Auckland, New Zealand
Auckland, New Zealand

Time filter

Source Type

Watt M.S.,Scion Research | Adams T.,MetService | Marshall H.,Interpine Forestry Ltd. | Pont D.,Scion Research | And 3 more authors.
New Zealand Journal of Forestry Science | Year: 2013

Background: Light Detection and Ranging (LiDAR) is an established technology that has been shown to provide accurate information on individual-tree and stand-level forest structure. Although LiDAR has been widely used to describe stand structural dimensions the utility of this technology to predict spatial variation in wood quality traits is largely unexplored. This study used LiDAR metrics to predict spatial variation in total stem volume (TSV) and outerwood stress-wave velocity (V) in an even-aged mature forest (25 yrs) of moderate size (stocked area of 217.8 ha). Outerwood stress-wave velocity is a good predictor of modulus of elasticity which is a key performance criterion for structural timber. Methods: Linear and non-linear models were developed to predict TSV and V. Models of TSV were developed from the full dataset that included 163 plots while models of V were developed from a subset of 32 plots in which V had been measured. Results: The best statistical models that included only LiDAR data, explained 60% and 37% of the variation in TSV and V, respectively. Addition of measured stand density to both models significantly improved the R2 to, respectively, 0.76 and 0.70 for TSV and V. The root-mean square error for the final models of TSV and V were, respectively, 64.0 m3 ha-1 and 0.086 km s-1. Conclusion: At the forest level LiDAR metrics were found to be useful for predicting both V and TSV. Further research should examine the link between LiDAR metrics and V across broader ranges of V to confirm these findings. © 2013 Watt et al.; licensee Springer.


Watt M.S.,Scion Research | Adams T.,Metservice | Watt P.,Indufor Asia Pacific Ltd | Marshall H.,Interpine
New Zealand Journal of Forestry Science | Year: 2013

Background: When aerial LiDAR data is used to construct Digital Elevation Models (DEMs) under vegetation, DEM quality will invariably suffer due to attenuation of the laser pulses by the land cover. Although the ratio of ground returns to outgoing pulses (GRper) is known to vary widely for forest applications, little research has quantified the influence of forest stand structure and site conditions on this ratio. An understanding of how these factors influence GRper is crucial for the development of accurate DEMs. Methods: Using an extensive national dataset obtained from New Zealand's plantation forests the objective of this research was to develop a multiple regression model of GRper that could be used to specify the necessary LiDAR pulse density for development of accurate DEMs. Results: Within the dataset GRper averaged 30.5% ranging from 0.73 to 92.2%. The final model of GRper included stand age, crop density, non-crop density and slope and accounted for 48% of the variance in GRper with root mean square error (RMSE) of 13.9%. The percentage of ground returns declined exponentially as stand age, crop and non-crop density increased and declined linearly with increases in slope. GRper was not substantially affected by either the pulse density, stand aspect or whether the stand comprised Pinus radiata or Pseudotsuga menziesii. Conclusion: The developed model highlights the sensitivity of GRper to stand and site conditions. This model is likely to be of considerable use in defining the optimal LiDAR pulse density across a range of forest environments. © 2013 Watt et al.; licensee Springer.


Watt P.,Indufor Asia Pacific Ltd | Meredith A.,Indufor Asia Pacific Ltd | Yang C.,Indufor Asia Pacific Ltd | Watt M.S.,Scion Research
New Zealand Journal of Forestry Science | Year: 2013

