco.2007.06.041 Get rights and content

Abstract

La Réunion Island (Indian Ocean) is one of the tropical oceanic islands that exhibits a lowland forest still not destroyed by human activities and by introduced species. In order to protect and conserve native intact habitats, information concerning their functioning is needed. We measure the biomass and litterfall of the Marelongue Reserve lowland rainforest on a permanent plot and examine what is the part played by the different species as well as the role of seasonality and cyclones. The total biomass estimated at 535 t/ha is relatively consistent with biomasses measured in other tropical forests. Over 92% of the biomass is due to 10 species only. The annual fine litterfall, measured at 7.6 t/ha during the two studied years, is similar to annual litterfall measured in similar ecosystems and is strongly affected by cyclones. The litterfall seasonality seemed to be linked to temperature fluctuations, rather than to either of the other two studied climatic factors (rainfall and wind speed). The results of this study will define valuable basic information for conservation and management of fragmented rainforest remnants.

Introduction

Insular tropical rainforests and particularly lowland forests are rapidly transformed or destroyed by man after their discovery. The impact of exotic species on ecosystems can be measured on such islands. The most documented of these are the Hawaiian islands (Mueller Dombois, 1971, Vitousek and Walker, 1987, Vitousek et al., 1993, Vitousek et al., 1995, Vitousek et al., 1997). La Réunion island, with an exceptionally high concentration of endemic species and a high destruction rate of native habitats, belongs to the 25 “hot spots” where conservation is considered a priority (Myers et al., 2000). However, in La Réunion Island, large tracts of native habitats (30% of the area) survive, as do sizeable remnants of lowland habitat, well preserved and free of alien animals. In order to conserve and manage intact native habitats, information concerning their functioning is needed.

Data concerning the ecology of the native lowland rainforest of La Réunion is provided by previous studies (Strasberg et al., 1995, Strasberg, 1996), which describe the diversity, size composition, spatial aggregation among trees, and tree mortality of this forest. The Marelongue reserve, created in 1981, forms the basis of the last band of intact lowland vegetation in the Mascarenes. It is uninterrupted forest from 100 m above sea level to the summit of the volcano. The reserve stretches from 100 to 700 m altitude over a total of 68 ha composed of three narrow bands of 18, 13 and 37 ha (Fig. 1). The vegetation of the reserve grows on a 500-year-old basaltic flow (Bachelery, 1999). The productivity of this lowland rainforest has never been documented. The main goals of this study are to measure the biomass and litterfall of the Marelongue forest, which has not been significantly transformed by alien species, and to determine, among the factors controlling the forest's productivity, what is the part played by the native tree species as well as the role of climatic factors (seasonality and cyclones).

Section snippets

Study site

The study was carried out in the natural reserve of Marelongue in the southeast of La Réunion Island (Mascarene Archipelago, Indian Ocean) on the flanks of the Piton de La Fournaise volcano (21°20′95S and 55°44′64E) (Fig. 1). The mean annual precipitation and temperature of the island are 4000 mm and 23 °C, respectively (Raunet, 1991). Two seasons can be distinguished: a warm and humid season from November to April, and a cool and dry season from May to October. The prevailing wind throughout the

Biomass

On the permanent plot, a total of 1079 canopy trees with diameters of 0.10 m or more were identified. The individuals all belonged to 41 species, belonging to 21 families, widely distributed over the tropical zone. The introduced species represent 5 species (only 18 individuals). Five species are indigenous and 31 are endemic to the Mascarenes (Bosser et al., 1976).

In the canopy, only three species (Labourdonnaisia callophylloïdes, Antirhea borbonica and Nuxia verticillata) account for more than

Composition and biomass of the marelongue forest

This study shows similar results to the one of Strasberg (1996), concerning the composition of the forest canopy. These two studies also show that, as opposed to most tropical forests, the Marelongue forest is characterized by a low number of lianas, an absence of trees with buttress roots and a high tree density.

