Modelling anthropogenic impacts on the growth of tropical rain forests - using an individual-oriented forest growth model for analyses of logging and fragmentation in three case studies Peter K¨ohler Der Andere Verlag, Osnabr¨uck, Germany, 2000 ISBN 3-934366-99-6 Zugl.: Kassel, Univ., Diss., 2000 Cover: Dawn in Danum Valley, Sabah (Borneo), Malaysia October 1997 taken by P. K¨ohler Modelling anthropogenic impacts on the growth of tropical rain forests - using an individual-oriented forest growth model for analyses of logging and fragmentation in three case studies Modellierung anthropogener Einflusse¨ auf das Wachstum tropischer Regenw¨alder - Analyse von Holznutzung und Fragmentierung in drei Fallstudien unter Verwendung eines individuen-orientierten Waldwachstumsmodells Inaugural-Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.) im Fachbereich Physik der Universit¨at Gesamthochschule Kassel vorgelegt von Dipl.-Phys. Peter K¨ohler aus Kassel Kassel, den 01.11.2000 Als Dissertation vom Fachbereich Physik angenommen am 01.11.2000 Erster Gutachter: Prof. Dr. Hartmut Bossel Zweiter Gutachter: Prof. Dr. Burkhart Fricke Drittes Mitglied der Promotionskommission: Dr. habil. Andreas Huth Viertes Mitglied der Promotionskommission: Prof. Dr. Albrecht Goldmann Tag der m¨undlichen Pr¨ufung: 01.11.2000 ”If everyone complains bitterness, then to whom is the world?” Zainal Abidin Jaafar Overview For answering questions concerning anthropogeneous impacts on tropical forest develop- ment the individual-oriented and process-based forest growth model Formind2.0 was developed. It simulates the spatio-temporal dynamics of uneven-aged mixed forest stands in areas of one hectare to several km2. The model describes forest dynamics as a mosaic of interacting forest patches of 20 m2×20 m2 in size. Within these patches trees are not spatial-explicitly distributed, and thus all compete for light and space following the gap model approach. Tree species diversity is aggregated into 5-20 plant functional types (PFT) on the basis of species maximum tree height and successional status. The carbon balance of each individual tree incl. photosynthesis and respiration is modelled explicitly. Thus, we can match measured diameter increment for different PFT, size and light condi- tions accurately. Allometric relationships connect above-ground biomass, stem diameter, tree height and crown dimensions. Beside increasing mortality through self-thinning in dense plots one of the main processes of mortality is gap creation by falling of large trees. This process as well as seed dispersal from mature trees interlinks neighbouring plots with each other. The model was parametrised for three different sites in South-East Asia and south- America: Sabah (Malaysia), Venezuela, and French Guiana. Model accuracy is tested with growth data from permanent sampling plots in Sabah. Sensitivity of various result variables on variation of most parameter values is tested and gives important insights into general model behaviour. Two examples of anthropogeneous impacts on tropical forest dynamics are management practise and fragmentation, both of major concern. Following applications are performed: Growth and yield of Venezuelan rain forest under various logging methods, intensities and cycles are analysed for their sustainability. Effects of logging (methods and cycles), fragmentation and recruitment assumptions on forest dynamics in Sabah are discussed. Finally, fragmentation impacts on mortality and recruitment are simulated and their effects on forest dynamic and biomass loss are evaluated for a forest site in French Guiana. Keywords: abandoned land; basal area; dipterocarp forest; edge effects; forest growth model; Formind; fragmentation; French Guiana; functional groups; individual-oriented model; logging impacts; logging scenarios; Malaysia; maximum height; model; mortal- ity; plant functional types; rain forest; recruitment; secondary succession; simulation; successional status; sustainable timber harvest; tropical rain forest. Contents 1 Introduction 15 2 An introduction to tropical rain forests 21 3 Concepts for the aggregation of tropical tree species into functional types and the application to Sabah’s lowland rain forests 25 4 The model Formind2.0 35 5 Comparison of measured and simulated growth on permanent plots in Sabah’s rain forests 51 6 Sustainable timber harvesting in Venezuela: a modelling approach 65 7 The effects of logging, fragmentation and recruitment on growth of di- pterocarp forest 85 8 Long-term response of tropical rain forests to the effects of fragmenta- tion: a simulation study 111 Summary 133 Zusammenfassung 139 Bibliography 145 AInventory data 165 B Lists of tree species 171 Danksagung 215 Account Chapters of this