Background: A number of data sources currently exist that can provide information on forest plantations at a range of scales over an entire rotation cycle. In particular, LiDAR is quickly becoming the technology of choice for harvest planning and providing local-scale estimates of forest structure. Its application is still limited as repeat annual acquisition at this scale is generally cost prohibitive. Development of temporally updateable models that can accurately project important metrics such as tree height between LiDAR acquisitions would be of considerable use to resource managers. The objective of this research was to develop models of Pinus radiata height using GIS spatial data supplemented with RapidEye satellite imagery. Methods: Multiple regression models were constructed to describe maximum canopy height (Hm) derived from LiDAR at two relatively distant study sites located in Kaingaroa and Tairua forests. A randomised selection of 300 m2 circular plots was made at both sites and average values of Hm within these plots were used for the modelling. Sources of information used for predicting Hm included stand age and spatial information describing environmental variables and stand productivity. This information was supplemented with spectra and vegetation ratios derived from high resolution RapidEye satellite imagery. Results: The most robust models of Hm that were developed for both sites included a combination of the crop age obtained from the stand GIS, Site Index (obtained from a GIS surface) and the red-edge vegetation ratio (REVI) The final models of Hm had respective R2 of 0.99 and 0.94 for the Kaingaroa and Tairua sites. At both sites, stand age was the strongest predictor of Hm. However, the inclusion of REVI from high resolution imagery did add an updatable temporal dimension to the model. Changes in REVI are sensitive to the impacts of abiotic and biotic factors that are not captured by stand age and Site Index. Conclusion: Applied operationally, this model can be used in a GIS environment to estimate tree height and identify areas of anomalous growth or disturbance caused by wind, snow, fire or disease. © 2013 Watt et al.


Watt M.S.,Scion Research | Meredith A.,Indufor Asia Pacific Ltd | Watt P.,Indufor Asia Pacific Ltd | Gunn A.,Blakely Pacific Ltd
New Zealand Journal of Forestry Science | Year: 2013

Background: Light Detection and Ranging (LiDAR) has been successfully used to describe a wide range of forest metrics at local, regional and national scales. However, little research has used this technology in young Douglas-fir stands to describe key stand characteristics used as criterion for operational thinning. The objective of this research was to develop models of Douglas-fir mean top height, basal area, volume, mean diameter (at breast height), green crown height and stand density from LiDAR and stand information. Methods: Data for this study were obtained from four widely separated young (age range of 9 to 17 years) Douglas-fir plantations in the South Island, New Zealand. LiDAR was acquired for the entire area and stand metrics were measured within 122 plots established across the study area. Spatially synchronous stand and LiDAR metrics were extracted from the plots. Using this dataset, multiple regression models were developed for each of the six stand metrics. Results: The final models constructed for mean top height, green crown height, total stem volume, mean diameter, basal area, and stand density had R2 values of 0.85, 0.79, 0.86, 0.86, 0.84 and 0.55, respectively, with root mean square errors of 1.02 m, 0.427 m, 20.2 m3 ha-1, 13.9 mm, 3.81 m2 ha-1 and 355 stems ha-1, respectively. With the exception of stand density, all relationships were relatively unbiased. Variables with the greatest contribution (with the partial R2 in brackets) to models of mean top height, green crown height, volume, mean diameter and basal area included the 75th (0.85), 1st (0.76), 10th (0.83), 95th (0.74), and 10th (0.72) LiDAR height percentiles. The LiDAR height interquartile distance was the most important contributor (partial R2 = 0.33) to the model of stand density. Conclusion: With the exception of stand density, the final models for stand metrics were sufficiently precise to be used for scheduling thinning operations. This study demonstrates the utility of LiDAR to accurately estimate key structural attributes of young Douglas-fir and to assist with forest management over a widely dispersed resource. © Watt et al.; licensee Springer.


Watt P.,Indufor Asia Pacific Ltd. | Watt M.S.,Scion Research
New Zealand Journal of Forestry | Year: 2012

Cost effective options for providing up-to-date information on the status of forest resources to improve planning are important. Recent advances in satellite technology have made it possible to acquire imagery on a regular basis at a spatial resolution that is likely to be useful to forest managers. This paper describes the use of RapidEye satellite imagery to identify new harvest areas. Processes are also demonstrated that use imagery to delineate stands, and areas within stands, with growth that differs from expected values.


