Simulation model of natural tropical forest dynamics

doi:10.1016/0304-3800(91)90127-M

Ecological Modelling

Volume 59, Issues 1–2, 1 December 1991, Pages 37-71

https://doi.org/10.1016/0304-3800(91)90127-M Get rights and content

Abstract

Assessments of the long-term natural regeneration of tropical forests following selective logging are today mostly based on extrapolation of limited empirical observations. Mounting evidence suggests that current logging policies overestimate forest regrowth by a wide margin, and are therefore not sustainable. The need for more reliable assessment of natural forest growth dynamics has led to the development of a vertically and spatially structured dynamic simulation model of natural forest development. The model accounts explicitly for biomass and tree numbers in five distinct canopy layers (seedlings, saplings, poles, main canopy, emergents). Energy accounting of assimilation and dissimilation rates leads to biomass, diameter, and height growth in each layer. Leaf mass in each layer determines photoproduction, and light and photosynthesis conditions in lower layers. Transitions to higher layers, mortalities in each layer, and seed events are accounted for. The basic model simulates the development dynamics of a forest gap. Parallel computation of spatially distributed single gap models with stochastic mortalities and seed dispersal provides a simulation of the spatial dynamics of the tropical forest mosaic. Simulation results are provided with parameters corresponding to unlogged and logged Malaysian lowland dipterocarp forest.

References (26)

H. Bossel
Dynamics of forest dieback: systems analysis and simulation
Ecol. Modelling
(1986)
H. Bossel et al.
Generic simulation model of forest growth, carbon and nitrogen dynamics, and application to tropical acacia and European spruce
Ecol. Modelling
(1989)
H.T. Valentine
Tree-growth models: Derivations employing the pine-model theory
J. Theor. Biol.
(1985)
S. Appanah et al.
Are tropical rainforests renewable? An enquiry through modelling
J. Trop. For. Sci.
(1990)
P.S. Ashton et al.
Staggered flowering in the dipterocarpaceae: New insights into floral induction and the evaluation of mast fruiting in the aseasonal tropics
Am. Nat.
(1988)
E. Assmann
Waldertragskunde
M. Begon et al.
Population Ecology. A unified study of animals and plants
H. Bossel et al.
Simulation of tropical natural forest dynamics: System structure, dynamic process model, and simulation results
H. Bossel et al.
Dynamic simulation of natural tropical forest development for a forest management and planning system

F. Hallé et al.

Tropical trees and forests

M. Jacobs

The tropical rain forest. A first encounter

R. Kato et al.

Plant biomass and growth increment studies in Pasoh Forest

Malay. Nat. J.

(1978)

Cited by (70)