The total biomass estimated at 535 t/ha is similar to but slightly higher than the biomass measured in other tropical forests, generally estimated at around 400 t/ha (Jung, 1969,

Conclusion

Our study indicates that the lowland tropical rain forest of La Réunion Island has a relatively high biomass (535 t/ha) and litter production (7.6 t/ha per year of fine litter) for a young insular ecosystem. A few species (N. verticillata, L. callophylloïdes, Syzygium sp. and A. salicifolia) are dominant in terms of abundance, biomass and litter production; these species are all indigenous and control the forest's productivity. The litterfall was found to fluctuate both inter-annually and

Acknowledgements

This study was supported by the Programme National Sols et Erosion (French programme 2001–2002), the Direction Régionale de l’Environnement-La Réunion, the Office National des Forêts-La Réunion and the Institut de Recherche pour le Développement. We would like to thank the helpful contribution of the Laboratory of Earth Sciences, University of Saint-Denis de la Réunion, the Conservatoire Botanique National des Mascarins and Météo France. The manuscript was greatly improved by J. Guiot and two

References (45)

S. Brown
Measuring carbon in forests: current status and future challenges
Environ. Pollut.
(2002)
J.W. Raich
Aboveground productivity and soil respiration in three Hawaiian rain forests
Forest Ecol. Manag.
(1998)
X. Xu et al.
Effect of typhoon disturbance on fine litterfall and related nutrient input in a subtropical forest on Okinawa Island, Japan
Basic Appl. Ecol.
(2004)
T. Arujo et al.
Comparison of formulae for biomass content determination in a tropical rain forest in the state of Parà, Brazil
Forest Ecol. Manag.
(1999)
P. Bachelery
Le fonctionnement des volcans boucliers. Exemple des volcans de la Réunion et de la Grande Comore, Thèse d’habilitation à diriger les recherches
(1999)
F. Bernhardt-Reversat et al.
La forêt sempervirent de Basse Côte d’Ivoire
F. Blasco et al.
Etude d’un écosystème forestier montagnard du sud de l’Inde
Ecology
(1975)
J. Bosser et al.
Flore des Mascareignes
(1976)
G. Bourgin et al.
Composition floristique strucure et régénération d’une forêt tropicale Rapport de Maîtrise
(2001)
S. Brown et al.
Biomass of tropical forests: a new estimate based on forest volumes
Science
(1984)

Cadet, L.J.T., 1977. La végétation de l’île de La réunion. Etude phytoécologique et phytosociologique, Thesis...

Cannell, M.G.R., 1982. World forest biomass and primary production data, London, p....

J. Chave

Dynamique spatio-temporelle de la forêt tropicale

(1999)

D.A. Clark et al.

Net primary production in tropical forests: an evaluation and synthesis of existing field data

Ecol. Appl.

(2001)

T.R. Crow

Common regressions to estimate tree biomass in tropical stands

Forest Sci.

(1978)

L.M. Dascanio et al.

Biomass structure and dry matter dynamics of a subtropical alluvial and exotic Ligustrum forests at the Rio de la Plata, Argentina

Vegetatio

(1994)

P. Detienne et al.

Identification des bois de l’île de la Réunion

(1993)

F. Hallé et al.

Tropical trees and forests

(1978)

A. Hladik

Importance des lianes dans la production foliaire de la forêt équatoriale du NE du Gabon

C. R. A. S. Paris

(1974)

S.P. Hubbell

A Unified Theory of Biodiversity and Biogeography

(2001)

G. Jung

Cycles biogéochimiques dans un écosystème de région tropicale sèche: Acacia albida (Del.). Sol ferrugineux tropical peu lessivé (Dior.)

Acta Oecologica/Oecologia Plantarum

(1969)

Kirman, S., 2003, Cycles biogéochimiques et biodiversité en forêt tropicale humide: étude d’une succession primaire sur...