Watt M.S.,Scion Research | Dash J.P.,Scion Research | Bhandari S.,Indufor Asia Pacific Ltd | Watt P.,Indufor Asia Pacific Ltd
Forest Ecology and Management | Year: 2015

Site Index (. SI) is one of the main measures of forest productivity used throughout the world. For even-age plantations Site Index is defined as the height of dominant trees at a given reference age. Site Index is normally determined from field measurements and expressed from these measurements at the resolution of the stand. Development of fine resolution spatial surfaces describing variation in productivity across broad landscapes would be of considerable use in improving stand management. Using data obtained from a large Pinus radiata D. Don forest located in the central North Island, New Zealand, the objective of this study was to compare the precision of parametric and non-parametric models of Site Index that included explanatory variables extracted from aerially acquired Light Detection and Ranging (LiDAR), satellite imagery or environmental surfaces and combinations of these three data sources. Models were constructed both with and without age as an explanatory variable as managers may not always have access to stand age. A total of 32 models (16 data sources. ×. two model methods) were constructed using data from 484 plots. Validation methods used to examine precision and bias of these models included leave one out cross validation and k-fold analysis.For all but one of the 16 data sources parametric models were found to be more precise than non-parametric models. Inclusion of stand age as an explanatory variable improved the precision of all but one model. For parametric models that included stand age, the R2 and RMSE (in brackets) for models with (i) all metrics derived from satellite imagery, (ii) environmental surface variables, (iii) variables derived from satellite imagery and environmental surfaces, (iv) LiDAR metrics and (v) all available variables were, respectively, 0.237 (2.850m), 0.613 (2.267m), 0.716 (2.025m), 0.883 (1.378m) and 0.801 (1.672m). These results show that LiDAR was the most useful data source for precise and unbiased prediction of Site Index. The parametric model created using variables derived from environmental surfaces and satellite imagery was also very precise showing that, in combination, these datasets may provide a useful alternative for predictions of Site Index when LiDAR data are not available. © 2015 Elsevier B.V.


Dash J.P.,Scion Research | Watt M.S.,Scion Research | Bhandari S.,Indufor Asia Pacific Ltd | Watt P.,Indufor Asia Pacific Ltd
Forestry | Year: 2016

The objective of this study was to compare the utility of combinations of data from airborne laser scanning (ALS), RapidEye satellite imagery and auxiliary environmental data to predict stand structure in a plantation forest. Both parametric and non-parametric modelling techniques that could simultaneously predict a multivariate response were employed and found to produce predictions with similar levels of accuracy. Response variables were derived from 463 field measurement plots that were used during model development; a further 60 randomly selected plots were set aside for validation of model performance. Candidate predictor variables were extracted from the ALS data, satellite data and auxiliary environmental data, and the variables with the greatest explanatory power were used to create six separate models based on combinations of the data sources. Model validation showed that models using RapidEye data only were the least precise and that adding auxiliary environmental data only led to a moderate improvement in model precision. The model precision observed was similar to those reported previously from studies using satellite data to predict stand structure. Models developed using data from ALS were by far the most precise and adding information from satellite data or auxiliary environmental data led to negligible improvement in the prediction of stand structure. Although the outputs of both model types were similar, the practical efficiencies of using the non-parametric approach make it appealing to meet the demands of managers of industrial plantation forest managers. © 2015 Institute of Chartered Foresters. All rights reserved.


Watt M.S.,Scion Research | Dash J.P.,Scion Research | Watt P.,Indufor Asia Pacific Ltd | Bhandari S.,Indufor Asia Pacific Ltd
New Zealand Journal of Forestry Science | Year: 2016

