JAPANESE JOURNAL OF ECOLOGY

Show abstractHide abstract

The degradation of tropical forests is a major environmental issue related to global warming and biodiversity conservation, and therefore deserves the attention of various international and non-governmental organizations. Commercial timber extraction is believed to be the main driver of degradation in tropical forests. Currently, production forests cover more area than protected forests. In addition, most of the remaining forests are designated for timber production by national forest services. Therefore, the forests remaining after selective logging will play important roles in maintaining and upgrading ecosystem services such as maintenance of carbon stocks and biodiversity in degraded tropical forests. Degraded forest structure and species composition caused by selective logging may be recovered through secondary forest succession. However, it remains unclear how long it takes tropical forests to recover from selective logging. This paper addresses this question from the viewpoints of forest structure, forest illumination, demographic parameters, biomass and tree biodiversity, using data derived from the Pasoh Forest Reserve, Malaysia where selective logging was performed in the 1950s. After selective logging the forest structure, forest illumination, and demographic parameters of this forest were still statistically distinguishable from unlogged forests even 50 years after a logging operation. Forest biomass had almost recovered to the pre-logging level 50 years after the logging operation concluded. However, after logging the biodiversity of the logged forest was quite different from that of the unlogged forest, suggesting that it will take quite a long time—more than 60 years—for biodiversity to recover after logging.

View full abstract

Download PDF (904K)

When constructing state-space models, variables in the models are explicitly divided into observable and non-observable variables. Observation errors are appropriately handled in state-space models. Consequently, the parameters of models are estimated with smaller biases. State-space models further enable us to divide the dynamics into the mainstream and tributary flows. We illustrate the effectiveness of state-space models using data for 50 years for two insect pests living in paddy fields: Chilo suppressalis (Walker) and Nephotettix cincticeps (Uhler). We discuss several problems that should be considered when applying state-space models to the dynamics of insect populations, including the initialization problem, detrending method, and model evaluation.

View full abstract

In this study, I review data and models used to estimate deer population dynamics, including state-space models. One common approach has involved cohort analysis and a Sex-Age-Kill (SAK) model, applied to data on the number of captured deer, broken down by sex and age. However, these methods are not spatially explicit, and the cost and effort involved in collecting suitable data are prohibitive. Thus, deer density indices that correlate with spatially explicit deer density are often used for population control—but few indices directly show deer density. Therefore, several state-space models have been developed to estimate deer density from the indices, and some studies have succeeded in estimating the spatio-temporal variation of deer density over a 5-km-square mesh unit. Overall, state-space models appear to be a suitable framework with which to estimate deer population dynamics.

View full abstract

CO₂ is used for carboxylation in chloroplast stroma, and diffuses in through stomata, intercellular air spaces, the cell wall, plasma membranes, and the chloroplast envelope. Stomatal resistance (R_s) and mesophyll resistance (R_m) occur in this pathway, the inverse of which are stomatal conductance (g_s) and mesophyll conductance (g_m), respectively, metrics used as indicators of conductivity. g_m is comparable to g_s and therefore important for photosynthesis. There are three ways to measureg_m: the chlorophyll fluorescence, curve-fitting, and stable carbon isotope methods. Here, we explain these methods and critical points in their respective measurements. The most reliable is the stable carbon isotope method, but parameter setting is difficult because small differences in parameters can result in large differences in g_m. We determined which parameters cause serious errors in estimating gm changes in response to CO₂. Errors in the CO₂ compensation point in the absence of mitochondrial respiration (Γ*) and mitochondrial respiration in light (R_d) exerted the most significant changes in gm at low CO₂ concentrations. Although these parameters are usually adopted from previous studies on limited numbers of species, they should be experimentally determined. Although calculating gm will still not be simple, precise measurements are required for understanding CO₂ diffusion in leaves.

View full abstract

Register with J-STAGE for free!