LoggingLab: An R package to simulate reduced-impact selective logging in tropical forests using forest inventory data
2024, Ecological Modelling
Even where Reduced-Impact Logging (RIL) practices are applied, selective logging causes substantial damage to tropical forests. To further reduce selective logging damage, the practices that cause the most damage need to be identified and alternatives tested. To this end, we developed the R package LoggingLab, a spatially-explicit and individual tree-based selective logging simulator and demonstrated its functions using data from French Guiana. LoggingLab explicitly simulates damage during each stage of the selective logging process taking into account topography and hydrography, which are main constraints on logging. Most LoggingLab parameters can be easily adjusted to a wide range of local contexts. LoggingLab can also be coupled with forest dynamics models to simulate the long- term effects of different selective logging scenarios.
CAMPUS-S – The model of ground layer vegetation populations in forest ecosystems and their contribution to the dynamics of carbon and nitrogen. I. Problem formulation and description of the model
2020, Ecological Modelling
Citation Excerpt :
According to the type of representation of space, time, and population elements, the mathematical models used in ecology can be divided into discrete and continuous. At present, a large number of models of forest ecosystems of various types have been developed, including EFIMOD (Komarov et al., 2003a; 2007), MixFor (Olchev et al., 2009), JABOVA (Botkin et al., 1972), FORMIX (Bossel and Krieger, 1991), iLand (Seidl et al., 2012), PICUS (Lexer, Hönninger, 2001), MARS (Baruth et al., 2006), and others. However, in none of the models of forest ecosystems known to the authors the contribution of the vegetation of the forest ground vegetation to the cycle of biophilic elements is explicitly taken into account.
The CAMPUS-S model was developed to simulate ground-layer vegetation populations in forest ecosystems and their contribution to carbon and nitrogen dynamics. The model uses several modeling techniques: matrix, l-systems and cellular automata modeling. The genetic species-specific features are defined for each ontogenetic stage (L-systems), while their implementation for certain individuals in the course of modeling depends on external conditions (cellular automata). To implement the variance of the transitions between ontogenetic stages, the matrix modeling method is used. The spatial resolution of the model is 1 cm² and the time step is one month. The model consists of two submodels: a submodel of the plant populations’ structure, and a submodel of carbon and nitrogen dynamics in plants and soil. The space in the CAMPUS-S model is heterogeneous. The map of the microrelief is dynamically loaded from an external source, according to which the temperature and the moisture of the soil are distributed across space during the simulation. To implement the feedback mechanism between plants and soil, integration with the Romul_Hum model is performed.
Forest dynamics in relation to meteorology and soil in the Gulf Coast of Mexico
2020, Science of the Total Environment
Citation Excerpt :
Forest dynamics has been studied and interpreted from various perspectives. Forest dynamics sometime refers to the forest deforestation and degradation (Giri et al., 2007) or the forest gap formation and closure (Bossel and Krieger, 1991; Yamamoto, 2000). Generally, the study of forest dynamics is focused on changes in forest structure and composition arising from natural or anthropogenic forces (Pretzsch, 2009).
Forest dynamics is complex, and the complexity could be a synthetic result of climate change. Specifically studying 11 forest type groups of the Gulf of Mexico coast region defined, we intended to explore and model the direct and indirect impacts of climate change on underlying forest dynamics. This study utilized normalized difference of vegetation index (NDVI) as a measurement indicator of forest dynamics, referring to the dynamics of canopy structure and phenology of forests, and for a given type of forests, seasonal and yearly NDVI values were applied to the quantification of its growth across the Gulf Coast. By utilizing geographically weighted regression (GWR) method, we related normalized difference vegetation index (NDVI) to precipitation, temperature, and silt and clay fractions in the soil. This study demonstrated an explanatory power of soil, besides the common macroclimate factors of precipitation, temperature, on explaining forest dynamics, which also revealed that the presence of spatiotemporal heterogeneity would affect model performance. Our results indicated that the model performance varied by forest type groups and seasons. The meteorology-soil model presented the best overall fit performance for White/Red/Jack Pine forests concerning R² (0.952), adjusted R² (0.905), Akaike information criterion (AIC, -1100) and residual sum of squares (RSS, 0.053) values. The comparative analysis of model performance also indicated that the meteorology-soil model has the best fit of data in summer. This study advanced the understanding of forests dynamics under conditions of climate change by highlighting the significance of soil, which is a significant confounding variable influencing forest activities but is often missed in forest-climate dynamics analysis.
Lessons learned from applying a forest gap model to understand ecosystem and carbon dynamics of complex tropical forests
2016, Ecological Modelling
