0657-B1

Simulation of Logging Impact in a Venezuelan Tropical Forest

H. Ramírez-Angulo and A. Torres-Lezama^[1]

Abstract

Tropical forest management has been strongly criticized over the last three decades, mainly because of the impact of selective logging on system sustainability. In order to analyse the effect of cutting cycle and timber harvesting in a humid deciduous forest of the Venezuelan western plains, an individual-based model (ZELIG) was applied. We used a data set with 36 years of measurement to parameterize the models. Two selective logging options were simulated: low impact and conventional logging. For both cases we harvested the total merchantable stand and considered vegetation damage caused by logging methods with regard to the minimum-felling diameter legally established for the country. Forest recovery was studied for 30-year (cutting cycle) and 60-year periods (rotation). We also calculated the benefit/cost relationship. The results show that the forest does not recovery in a 30-year period and conventional logging requires more than 60 years. However, from an economic point of view, the option with the most impact was more profitable in the short term.

Introduction

Forest management involves a planning process where objectives are defined, information is gathered, options are selected and actions are described. These actions are taken and evaluated with results that contribute to feed the process back and determine future effects of current decisions; therefore, good planning is required at the management unit level and its environment (Sabogal 1998).

A particular area of interest is the long-term impact of logging. Appropriate planning of these activities is a key element for successful introduction of sustainable utilization oriented towards reducing general costs and stand and soil damage, and to increase productivity by means of more efficient operations to simultaneously achieve higher value products (Heinrich 1998).

Silvicultural practices that are more sensible to conservation needs require changes in approach from cutting regimes associated to simple tree harvesting to the perpetuation of stand structure and species composition (Lindenmayer 1999). In this sense, the cutting cycle is one of the most important decisions. Very short cycles ignore capacity of forest recovery and degrade it (Huth and Ditzer 2001).

In order to develop practices for sustainable management it is necessary to know the processes influenced by management and to evaluate potential long-term consequences. Tree harvesting may affect forest stand growth through effects associated with soil compaction, microclimate changes and differences in nutrient reserves. The complexity of tropical forest ecosystems, however, makes experimentation difficult. Thus, models are a useful tool, because the impact mentioned above can be analyzed and evaluated through simulation of different logging options.

Simulation models have been scarcely applied to the neotropical forests (but see Doyle 1981; Ramírez 1995 and Campos et al. 1998). This relative shortage may be explained, among other reasons, by: a) there are few data records in the tropical forests related to growth and other variables that allow model formulation; and b) these ecosystems are more complex than temperate forests (Liu and Ashton 1995).

This complexity is expressed by species diversity as well as functional biology. Species exhibit many differences in growth form, solar radiation requirements and reproduction patterns (Field and Vásquez-Yanes 1993). Therefore, it is convenient to simplify the simulation process through the use of functional groups defined according to appropriate criteria (Acevedo et al. 1996).

In contrast to many tropical situations, where information is practically lacking, in Venezuela we have a network of 67 permanent plots with a record period close to forty years. These plots were established, by the Silviculture Institute of the Universidad de Los Andes, in nine life zones under different disturbance degrees (from primary forest to severely disturbed forest).

The objectives of this work are to evaluate the ecological and economics impact of current cutting cycles and logging methods in a humid deciduous seasonal Venezuelan forest, specifically in the University Forest "El Caimital", by means of an individual-based model, the ZELIG model (Urban 1993).

Materials and Methods

Study site

This remnant forest is located to the northeast of Barinas State (8º40´ N, 70º13´W), at an elevation of 170 m above sea level. The relief is flat with small depressions which are flooded in the rainy season. Terrain slope does not exceed 3%. The climate belongs to the seasonal tropical, with a mean annual temperature of 26.7ºC and an average annual rainfall of 1,590 mm. Precipitation is strongly seasonal, characterized by a very humid period (May-November) and a very dry period (December-April).

The life zone is classified as tropical dry forest transition to humid, after Holdridge (1967). The vegetation is typical of a tropical moist deciduous forest. Apparently, "El Caimital" is a secondary forest established over former agricultural land. This is supported by the abundance of coffee plants. In the 1940's and 1950's the forest was selectively logged. The best trees of the major commercial timbers were harvested (Cedrela odorata and Swietenia macrophylla) (Finol 1964).

Species selection

In a previous work we used a data set with over 35 years of measurement from the University Forest "El Caimital" to parameterize the model (Ramírez 1995 and Ramírez et al. 1997). Due to the large number of tree species (64), and the difficulty to estimate parameter values for such a number, we selected the top 14 species above a threshold of Value Index (VI = 75%). This index is calculated dividing relative basal area by relative stand density. The remaining species were classified in six ecological groups, according to their light requirements and maximum tree height.

