We estimated demographic parameters across entire size classes for 17 tree species (Betula, Carpinus, Fagus, Quercus, Castanea, Acer, Cerasus, Swida, Kalopanax, and Styrax) using a dataset over 18 years obtained from the Ogawa Forest Reserve, an old-growth deciduous forest located in the southern part of the Abukuma Mountains, Ibaraki Prefecture, central Japan (36° 56', 140° 35', 610 m in elevation) ¹⁾. Size classes were represented by 12 categories defined based on age, height, and diameter at breast height (DBH): new seedling (age < 1), aged seedling (age ≥ 1 and height < 30 cm), sapling (height 30 cm to 2 m), juvenile (height ≥ 2 m and DBH < 5 cm), D10 (DBH 5–15 cm, D10 was denoted by the midpoint of this range; later classes were similarly defined), D20, D30, D40, D50, D60, D70, and D80 (including trees with DBH ≥ 85 cm). We derived the annual mortality and growth index (i.e., the probability of a living tree transitioning to the next size class) for each species and size class using estimates of transition probabilities between size classes (i.e., stasis, progression and retrogression)²⁾. The stem densities of these 17 species in each size class are also presented³⁾.

File list

1) dataset - number of trees stagnating, progressing and retrogressing by size class.csv

2) estimates - mean and 95 percent credible intervals of demographic parameters by species and size class.csv

3) estimates - mean density of stems by species and size class.csv

The study site was located in the Ogawa Forest Reserve, an old-growth deciduous forest located in the southern part of the Abukuma Mountains, Ibaraki Prefecture, central Japan (36° 56', 140° 35', 610 m in elevation). The mean monthly temperature is 10.7°C, with a high of 22.6°C in August and a low of −0.9°C in February. Annual precipitation is approximately 1,910 mm; August and September are the wettest months and December and January are the driest. Maximum snow depth often reaches 50 cm. The dominant soil types are Andic Haplumbrepts and others (e.g., Andic Dystrochrepts). The forest canopy is mostly closed; only 9–15% of the canopy was gap at the time of this study.

We make available here 3 files. Details of the field methods and the analysis are described in the journal article.

1) Data used in the study. The number of trees for each species, size class and year is shown with their fates during the measurement period.

2) The mean density (ha^-1) of stems at each size class averaged over the study period. Size classes were defined as follows: new seedling (age < 1 year), aged seedlings (age ≥ 1 year and height < 30 cm), saplings (height of 30 cm to 2 m), juvenile (height ≥ 2 m and DBH < 5 cm), and D10 (DBH 5 cm to 15 cm). Stems with DBH ≥ 15 cm at D10 cm were similarly defined. The largest class, D80, includes stems of ≥ 85 cm DBH. For size classes from seedling to juvenile, data of 1987-1989 were not used because of the limited sample sizes.

3) Estimated annual probabilities of stasis, progression, and retrogression and derived mortality and growth values by species and size class. The 95% confidence intervals are also provided. The rightmost column indicates the total sample size for each species and each size class summed over the study period.


On June 18, the second file "mean_and_credible_intervals_of_estimates_and_derivatives_to_prepare.csv" have been replaced with "mean_and_credible_intervals_of_estimates_and_derivatives_to_prepare_corrected.csv" where mortality values at specific maximum life stages have been corrected