Cited by (7)

Improving litterfall production prediction in China under variable environmental conditions using machine learning algorithms
2022, Journal of Environmental Management
Citation Excerpt :
Enhancing our understanding of litterfall production in forests is of great importance for improving forest management and knowledge of the carbon cycle. Previous studies have developed regression models to identify the key drivers of litterfall and predict litterfall production (Starr et al., 2005; Kirman et al., 2007; Chave et al., 2010); however, the interactions among many abiotic and biotic factors that influence forest litterfall and the relationship between these factors and litterfall production are nonlinear and complex (Guo et al., 2019). Previous studies typically considered a limited number of abiotic and biotic features (Liu et al., 2004; Chave et al., 2010; Kirman et al., 2007; You et al., 2017); therefore, the complex nonlinear relationships involved in the prediction of litterfall production were not adequately characterized.
Litterfall production is a major process within forest ecosystems that plays a crucial role in the global carbon cycle. Accordingly, studies have explored the abiotic and biotic features that influence litterfall production. In addition to traditional statistical models, the rapid development of nonparametric and nonlinear machine learning models, such as random forest (RF), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost), have provided new methods of predicting the production of forest litterfall. Here, we evaluated the ability of the abovementioned models and mixed effect random forest (MERF) models to predict total annual litterfall production—based on several abiotic and biotic features—using 968 records from 314 forest sites covering the full geographical range of Chinese forests. In general, machine learning models were found to outperform linear mixed models. In particular, the MERF models ranked the highest in terms of performance (R² = 0.7), which may be attributed to their ability to characterize nonlinear relationships between features and litterfall production. The key drivers were climate-related features and forest age, with the mean annual temperature and age positively correlated with litterfall production. Furthermore, the correlation between forest type and litterfall production was more significant for needleleaf forests than for other forest types. For needleleaf and broadleaf forests in several regions in China, the future litterfall production was predicted to be the highest under IPCC representative concentration pathway (RCP) 8.5, followed by RCP 4.5, RCP 2.6, and the original scenarios (sample data). Improved models to better understand and estimate litterfall production in forests at present and in the future are required for forest management planning to minimize the negative impacts of climate change on forest ecosystems.
An assessment of biological control of Rubus alceifolius invasion on Réunion Island (Mascarene archipelago)
2021, Biological Control
Biological control can be an effective conservation strategy to manage invasive species when other conventional methods cannot be implemented. Estimating the benefits of biological control for the conservation of native biodiversity remains a challenging task. Following the introduction of a defoliating sawfly, Cibdela janthina, for biological control of invasive giant bramble, Rubus alceifolius, on Réunion Island (Mascarene Archipelago), we monitored the consequent change in R. alceifolius cover as well as species richness and cover of non-native species and native species within a set of patches along an altitudinal gradient in the National Park of Réunion Island. There was an increase in native species richness and cover together with a decrease in R. alceifolius cover, showing that biological control positively impacted the native plant communities. The effectiveness of C. janthina was influenced by the altitudinal range, the surface area of the patches and the location of each patch, i.e forest edge or forest matrix. We discuss the extent to which the use of biological control within an integrated approach can optimize benefits for biodiversity conservation in the short to medium term by selecting complementary and standardised protocols.
Incipient weathering by Stereocaulon vulcani at Réunion volcanic island
2014, Chemical Geology
Citation Excerpt :
The first seedlings of the pioneer tree species Agarista salicifolia were seen, represented by 5750 individuals/ha. For comparison, on the 500-year-old flow they represent only 44 individuals ha− 1 (Kirman et al., 2007). The composition of the colonizing plants on this 24-year-old flows was similar to other young undisturbed lava flows studied at low elevations (Cadet, 1977; Strasberg et al., 1995; Strasberg, 1996).