Model description

ZELIG is derived from the FORET model (Shugart and West 1977). It is a basic model, easy to implement and developed with versatility in the application. It is based upon two conceptual sub-models, the first emphasizes the solar radiation regime and the second the soil water balance (Ramírez 1995).

The required parameters to run the model are growth rates, height-diameter alometric relations, and relative growth constraints regarding solar radiation, drought, nutrients, temperature and species reproductive capacity. In this work we used the parameterization carried out by Ramírez (1995) and described in Ramírez et al. (1997). The values of the parameters were estimated, and a good adjustment was reached after contrasting the model results with the field data.

Logging options

Based on this parameterization the model was applied to predict selective logging effects on forest dynamics. Two selective logging options were simulated with regard to the minimum felling diameters (mfd) legally established in Venezuela. Approximately 30 trees per hectare were harvested in both options; this resembles common logging practices in the country. In both options we considered the vegetation damage caused by logging.

Option 1: we assumed low impact logging and that the total merchantable stand was harvested. The mfd was of 60 cm dbh for Bombacopsis quinata and Cedrela odorata, and 40 cm dbh for the remaining species.

Option 2: the forest stand was harvested as above, but a conventional logging was assumed (see Kammesheidt et al. 2001).

Furthermore, two cutting cycles were considered: the traditional cutting cycle in the country (30 years) and the rotation (60 years. We determined the degree of ecological sustainability based upon the level of remaining forest recovery with respect to the initial stand values of the non logged forest. In order to determine the benefit/cost relationship, the harvest total value was calculated regarding the standing timber prices. Logging damage was estimated as a function of the harvested tree number and the logging option. Several investigations reveal that damage to the remaining forest by logging operations oscillate between 20 y 80%. As references we used the evaluations of Uhl et al. (1991), Verissimo et al. (1992), Huth and Ditzer (2001). Based on "El Caimital" historical references we assumed a forest age of ca. 260 years. The current conditions were simulated by Ramírez et al. (1997) and Ramírez and Torres (1998); starting from those conditions the proposed logging options were simulated.

For each logging option the following parameters were evaluated: mean stand density, mean basal area and total timber biomass.

Results

In the initial harvest about 30% of the basal area (BA) was removed. This agrees with Dawkins (1963) and de Graff (1986) prescriptions to achieve a satisfactory forest response.

In the simulation, for the intervention year, the BA value reached 27.57 m² ha^-1 (included the category 0 - 10 cm dbh), very close to the value of 25 m² ha^-1 found by Veillon (1985) for the same forest (starting at 10 cm dbh). For both options the forest recovery degree was similar; for Option 1 it took 100 years, whereas for Option 2 120 years were needed (Figure 1). Stand biomass simulated at the time of harvest reached 314.82 Mg ha^-1, for individuals with dbh > 10 cm. Trends for this variable were similar for both logging options, with a recovery period of 160 and 180 years, respectively (Figure 2).

Figure 1: Basal area simulation for 500 years, in the University Forest "El Caimital", Venezuela.

Figure 2. Simulation of woody biomass for 500 years, in the University Forest "El Caimital", Venezuela

Timber supply is expected to be lower for the next cutting cycle (30 years), with conventional logging yielding only 56% of the volume harvested in the first operation (Figure 3). For a 60 years felling cycle this logging method does not allow to reach yet the initial values.

Figure 3. Harvested volume in different logging options and intervention years

Regarding the economic analysis of the different logging combinations, the benefit/cost relationship showed that in all cases Option 2 was the most favourable, probably because low impact logging was more costly than the traditional practice (Table 1). All the combinations are profitable and differences between combinations are small. Nevertheless, if we look closely at the merchantable biomass in each option we find that Option 1 is the best, from an ecological point of view.

Table 1. Benefit/cost relationship for the simulated timber exploitation combinations at the University Forest "El Caimital", Venezuela.

Discussion

Logging damages are very high, particularly if we consider that only in the low impact option a volume similar to the original one is harvested after 60 years. The major differences are observed for conventional logging where even after 100 years this value is not similar to the low impact logging. Regarding stand biomass, the results are similar to those found by Uhl and Saldarriaga (1986), at San Carlos de Río Negro (Venezuela), concerning a linear increment of biomass at the beginning of forest recovery, increasing from 15 (at two years) to 150 Mg ha^-1 at 60 years. This represents a 59% of the mature forest biomass. They estimate that an abandoned field needs from 140 to 200 years to reach a biomass similar to the mature undisturbed forest. This period could even be longer depending upon the degree of disturbance; in severe interventions more than 1000 years may be required to reach the values of the natural forest. On the other hand, we found that planning of the exploitation although may be more costly it allows a faster forest recovery.

