Yasumasa Hirata

Stand volume estimation by combining low laser-sampling density LiDAR data with QuickBird panchromatic imagery in closed-canopy Japanese cedar (Cryptomeria japonica) plantations

This study proposes a simple method for stand volume estimation by combining low laser-sampling density Light Detection and Ranging (LiDAR) data (i.e. 1 hit per 4 m2) with high-resolution optical imagery (i.e. 0.6 m) in coniferous plantations. The study area was in closed-canopy, mountainous, Japanese cedar (Cryptomeria japonica) plantations on undulating terrain at an elevation of 135–391 m above sea level. A total of 25 circular plots (0.04 ha) were established and stand volumes within plots were investigated in the field. The field-measured, plot-level stand volume ranged from 262.8 to 984.0 m3 ha−1 and the average value was 555.7 m3 ha−1. We used the measurements as validation data to evaluate estimates derived from an empirical regression model that was constructed on the basis of the allometry between crown diameter and diameter at breast height (DBH) from previous research. As a result, stand volume at various stand conditions could be estimated precisely regardless of different laser footprint sizes of 0.16–0.47 m when combining low-density LiDAR data with QuickBird panchromatic imagery. The maximum random error and root mean square error (RMSE) in stand volume estimates by data combination were 10% and 39% of the average stand volume, respectively. Thus, this method based on allometry and using low-density LiDAR data and high-resolution optical imagery could be capable of offering precise stand volume estimates in coniferous forests in different localities.

Effects of elevation and postharvest disturbance on the composition of vegetation established after the clear-cut harvest of conifer plantations in southern Shikoku, Japan

Large areas of previously clear-cut conifer plantations have been recently abandoned in Japan. We investigated the vegetation in the clear-cut sites and examined the environmental factors affecting species composition of the vegetation. We set up 32 study sites, each composed of several study plots (5 × 5 m), ranging from 220 m to 1060 m a.s.l. Elevation and warmth index (cumulated thermal quantity) were the primary factors affecting the species composition, with clear-cut areas showing a smaller effect in the nonmetric multidimensional scaling (NMS) ordination. Two-way indicator species analysis (TWINSPAN) divided the 32 study sites into ten vegetation groups, clustering the sites by elevation or by postharvest disturbances (i.e., replanting or browsing of Sika deer). Deciduous trees and shrubs were significant in the vegetation cover at higher elevations, while they were less so in areas of high Sika deer populations. We also investigated the abundance of old-growth species, which are expected to regenerate where the clear-cut site is abandoned. Evergreen Quercus and Castanopsis saplings were abundant at low elevations (<600 m), suggesting that they will successfully regenerate. The sapling densities of Abies firma and Betula grossa were significantly large where a clear-cut site was adjacent to natural forest, which is expected to act as a seed source. This implies that degraded deciduous forests may establish after clear-cutting at intermediate and high elevations (>600 m) if the clear-cut site is distinct from seed sources. It is argued that the preservation of natural forests is critical for the regeneration of old-growth species.

Estimation of Airborne Lidar-Derived Tropical Forest Canopy Height Using Landsat Time Series in Cambodia

In this study, we test and demonstrate the utility of disturbance and recovery information derived from annual Landsat time series to predict current forest vertical structure (as compared to the more common approaches, that consider a sample of airborne Lidar and single-date Landsat derived variables). Mean Canopy Height (MCH) was estimated separately using single date, time series, and the combination of single date and time series variables in multiple regression and random forest (RF) models. The combination of single date and time series variables, which integrate disturbance history over the entire time series, overall provided better MCH prediction than using either of the two sets of variables separately. In general, the RF models resulted in improved performance in all estimates over those using multiple regression. The lowest validation error was obtained using Landsat time series variables in a RF model (R2 = 0.75 and RMSE = 2.81 m). Combining single date and time series data was more effective when the RF model was used (opposed to multiple regression). The RMSE for RF mean canopy height prediction was reduced by 13.5% when combining the two sets of variables as compared to the 3.6% RMSE decline presented by multiple regression. This study demonstrates the value of airborne Lidar and long term Landsat observations to generate estimates of forest canopy height using the random forest algorithm.

