We developed a height-diameter curve estimation model for even-aged hinoki (Chamaecyparis obtusa Sieb.) plantations to develop a height-diameter curve that could be applied to any stands at any stand age using diameter distribution and stand characteristics. We used generalized allometric equation for the height-diameter curve equation and assumed that allometric coefficient was equal to 1 (constant) and that maximum height was the upper-asymptote of the site index curve. We performed stepwise multiple regression analysis to derive the remaining parameter related to the slope of curves (referred to here as the 'shape parameter'). For the multiple regression analysis, we used stand age, density, mean basal area and site index as explanatory variables. To validate the model, we estimated the mean tree height of each DBH class and compared the predicted and observed heights. Furthermore, we predicted stand volume by combining the height-diameter curve estimation model and a diameter distribution prediction system developed in a previous study, and compared the predicted and observed data. From the regression analysis, we obtained a linear equation for the shape parameter as a function of stand age, density and mean basal area. The results of model validation indicate that the height-diameter curve estimation model is capable of estimating mean tree height for DBH classes with less bias, and would be useful for diameter distribution-based stand volume prediction. The results of stand volume prediction indicated that when initial stand age was greater than 30 years, the prediction error of both basal area and stand volume were small, while they were over-predicted when initial stand age was less than 30. These results suggest that the predictive error for stand volume was dependent upon the diameter distribution prediction system.

View full abstract

Forest stands must be evaluated before making zoning or harvest plans. This paper evaluated the net income of managed sugi (Japanese cedar, Cryptomeria japonica D. DON) plantations in private forests in a snowy region and classified sugi plantations according to the probability of harvesting, using a geographic information system (GIS), in a case study of private forests in Sanpoku Town, Niigata Prefecture. The town includes both low coastal and mountainous inland areas and the snow depth varies markedly. Two transfer points were assumed: a timber market and a local wood-related complex. First, log production was predicted according to site quality, using an existing stand density management diagram, as well as height growth curves to obtain log volumes according to log length and top diameter class. Gross income was estimated from log prices and the predicted log production according to site quality. Then, the total cost from planting through to logging was estimated for standard conditions of productivity, necessary workforce and wages, using four factors: site quality, yarding distance, the deepest snow depth and slope angle. The cost of plantation forest management included the regeneration cost for establishing new plantations and the logging cost for final cutting. Forest net income was evaluated for three cutting ages: short, intermediate and long rotation. Finally, sugi plantations were classified according to the probability of harvest and the area of each class was obtained using GIS on the assumption of present, 10% higher and 10% lower log prices. Site quality was the more important factor in sugi plantation management; it was more important than yarding distance or the deepest snow depth. Prices 10% lower than present prices on the Niigata Timber Market were the minimum necessary to maintain timber production from sugi plantations in private forests, given present costs and subsidies.

View full abstract

Farmers' behavior and their socio-economic conditions were studied focusing on the agricultural land use patterns and production, as well as their basic needs. An interviewing survey of farmers living in some selected forest village projects was conducted using participatory rural appraisal tools and techniques. Findings showed that field crops was found to be the major agricultural land use in all sizes of landholding, especially in the North and Northeast. On the other hand, most of the farmers were more likely to grow field crops at first thought, regardless of the size of landholding. Unlikely, paddy rice (followed by field crops) was the major agricultural land use in the Central region, whereas rubber plantation was the major one in the South. The difference in agricultural land use by regions can be attributed to the geographical, socio-economic, and environmental characteristics of each region. The species grown in degraded forests by the farmers were not much different between the North and the Northeast regions, except more fruit trees were planted in the North. Apart from corn and cassava, cotton and sugarcane were found to be the main cash crops in the Central region, and some fast growing tree species like Eucalyptus sp. could be generally planted in many areas. In the South, rambutan, coconut and durian were the most popular fruit trees of the region usually grown in areas of small landholding sizes. Regarding the basic needs of farmers, the most wanted ones were income; followed by land and water respectively. However, the three important basic needs mentioned here were together at all times when farmers' need was assessed.

View full abstract

To confirm that the vertical distribution of stem cross-sectional area increment (CSAI) for Hinoki cypress can be described by two straight lines derived from three tree attributes: sunny crown length (CL), stem volume increment (SVI), and CSAI at the base of the sunny crown (CSAI_(CL)), two model were constructed and compared. One is the unrestricted model for describing the vertical distribution of CSAI by two straight lines (CSAI_(z)=az and CSAI_(z)=b+cz where CSAI_(z) is the CSAI at a distance z from the apex, and a, b, and c are parameters), and the other is the restricted model imposed two restrictions on the unrestricted model: (i) the two straight lines intersect at the coordinates (CL, CSAI_(CL)), (ii) the area of the diagram formed from the two straight lines is equal to the SVI. AKAIKE's information criterion (AIC) and root mean square error (RMSE) were used to identify the better model. Data were obtained from 112 sample trees selected in six pure stands of even-aged Hinoki cypress. Parameters in the unrestricted model were estimated from CSAI dataset using least-squares method, and those in the restricted model were calculated from measurements of CL, SVI, and CSAI_(CL) for each sample tree. The unrestricted model fit well to CSAI observations, indicating that the vertical distribution of CSAI can be described by two straight lines. The AIC value for the restricted model was smaller than that for the unrestricted model, indicating that the restricted model is superior to the unrestricted model in describing the vertical distribution of CSAI. Although the restricted model returned a larger RMSE value than the unrestricted model did, the difference in RMSE was small. In conclusion, the vertical distribution of CSAI for Hinoki cypress can be described by two straight lines derived from CL, SVI, and CSAI_(CL).

View full abstract