The use of current cutting cycles in tropical forests has been criticized for several authors (e.g. Kammesheidt et al. 1999, Huth and Ditzer, 2001), because apparently so short periods do not allow forest recovery; these results seem to confirm this statement. In the studied forest at least twice as much time than the cutting cycle established in Venezuela is required under a low impact logging method.

All these findings allow us to say that in order to implement a sound forest management, without hampering profitability, some measures must be taken to assure the achievement of this objective. These measures are fully justified if we consider that tropical forests, in addition to producing a series of timber resources, offer environmental services. In consequence, when forests are either destroyed or degraded, the losses go far beyond the economic aspect.

Besides, the inclusion of environmental assessment in the economical analyses, measures could be taken to favour products coming from forests managed with sustainability criteria. For instance, to assign better prices to the timber from such forests, to promote local manufacture of high aggregated value products, to pay lower taxes and to develop incentives for product exportation.

Conclusions

Low impact logging (Option 1) allows a faster ecosystem recovery, although full recovery takes at least 100 years. From the economic point of view, however, this is not the best option. Taking into account that the ecological aspect is very important, even though the economical results are satisfactory for both options, it is essential to manage the forest in order to assure the maintenance of the original ecological conditions as much as possible. This could be achieved, among other strategies, through lower intensities of exploitation and sound planning of the logging activities. Similarly, we found that cutting cycles established in the management plans in Venezuela are very short and could compromise resource sustainability, since the forest needs longer to recover from the intervention.

Our results show that economic and ecological aspects may produce conflicting results; therefore, it is essential to apply different economical approaches, in order to assure sustainable forest management without compromising profitability. This implies the necessity of recognizing that a trade-off must be reached between timber production and other forest values in order to assure good management and the welfare of society. Nowadays, forest management more than ever has a singular importance and represents the most realistic way of assuring that goods and services from the forest are made available for mankind and supplying habitats to the flora and fauna that constitute such forest ecosystems.

Simulation models have shown to be a useful tool for the analysis of different forest management strategies. The Zelig model may be utilized to estimate the effects of different logging options. For the forest 'El Caimital´, according to the indicators found for impact and timber yield, at least a 60 years cutting cycle is required as well as a change of the conventional logging practices to low impact techniques. It is important to point out, however, that even applying such techniques, shorter felling cycles impact the forest in a negative and considerable way.

The logging options evaluated in this paper will be further explored in the search for better solutions to the forest management problem. In the transition towards sustainable forest development a trade-off must be found between economical short-term profits and long-term development.

References

ACEVEDO, M.F., D.L. URBAN and H. SHUGART. 1996.Models of forest dynamics based on roles of tree species. Ecological Modeling 87:267-284.

CAMPOS ARCE, J.J., F. FINEGAN, M. CAMACHO and D. QUIRóS. 1998. Sostenibilidad del manejo de bosques naturales: resultados sobre la factibilidad ecológica y económica en Costa Rica. In: Actas I Congreso Latinoamericano IUFRO (Tema 3).

DAWKINS, H.C. 1963. The productivity of tropical high-forest trees and their reaction to controllable environment. PhD dissertation, University of Oxford, 155 p.

DOYLE, T. W. 1981. The role of disturbance in the gap dynamics of a montane rainforest: an application of a tropical forest succession model. In: West, D. C.; Shugart, H. H. and D. B. Botkin (eds.) Forest Succession concepts and applications. Springer-Verlag. New York, p.56-73.

FIELD, C. and C. VASQUEZ-YANES. 1993.Species of the genus Piper provide a model to study how plants can grow in different kinds of rainforest habitats. Interciencia 18:230-236.

FINOL, H. 1964. Estudio silvicultural de algunas especies comerciales en el Bosque Universitario "El Caimital", Edo. Barinas. Revista Forestal Venezolana 7:17-63,

GRAAF, de N.R. 1986. A silvicultural system for natural regeneration of tropical rain forest in Suriname. Agricultural University of Wageningen, 250 p.