Microhabitat use of wood mice ranging from a reserved belt with evergreen broad-leaved trees to a coniferous plantation

Wood mice Apodemus speciosus and Apodemus argenteus are potentially important seed dispersers and predators of Quercus and Castanopsis in Japan. We investigated the existence of two species of wood mice in warm-temperate forests ranging from a reserved belt of evergreen broad-leaved trees to a coniferous plantation, and analyzed the relationship between wood mouse occurrence and environmental factors to confirm their microhabitat use. We used two-way analysis of variance to analyze differences in the captured number of each wood mouse species in two trapping seasons as well as two stand types to confirm the interaction between the stand type and trapping season. Apodemus speciosus were often captured in the reserved belt, while captures of A. argenteus were independent of season and stand types. It is reasonable to conclude from the results of the trends in occurrence that the two species of wood mice showed different uses of the microhabitat. The result of linear discriminant analysis made it clear that the distance from the reserved belt had much to do with the difference in microhabitat use between the two species in the fruiting season of Quercus and Castanopsis. On average, A. speciosus moved 19.9 m during nonfruiting and 61.3 m during fruiting, while A. argenteus moved 8.1 m during nonfruiting, and 29.0 m during fruiting from analysis of the recapture position. The results indicate that both species of mice move around during the fruiting season more than in the nonfruiting season.

A Growth Prediction System for Local Stand Volume Derived from LIDAR Data

Recent advances in light detection and ranging (LIDAR) technology have enabled the estimation of valuable canopy parameters (e.g., crown diameter, leaf area, and canopy structure) that are difficult to obtain through in situ surveys. The objective of this study was to assess the utility of LIDAR-derived measurements of crown and growth parameters to model and predict the growth of sugi (Cryptomeria japonica) stands located in the University of Tokyo Forest, Chiba Prefecture, Japan. Initially, we confirmed that crown lengths and widths of trees in stands of various densities obtained from LIDAR data correlated with those measured in situ. Then, we developed a crown growth model from repeated LIDAR measurements of stands, suggesting that LIDAR data are adequate for this purpose, and indicating that crown surface area and tree volume growth were linearly related (R2 = 0.90; p < 0.01; RMSE tree volume < 0.02 m3). The model also provided robust predictions of the volume growth of local forests in 10 × 10 m plots based on LIDAR-derived estimates of crown surface areas. Future work should test the applicability of this growth model to facilitate practical forest management.

The extraction of canopy-understory vegetation-topography structure using helicopter-borne LIDAR measurement between a plantation and a broad-leaved forest

The relationship between stratification of canopy layers, understory vegetation and topography was investigated using helicopter-borne laser scanner data. The study plot was established ranging from an evergreen broad-leaved forest to a plantation of hinoki cypress (Chamaecyparis obtuse). LIDAR measurement was conducted with high density of footprints (23.4 points/m/sup 2/) and both first pulse data and last pulse data were recorded. A local minimum filter was used to generate a digital elevation model (DEM). Adjoining spaces with 1 m wide along a certain direction in the stand were assumed and whole measurement data within every space were projected to a corresponding vertical plane to comprehend a canopy structure. It was found that through this processing that a series of vertical projective planes described stratification of canopy layers, gap structure, distribution of understory vegetation as well as topography.

Natural Forest Mapping in the Andes (Peru): A Comparison of the Performance of Machine-Learning Algorithms

The Andes mountain forests are sparse relict populations of tree species that grow in association with local native shrubland species. The identification of forest conditions for conservation in areas such as these is based on remote sensing techniques and classification methods. However, the classification of Andes mountain forests is difficult because of noise in the reflectance data within land cover classes. The noise is the result of variations in terrain illumination resulting from complex topography and the mixture of different land cover types occurring at the sub-pixel level. Considering these issues, the selection of an optimum classification method to obtain accurate results is very important to support conservation activities. We carried out comparative non-parametric statistical analyses on the performance of several classifiers produced by three supervised machine-learning algorithms: Random Forest (RF), Support Vector Machine (SVM), and k-Nearest Neighbor (kNN). The SVM and RF methods were not significantly different in their ability to separate Andes mountain forest and shrubland land cover classes, and their best classifiers showed a significantly better classification accuracy (AUC values 0.81 and 0.79 respectively) than the one produced by the kNN method (AUC value 0.75) because the latter was more sensitive to noisy training data.