thesis have been published as follows: Chapter 3: K¨ohler, P., Ditzer, T., & Huth, A. 2000b. Concepts for the aggregation of tropical tree species into functional types and the application on Sabah’s lowland rain forests. Journal of Tropical Ecology, 16(4), 591-602. Chapter 5: K¨ohler, P., Ditzer, T., Ong, R. C., & Huth, A. 2001. Comparison of mea- sured and simulated growth on permanent plots in Sabah’s rain forests. Forest Ecology and Management, 142(1-3), in press. Chapter 6: Kammesheidt, L., K¨ohler, P., & Huth, A. 2000. Sustainable timber harvest- ing in Venezuela: a modelling approach. Journal of Applied Ecology, in press. Chapter 7: K¨ohler, P., Ditzer, T., & Huth, A. 2000c. The effects of logging, fragmen- tation and recruitment on growth of dipterocarp forest. Journal of Ecology, submitted. Chapter 8: K¨ohler, P., Chave, J., Riera, B., & Huth, A. 2000a. Long-term response of tropical rain forests to the effects of fragmentation: a simulation study. To be submitted. Other publications by the author related to the topics of this thesis: Ditzer, T., Glauner, R., F¨orster, M., K¨ohler, P., & Huth, A. 2000. The process-based stand growth model FORMIX3-Q applied in a GIS-environment for growth and yield analysis in a tropical rain forest. Tree Physiology, 20, 367–381. K¨ohler, P. 1996. Ein individuenbasiertes Wachstumsmodell zur Simulation tropischer Regenw¨alder. Diploma thesis, University of Kassel, Germany. K¨ohler, P. 1997. An individual based rain forest model: Formind. in Hahn-Schilling, B. (editor), Forest management with growth models. Malaysian-German Technical Cooperation Project, Forest Department of Sarawak, Malaysia, Kuching, Malaysia. K¨ohler, P. 1998. Parameter research for the tropical rain forest growth model FORMIX4. Research report P9801, Center for Environmental Systems Research, University of Kassel, Germany. K¨ohler, P. & Huth, A. 1998a. The effect of tree species grouping in tropical rain for- est modelling - Simulation with the individual based model Formind. Ecological Modelling, 109, 301–321. K¨ohler, P. & Huth, A. 1998b. An individual based rain forest model - concepts and sim- ulation results. In: Kastner-Maresch, A., Kurth, W., Sonntag, M., & Breckling, B. (editors), Individual-based structural and functional models in ecology, number 52 in Bayreuther Forum Okologie.¨ Bayreuther Institut f¨ur Terrestrische Okosystem-¨ forschung, Bayreuth, 35–51. Chapter 1 Introduction Introduction Goldammer 1999; Laurance & Fearnside 1999; Nepstad et al. 1999; Hashimotio et al. The use of natural resources change our en- 2000). vironment directly and indirectly through With 36000 000 km 2 of forests, cov- effects which are not fully understood so far. ering a quarter of the total land sur- Climate change and an increase in mean face on earth, beside the oceans forests global temperature, the amount of carbon- are the biggest ecosystems on our planet. dioxide in the atmosphere, or rising sea lev- About 475 to 825 billion tons of carbon els are some examples of occurring effects are bound in the forests and thus they are (Fan et al. 1998; Chavez et al. 1999; Malhi & the biggest above-ground carbon storages Grace 2000). These anthropogeneous influ- (Murphy 1975; Enquete-Kommission 1990; ences will change our environment for cen- Grace et al. 1995; Fan et al. 1998; Pren- turies. Plants might react adaptivly to their tice & Lloyd 1998; Alexandrov et al. 1999b, changing environment (Pastor & Post 1988; 1999b). A further reduction of woodland Friend 1997; Cao & Woodward 1998; Tian and, following this, an increasing release of et al. 1998; DeLucia et al. 1999; Pounds carbon in the form of carbondioxide would et al. 1999; Stil et al. 1999; Hashimotio et al. certainly intensify climate changing effects. 2000). Huge ecosystems like forests might Currently, annual release of carbon and its buffer changes, caused for example by ex- input in the atmosphere are estimated at traction of timber woods over a long period. seven billion tons. About 20 % of the release But if natural catastrophes occur in ecosys- is caused by global deforestation (Enquete- tems, which have already been weakened, Kommission 1994; Houghton et al. 2000). damage is more dramatic than ever thought There are various reasons which make before (Phillips & Gentry 1994; Laurance forests worth protecting and sustainable et al. 1997; Phillips et al. 1998; Peres 1999; management desirable. Forests produce Gascon et al. 2000). Thus, forest fires in timber, firewood and food, act as lo- the Amazonian rain forest and the Indo- cal climate regulator, prevent erosion, and Malayan archipelago in the years 1997/1998 are important water storages. Addition- spread very fast. El Ni˜no, the Great South- ally, tropical rain forests are remarkable