Search

Questions: Previous studies have shown that mire vegetation responds to fine‐scale differences in topography; however, very few studies have examined the relationship between multiple topographic conditions and vegetation distribution patterns by using topographic information with a resolution finer than microtopographic patches of sub‐meter scale in mires. Is it possible to predict vegetation distribution patterns in a mire using topographic information derived from drone imagery? How do topographic conditions affect the emergence of vegetation types? Location: 3 km2 of sloping mire in eastern Hokkaido, northern Japan. Methods: Digital surface models (DSMs) and orthomosaics were created using an affordable drone (Phantom3 advanced) and photogrammetry. We calculated 23 variables reflecting topographic conditions from DSMs. The locations of four vegetation types (Moliniopsis japonica–Eriophorum vaginatum, ME; Moliniopsis japonica–Carex lasiocarpa, MC; Sphagnum fuscum–Ledum palustre, SL; and Carex lyngbyei–Thelypteris palustris, CT) were recorded by field survey and orthomosaics. Random forest models that predicted vegetation types from topographic variables were created and we evaluated the prediction accuracy, importance of variables and effect of variables on the occurrence probability of the vegetation types. Results: All predictive indicators (recall, precision, overall accuracy, kappa) were very high (>0.9). Important variables for modeling were altitude, relative height to the surroundings, slopes at a scale of several 10 m, soil wetness index and terrain ruggedness. Each vegetation type appeared under different topographic conditions; altitude was a key factor in ME and MC, and slope in CT. SL showed patchy distribution predictions on hummocks. Conclusions: Vegetation distribution in the mire was strongly related to topographic information based on the DSMs. Analysis of the variable effects indicates that the emergence of the vegetation types was explained by a combination of multiple topographic conditions. This approach has the outstanding advantages of obtaining many survey points quickly and retrieving topographic conditions difficult to measure using on‐ground field surveys. [ABSTRACT FROM AUTHOR]

In cool-temperate regions of Japan, riparian forests tend to have high tree species richness, but it is unknown if this trend is consistent in warm-temperate regions. Here, we assessed tree species richness and forest structure in three topographic units reflecting the ridge to valley gradient (slope crests and upper and lower slopes) in a mature broadleaf evergreen forest in southwest Japan. We established a 0.59-ha valley bottom plot that was primarily comprised of a lower slope unit, then surveyed and mapped all living trees with a diameter at breast height > 5 cm. Resulting compositional and species richness data were compared to those from an adjacent permanent plot that included the slope crest and upper slope units, provided by the Japanese Ministry of the Environment. In the valley bottom, debris flow terraces had the highest tree density and served as important habitat for broadleaf deciduous trees. Tree species compositions were broadly similar in that nearly all species were found in each topographic unit. However, there were obvious differences in species densities among topographic units, with the lowest densities found in the lower slope unit. Species–area curves indicated that tree species richness was lowest in the lower slope unit. Species–individual curves indicated that all three units had very similar total species richness. Even when forbs were included, understory species richness in the lower slope unit remained low. Our results suggest that unlike broadleaf deciduous cool-temperate forests, broadleaf evergreen riparian forests in warm-temperate regions may not necessarily relate to greater species richness. [ABSTRACT FROM AUTHOR]

In the epicentral region of subduction earthquakes, large-scale landslides are triggered by strong shaking of slopes weakened by flexural toppling. The distribution of flexural-toppled slopes is the primary information for predicting and mitigating landslide damage beforehand. This study proposes and tests a simple algorithm for extracting flexural-toppled slopes on a regional basis from a 10-m resolution digital elevation model by windows of two sizes. Local relief is extracted with a large window (200 × 200 pixels), and microtopography is extracted with a small window (3 × 3 pixels). The algorithm was derived from the region of the Kanagi-kuzure landslide in Shikoku, and tested at Aka-kuzure landslide in Honshu in Japan. Both areas are underlain by strongly tilted sedimentary rocks in epicentral regions of subduction earthquakes along the Nankai Trough. [ABSTRACT FROM AUTHOR]