The impact of early land plants and fungi in increasing global weathering is still debated, particularly before the advent of vascular plants during the Devonian. Here we present a study of the incipient weathering of basalt by Stereocaulon vulcani, a native colonizing lichen on Réunion Island (Indian Ocean). We analyzed the chemistry and mineralogy of a 24-year-old flow located at low altitude that was mostly covered by S. vulcani with aboveground biomass of 6249 kg ha^− 1. The chemical composition of S. vulcani showed that besides C and N, Si and Fe were the dominant elements. The Si stored in the aboveground pioneer vegetation gives 27 kg ha^− 1, comparable to the Si stored in the old-growth at the Marelongue Reserve (Meunier et al., 2010). On thin sections, the inner part of S. vulcani was mostly composed of Si while Fe coatings observed at the base of the thallus may be the result of wind blow dust interception as suggested by Cochran and Berner (1992). Using BSE images on SEM, we showed evidence of dissolution of the glass matrix at the basalt–thallus contact. The quantification of porosity by digital imagery showed a variation between 7% in the unweathered zone to near to 40% at the surface. A maximum denudation rate of 6.7 μm year^− 1 is estimated to fall within the range of the values reported in the literature. Using our data for analog to the past, we suggest that early land plants would have been capable to mobilize Si in a similar proportion as vascular plants and should have significantly affected the weathering of land before the advent of vascular plants.
The output and bio-cycling of Si in a tropical rain forest developed on young basalt flows (La Reunion Island)
2010, Geoderma
Citation Excerpt :
In order to do so, and for each species, a mixture of leaves collected at different dates and from different traps was ground to powder. Wood of the 6 main species, contributing more than 84% to the total aboveground woody biomass (Kirman et al. 2007) was sampled and dried to constant weight. Three soil and one basalt samples were randomly collected on the permanent plot.
The importance of plant bio-cycling in the biogeochemical cycle of Si has been pointed out by recent studies. We analysed the litter and wood chemical composition of a lowland rainforest (Marelongue Reserve, Réunion Island, Indian ocean) not impacted by human activities, developed on a recent basalt flow dated ci 500 years. The amount of elements stored in the above ground biomass falls within the range of tropical forests developed on moderately fertile older soils with Ca, K, Mg and Si values respectively at 1458 kg ha⁻¹, 1506 kg ha⁻¹, 246 kg ha⁻¹and 17 kg ha⁻¹. The estimated mineral uptake of the old trees gives a flux equivalent to 5 ± 2% of the elements returned to the soil with litterfall. The four species that contribute the most to the litterfall with 70% of the leaf fall represent only 45% of the annual Si bio cycling. The result demonstrates that species that are not highly productive in a rain forest could be significant for the bio-cycling of parent-rock elements such as Si. A mass balance calculation shows that the dissolved silicon (DSi) issued from the dissolution of basalt is 1.8 times the DSi from litterfall. Here, the biogeochemical cycle of Si is characterized by a limited contribution of the bio-cycling compartment compared to other cases reported for rainforests.
Trees as net sinks for methane (CH<inf>4</inf>) and nitrous oxide (N<inf>2</inf>O) in the lowland tropical rain forest on volcanic Réunion Island
2021, New Phytologist
Mineralization and nutrient release of leaf litter and fallen bromeliads in a pine - Oak forest in Oaxaca, Mexico
2011, Agrochimica

View all citing articles on Scopus

View full text

Biomass and litterfall in a native lowland rainforest: Marelongue Reserve, La Réunion Island, Indian Ocean

Abstract

Introduction

Section snippets

Study site

Biomass

Composition and biomass of the marelongue forest

Conclusion

Acknowledgements

Environ. Pollut.

Forest Ecol. Manag.

Basic Appl. Ecol.

Comparison of formulae for biomass content determination in a tropical rain forest in the state of Parà, Brazil

Forest Ecol. Manag.

Le fonctionnement des volcans boucliers. Exemple des volcans de la Réunion et de la Grande Comore, Thèse d’habilitation à diriger les recherches

La forêt sempervirent de Basse Côte d’Ivoire

Etude d’un écosystème forestier montagnard du sud de l’Inde

Ecology

Flore des Mascareignes

Composition floristique strucure et régénération d’une forêt tropicale Rapport de Maîtrise

Biomass of tropical forests: a new estimate based on forest volumes

Science