Forests worldwide are threatened by various environmental and anthropogenic hazards, especially tropical forests. Knowledge on the impacts of these hazards on forest structure and dynamics has been compiled in empirical studies. However, the results of these studies are often not sufficient for long-term projections and extrapolations to large spatial scales especially for unprecedented environmental conditions, which require both the identification and understanding of key underlying processes. Forest models bridge this gap by incorporating multiple ecological processes in a dynamic framework (i.e. including a realistic model structure) and addressing the complexity of forest ecosystems. Here, we describe the evolution of the individual-based and process-based forest gap model FORMIND and its application to tropical forests. At its core, the model includes physiological processes on tree level (photosynthesis, respiration, tree growth, mortality, regeneration, competition). During the past two decades, FORMIND has been used to address various scientific questions arising from different forest types by continuously extending the model structure. The model applications thus provided understanding in three main aspects: (1) the grouping of single tree species into plant functional types is a successful approach to reduce complexity in vegetation models, (2) structural realism was necessary to analyze impacts of natural and anthropogenic disturbances such as logging, fragmentation, or drought, and (3) complex ecological processes such as carbon fluxes in tropical forests – starting from the individual tree level up to the entire forest ecosystem – can be explored as a function of forest structure, species composition and disturbance regime. Overall, this review shows how the evolution of long-term modelling projects not only provides scientific understanding of forest ecosystems, but also provides benefits for ecological theory and empirical study design.
Habitat complexity, environmental change and personality: A tropical perspective
2015, Behavioural Processes
Citation Excerpt :
Tropical rainforests are highly complex three-dimensional habitats (Rader and Krockenberger, 2006) that are generally commonly divided into five distinguishable layers (as per Bossel and Krieger, 1991): (1) seedling and shrubs; (2) saplings; (3) subcanopy; (4) canopy; and (5) emergents (trees that rise above the canopy, typically >36 m in height). Resource availability in these environments is largely driven by the vertical stratification of trees, as well as complex processes for pollination, seed dispersal and regeneration, which are influenced by light conditions and microclimates (Bossel and Krieger, 1991). A substantial fraction of light is absorbed at each of the five layers and, since light availability drives plant recruitment and limits seedling growth and survival (Kitajima, 1996), a higher abundance of resources is often found within the upper canopy and emergent layers (Rader and Krockenberger, 2006).
Tropical rainforests are species-rich, complex ecosystems. They are increasingly being negatively affected by anthropogenic activity, which is rapidly and unpredictably altering their structure and complexity. These changes in habitat state may expose tropical animals to novel and unpredictable conditions, potentially increasing their extinction risk. However, an animal's ability to cope with environmental change may be linked to its personality. While numerous studies have investigated environmental influences on animal personalities, few are focused on tropical species. In this review, we consider how behavioural syndromes in tropical species might facilitate coping under, and adapting to, increasing disturbance. Given the complexity of tropical rainforests, we first discuss how habitat complexity influences personality traits and physiological stress in general. We then explore the ecological and evolutionary implications of personality in the tropics in the context of behavioural flexibility, range expansion and speciation. Finally, we discuss the impact that anthropogenic environmental change may have on the ecological integrity of tropical rainforests, positing scenarios for species persistence. Maintaining tropical rainforest complexity is crucial for driving behavioural flexibility and personality type, both of which are likely to be key factors facilitating long term persistence in disturbed habitats.
Modelling mixed forest stands: Methodological challenges and approaches
2015, Developments in Environmental Modelling
Citation Excerpt :
Calibration is performed with meteorological data. The original purpose of the FORMIX model was to represent the growth of tropical natural forest before and after logging (Bossel and Krieger, 1991). Through simulations of different logging scenarios with different cycles and intensities, the impact of different logging strategies can be predicted and forest management concepts can be evaluated (Ditzer et al., 2000; Huth and Ditzer, 2001; Tietjen and Huth, 2006; Huth and Tietejen, 2007).
Mixedwoods are increasingly becoming one of the major focuses in ecological modelling studies, due to their higher productivity and resiliency compared to monocultures. In this chapter, we provide a review on the state-of-the-art of models simulating mixedwood stands. After reviewing almost 400 peer-reviewed publications, it is clear that most modelling exercises are done for explanatory uses rather than for decision-making support tools. Among them, North America and central Europe are the main model development centers. It is also clear that more model portability is needed, as even the most widely distributed models tend to be used in the same region where they were originally developed. Finally, the four models used more widely for simulating mixed forests stands were identified: two have been applied mostly in boreal/temperate ecosystems (FORECAST and SILVA), whereas the other two have been used in tropical/subtropical environments (FORMIND and FORMIX). A comparison of their different modelling approaches is provided, and multimodel exercises are suggested as a way to both compare model performance and also to reduce simulation uncertainty due to model selection.

View all citing articles on Scopus

View full text

Simulation model of natural tropical forest dynamics

Abstract

Ecol. Modelling

Ecol. Modelling

J. Theor. Biol.

Are tropical rainforests renewable? An enquiry through modelling

J. Trop. For. Sci.

Staggered flowering in the dipterocarpaceae: New insights into floral induction and the evaluation of mast fruiting in the aseasonal tropics

Am. Nat.

Waldertragskunde

Population Ecology. A unified study of animals and plants

Simulation of tropical natural forest dynamics: System structure, dynamic process model, and simulation results

Dynamic simulation of natural tropical forest development for a forest management and planning system

Tropical trees and forests

The tropical rain forest. A first encounter

Plant biomass and growth increment studies in Pasoh Forest

Malay. Nat. J.