Natural regeneration of canopy tree species is generally poor under a selective cutting regime practised in mixed conifer–hardwood stands on Hokkaido, northern Japan, and is prevented by dense bamboo layers in the understory. To promote tree regeneration, we scarified soil in ten sites, with elevations ranging from 420 to 760 m, in 1979, with a rake-equipped bulldozer to produce three different artificial microtopographies (concave, flat, and convex). Twenty-six years after scarification, six canopy tree species ( Acer mono, Abies sachalinensis, Betula ermanii, B. maximowicziana, Kalopanax pictus, and Picea jezoensis) had become well established. Abies sachalinensis and B. ermanii regenerated at all sites. Picea jezoensis was well established in high-elevation sites, whereas other hardwood tree species ( A. mono, B. maximowicziana, and K. pictus) were well established at low-elevation sites. The Betula species dominated upper layers of regenerated trees, and two conifer species ( A. sachalinensis and P. jezoensis) occupied lower layers. Regenerated densities of four (of six) major tree species were positively correlated with standing timber volume of mother trees, whereas those of shade-intolerant species (two Betula species and K. pictus) were negatively correlated with standing timber volume of all canopy trees. Small-seeded species ( P. jezoensis and the two Betula species) had higher regeneration densities on convex sites than on other microtopographies. Adequate seed supply, convex substratum for small-seeded species, and control of light conditions for shade-intolerant species should be taken into account when planning regeneration of desirable tree species in central Hokkaido. [ABSTRACT FROM AUTHOR]

With the industrialization in East Asia, greater amounts of ozone and acidic aerosols are transported to Japan. Yakushima Island (30°20′N, 130°31′E) is affected directly by long-range transboundary air pollution from the continent because there are no industrial areas and sources of pollution between the continent and the island. Recently, there has been a significant decline of Pinus amamiana in the island and the actual condition of air pollution was investigated. A significant amount of air pollutants were detected in winter. The daily mean tropospheric O3 concentration was nearly 100 ppb in winter. A large amount of non-sea salt sulfate (nss-SO4 2−) was deposited on the needle surfaces in the northwest part of the island and dry deposition rates of SO4 2− were remarkably higher in February than in November. About half of the increase in SO4 2− was nss-SO4 2− and the rate in February was over four to five times as high as that in November. The load of dry deposition on the crown of trees was strongly influenced by the topographic position of the slope. The amount of deposited SO4 2− was larger on trees growing on ridges and the tendency of P. amamiana to grow on ridges where it sticks out over 20 m enhances the load of dry deposition and O3 on the needle surfaces. The potassium leakage and ethylene emission from needles were remarkably increased and the needle mass per area decreased from November to February. It was suggested that the additional input of nss-SO4 2− acidifies the needle surface and accelerates leaf surface deterioration and leaching from needles and soils. [ABSTRACT FROM AUTHOR]