HEINRICH, R. 1998. Aprovechamiento ambientalmente apropiado para mantener los bosques tropicales. In: BOLFOR; CIFOR; IUFRO. 1998. Memorias del Simposio Internacional sobre Posibilidades para el Manejo Forestal Sostenible en América Tropical. Editora El País. Santa Cruz, Bolivia, p.5-15.

HOLDRIDGE, L. 1967. Life zone ecology. Tropical Science Center. San José, Costa Rica, 206 p.

HUTH,A. and T. DITZER. 2001.Long-term impacts of logging in a tropical rain forest - a simulation study. Forest Ecology and Management 142:33-51.

KAMMESHEIDT, L. 1998.Stand structure and spatial pattern of commercial species in logged and unlogged Venezuelan forest. For. Ecol. Manag. 109:163-174.

KAMMESHEIDT, L., A. TORRES-LEZAMA and W. FRANCO. 1999. Impacto de la Explotación Selectiva sobre la diversidad de especies arbóreas en un bosque de los llanos occidentales venezolanos. Revista Forestal Venezolana 43:59-67.

KAMMESHEIDT, L., A. TORRES-LEZAMA, W. FRANCO and M. PLONCZAK. 2001. Logging history of the western Venezuelan plains and the prospects for a sustainable forest management. Forest Ecology and Management. 148:1-20

LINDENMAYER, D.B. 1999. Future directions for biodiversity conservation in managed forests: indicator species, impact studies and monitoring programs. Forest Ecology and Management 15:277-287.

LIU, J. and P.S. ASHTON. 1995.Individual-based simulation models for forest succession and management. Forest Ecology and Management 73:157-175.

ORGANIZACIÓN PARA LA PREINVERSIÓN EN AMERICA LATINA Y EL CARIBE (OPALC). 1995. Impacto de la política macroeconómica en los recursos forestales. INEFAN. Quito, 64 p.

RAMÍREZ, H., 1995. Aplicación de un modelo de simulación de base individual a la dinámica del bosque tropical: un caso de los llanos venezolanos. Tesis M.Sc. Centro de Estudios Forestales de Postgrado. Facultad de Ciencias Forestales y Ambientales. Universidad de Los Andes. Mérida, Venezuela, 64p.

RAMÍREZ, H., A. TORRES-LEZAMA and M. ACEVEDO. 1997. Simulación de la dinámica de grupos de especies vegetales en un bosque de los llanos occidentales venezolanos. Ecotrópicos 10:9-20.

RAMÍREZ, H. and A. TORRES-LEZAMA. 1998. Parametrización de un modelo de simulación de base individual como base para el manejo de un bosque tropical venezolano. In: BOLFOR; CIFOR; IUFRO. 1998. Memorias del Simposio Internacional sobre Posibilidades para el Manejo Forestal Sostenible en América Tropical. Editora El País. Santa Cruz, Bolivia. p. 224-233.

SABOGAL, C. 1998. Planes de manejo forestal y necesidades de información para el manejo operacional. In: BOLFOR; CIFOR; IUFRO. 1998. Memorias del Simposio Internacional sobre Posibilidades para el Manejo Forestal Sostenible en América Tropical. Editora El País. Santa Cruz, Bolivia. p.135-147.

SHUGART, H.H. and D.C. WEST. 1977.Development of an appalachian deciduous forest succession model and its application to assessment of the impact of the chestnut blight. Journal of Environmental Management 5:161-179.

SILVA, R., 1974. Tabla de volumen para árboles en pie del bosque tropófito de los llanos occidentales venezolanos. Escuela de Ingeniería Forestal. ULA. Mérida, Venezuela, s.p.

UHL C.; SALDARRIAGA, J. 1986.Fragilidad de la pluviselva amazónica. Investigación y Ciencia 121:72-81.

UHL, C.; VERíSSIMO, A.; MATTOS, M.; BRANDINO, Z.; GUIMARAES, I.C. 1991.Social, economic and ecological consequences of selective logging in an Amazon frontier: The case of Tailandia. Forest Ecology and Management 46:243-173.

URBAN, D.L. 1993. A User's Guide to ZELIG version 2. Department of Forest Sciences. Colorado State University. Fort Collins, USA, 73p.

VEILLON J.P. 1985. El crecimiento de algunos bosques naturales de Venezuela en relación con los parámetros del medio ambiente. Revista Forestal Venezolana 29:5-120.

VERíSSIMO, A.; BARRETO, P.; MATTOS, M.; TARIFA, R.; UHL, C. 1985. Logging impacts and prospects for sustainable forest management in an old amazonian frontier: the case of Paragominas. Forest Ecology and Management 55:169-199.