Object-Based Mapping of Aboveground Biomass in Tropical Forests Using LiDAR and Very-High-Spatial-Resolution Satellite Data

Developing countries that intend to implement the United Nations REDD-plus (Reducing Emissions from Deforestation and forest Degradation, and the role of forest conservation, sustainable management of forests, and enhancement of forest carbon stocks) framework and obtain economic incentives are required to estimate changes in forest carbon stocks based on the IPCC guidelines. In this study, we developed a method to support REDD-plus implementation by estimating tropical forest aboveground biomass (AGB) by combining airborne LiDAR with very-high-spatial-resolution satellite data. We acquired QuickBird satellite images of Kampong Thom, Cambodia in 2011 and airborne LiDAR measurements in some parts of the same area. After haze reduction and atmospheric correction of the satellite data, we calibrated reflectance values from the mean reflectance of the objects (obtained by segmentation from areas of overlap between dates) to reduce the effects of the observation angle and solar elevation. Then, we performed object-based classification using the satellite data (overall accuracy = 77.0%, versus 92.9% for distinguishing forest from non-forest land). We used a two-step method to estimate AGB and map it in a tropical environment in Cambodia. First, we created a multiple-regression model to estimate AGB from the LiDAR data and plotted field-surveyed AGB values against AGB values predicted by the LiDAR-based model (R2 = 0.90, RMSE = 38.7 Mg/ha), and calculated reflectance values in each band of the satellite data for the analyzed objects. Then, we created a multiple-regression model using AGB predicted by the LiDAR-based model as the dependent variable and the mean and standard deviation of the reflectance values in each band of the satellite data as the explanatory variables (R2 = 0.73, RMSE = 42.8 Mg/ha). We calculated AGB of all objects, divided the results into density classes, and mapped the resulting AGB distribution. Our results suggest that this approach can provide the forest carbon stock per unit area values required to support REDD-plus.

Estimation of Above-Ground Biomass of a Tropical Forest in Northern Borneo Using High-resolution Satellite Image

Abstract Estimating above-ground biomass is important in establishing an applicable methodology of Measurement, Reporting and Verification (MRV) System for Reducing Emissions from Deforestation and Forest Degradation-Plus (REDD+). We developed an estimation model of diameter at breast height (DBH) from IKONOS-2 image that led to above-ground biomass estimation (AGB). The IKONOS image was preprocessed with dark object subtraction and topographic effect correction prior to watershed segmentation for tree crown delineation. Compared to the field observation, the overall segmentation accuracy was 64%. Crown detection percent had a strong negative correlation to tree density. In addition, satellite-based crown area had the highest correlation with the field measured DBH. We then developed the DBH allometric model that explained 74% of the data variance. In average, the estimated DBH was very similar to the measured DBH as well as for AGB. Overall, this method can potentially be applied to estimate AGB over a relatively large and remote tropical forest in Northern Borneo.Key Words: tree crown delineation, biomass estimation, IKONOS-2

i-LOVE: ISS-JEM lidar for observation of vegetation environment

It is very important to watch the spatial distribution of vegetation biomass and changes in biomass over time, representing invaluable information to improve present assessments and future projections of the terrestrial carbon cycle. A space lidar is well known as a powerful remote sensing technology for measuring the canopy height accurately. This paper describes the ISS(International Space Station)-JEM(Japanese Experimental Module)-EF(Exposed Facility) borne vegetation lidar using a two dimensional array detector in order to reduce the root mean square error (RMSE) of tree height due to sloped surface.