• Distances and directions from volcanos are effective covariates for SOC mapping. • 3D mapping with 10-m resolution can reproduce SOC along with microtopography. • SOC may accumulate more on ridges, highland slopes, and landslide masses. Geographic information on soil organic carbon (SOC) is important for climate change research and understanding the global carbon cycle. At present, this information is limited for forest soils in mountainous regions that have complex topographies and are affected by tephra deposition. This study predicted the SOC concentration (OC), bulk density (BD), apparent rock coarse fraction (CF), and SOC stock (OCS) at 0–5, 5–15, and 15–30 cm in the soil horizon for the forested areas (ca. 230 000 km2) of Japan. This was carried out with a 10 m grid resolution using a digital soil mapping (DSM) approach. To determine the appropriate spatial prediction model, we evaluated the accuracy of a two dimensional approach for ordinary kriging (OK) and regression kriging (RK), and a three dimensional approach for random forest model (RF). The RF was based on topography, climate and vegetation factors (RF env), distance and direction from 152 volcanos (RF env+vol), distance from survey points (RF env+sp); these factors were also combined (RF all) as explanatory attributes. The results demonstrated an improvement in the average root mean square errors (RMSEs) of RF env+vol , RF env+sp , and RF all by approximately 12%, 14%, 5%, and 21% for OC, BD, CF, and OCS using 10-fold cross-validation on a site-by-site basis, respectively, compared with OK. Based on the relatively small difference in improvement among RF env+vol , RF env+sp , and RF all (<0.5%), and the considerably high computational cost of RF env+sp and RF all , RF env+vol was considered as the appropriate model for the area. The R2 of RF env+vol was calculated at 0.59, 0.44, 0.30, and 0.38 in OC, BD, CF, and OCS, respectively, using 10-fold cross validation on a site-by-site basis. The predicted map by RF env+vol accurately reproduced the large spatial variation in OCS at small watershed, regional, and national scales; this have been derived from the effect of microtopography, tephra deposition, and the gradient of temperature with latitude, respectively. A larger OCS was predicted to accumulate on the ridges, highland slopes, near volcanos, and higher latitudes than at other areas. The mean OCS was estimated to be 7.6 kg·m−2 for a soil depth of 0–30 cm in Japanese forests. The results also suggest that the fine-scale three dimensional random forest approach using the distances from volcanoes showed promise for regional-scale OCS prediction. This approach was considered particularly effective in hills and mountainous areas that are strongly exposed to tephra deposition, this includes the Pacific Rim regions and other volcanic countries. [ABSTRACT FROM AUTHOR]

• In this study we have exploited LiDAR derived canopy height data at landscape level (230 km2). • Canopy of conifer plantation was higher than other forest types. • Taller trees were found in the concave valleys and at the intermediate elevations. • Along with topography, neighboring tree density and distance from canopy gap shape canopy height. • The results of this study will provide important insights about vulnerability of forests. Spatial variations of maximum canopy height are key indicators of many ecosystem processes such as above-ground biomass and productivity. With rapid advances in remote sensing techniques, such as airborne laser scanning (ALS), we can now measure canopy height, tree density, and canopy gaps from local to large scales. But these data are yet to be fully exploited for a better understanding of canopy height variations in relation to fine topographical heterogeneities and neighborhood conditions at the landscape level (>100 km2). In this study, we analyzed ALS data to extract spatial variations in canopy height and topographic features for a forest-dominated area of 230 km2 in Kyoto, Japan. The study area spanned an elevational range from 70 to 1000 m a.s.l and included four forest types, i.e., cedar or cypress plantations (PL), natural evergreen needle leaf forests (ENF), deciduous broadleaved forests (DBF), and evergreen broadleaved forests (EBF). PL had a greater mean canopy height than ENF, both of which were taller than DBF and EBF. Canopy height exhibited a hump-shaped relationship with elevation with the peak value at mid-altitude (461–560 m) in all forest types. Canopy height was greater by 5-10 m in the concave valleys than on the convex slopes and ridges. Trees were taller at greater distances from forest gaps and with lower tree density in the neighborhood. Local tree density, distance from the nearest canopy gap, and topographic curvatures were the most significant predictors in the best-fitted model (Random Forest: R2 = 0.41). The analytical approach developed in this study provides a useful tool for planning proper land use and maintaining a healthy forest stand at the regional scale. [ABSTRACT FROM AUTHOR]

The article presents a discussion of a study on the impact of microtopography and erosion on seedling emergence and survival in volcano Usu, Japan. The study focused on seedling emergence and survival ten years after the eruptions of the volcano from 1977 to 1978. It was found that seed germination rate were higher in Polygonum sachalinense when planted on finer-particle ground surface under laboratory conditions. It is noted that the amount of organic matter in volcanic deposits were minimal, after 11 years since the eruption.