Background: An understanding of how plantation productivity varies spatially is important for forest planning, management and projection of future plantation yields and returns. The 300 Index is a volume productivity index developed for Pinus radiata D.Don that has been widely used within New Zealand to assess site productivity. Although the 300 Index is routinely characterised at the stand level, little research has investigated if remotely sensed data sources can be used in combination with environmental layers to precisely predict this metric at fine spatial resolution. Methods: This study uses an extensive dataset obtained from P. radiata plantations in the central North Island, New Zealand. Using this dataset, the objective of this research was to compare the precision of parametric and non-parametric models of the 300 Index that included explanatory variables extracted from aerially acquired light detection and ranging (LiDAR), satellite imagery (RapidEye) at 5-m resolution or environmental layers and combinations of these three data sources. Models were constructed both with and without stand age as an explanatory variable as managers may not always have access to stand age. A total of 28 models (14 data sources × two model methods) were constructed using data from 433 plots. Precision and bias of these models was determined using an independent dataset of 60 plots. Results: Of the non-parametric methods tested (k-most similar neighbour (k-MSN), k-nearest neighbour (k-NN)), k-NN using an optimised value of k-most precisely predicted the 300 Index for 11 of the 14 constructed models. The use of k-NN was found to be more precise than parametric models when age was not available but of overall similar precision to parametric models when stand age was available as a predictor. For models including stand age, the inclusion of LiDAR resulted in the most precise model (mean R2 = 0.789; root mean square error (RMSE) = 2.48 m3 ha−1 year−1) while for models without stand age, metrics extracted from both satellite imagery and environmental layers produced the most precise model of the 300 Index (R2 = 0.65; RMSE = 3.21 m3 ha−1 year−1). Conclusions: Results clearly show that models constructed from LiDAR provide the most precise means of estimating the 300 Index. However, in many situations, LiDAR is too expensive to acquire or stand age, which is used as a reference for linking LiDAR to 300 Index, is not available as an independent variable. Under these circumstances, results show that precise models can be constructed from variables derived from the combination of satellite imagery and environmental surfaces. © 2016, Watt et al.


Watt M.S.,Scion Research | Meredith A.,Indufor Asia Pacific Ltd | Watt P.,Indufor Asia Pacific Ltd | Gunn A.,Blakely Pacific Ltd
New Zealand Journal of Forestry Science | Year: 2014

Conclusion: This study was undertaken in highly stocked unthinned Douglas-fir stands located in areas with complex topography. Consequently, the pulse density thresholds described here are likely to be conservative and could be used to guide acquisition of high-quality LiDAR datasets for this species.Background: LiDAR is an established technology that is increasingly being used to characterise spatial variation in stand metrics used in forest inventory. As the cost of LiDAR acquisition markedly declines with LiDAR pulse density, it is useful to identify how far pulse density can be reduced without compromising the precision of relationships between LiDAR and stand metrics. Using plot measurements and LiDAR data obtained from highly stocked and unthinned Douglas-fir plantations (Pseudotsuga menziesii [Mirb.] Franco), the objective of this research was to characterise the precision of regressions between LiDAR metrics and stand metrics (mean top height, Hm, volume, V and mean diameter, D) under a range of pulse densities using Digital Terrain Models (DTMs) representing two common scenarios. Under the first scenario, which represents an initial acquisition, the point cloud was sequentially culled and used for creation of a DTM and corresponding LiDAR cloud metrics. In the second scenario, which represents a subsequent acquisition, a DTM generated at high pulse density (10 pulses m−2) was used for the creation of the corresponding LIDAR cloud metrics.Methods: Models describing the precision of regressions between LiDAR metrics and stand metrics were developed at 10 pulses m−2. LiDAR data were culled to pulse densities ranging from 10 to 0.01 pulses m−2and the impact of culling on the precision of these regressions was examined under the two scenarios.Results: For the scenario with the culled DTM, precision of the three models remained stable until densities of 2 – 3 pulses m−2were reached. Below this threshold, there was a gradual decline in precision to pulse densities of 0.7 – 1 pulses m−2at which point the R2 was 95% of the maximum values. Further culling of the data resulted in a sharp decline in model precision for all three regressions. For the scenario where the DTM was held at a high pulse density, little change in the precision of the regressions was found until pulse densities of 0.04 to 0.2 pulses m−2were reached. There was a sharp decline in precision below pulse densities of 0.04 pulses m−2for all three models. © 2014, Watt et al..


Watt P.,Indufor Asia Pacific Ltd | Watt M.S.,Scion Research
International Journal of Remote Sensing | Year: 2013

Light detection and ranging (lidar) has been successfully used to describe a wide range of forest metrics at local scales. However, little research has tested the general applicability of this technology to describe commercially important stand dimensions, such as total stem volume (V), at national levels across broad environmental gradients. Using an extensive national data set covering the spatial extent of Pinus radiata plantation forests in New Zealand, the key objectives of this study were to (1) develop regression models to best describe V for P. radiata from lidar metrics and (2) investigate whether these relationships could be improved using coincident environmental and stand-level information. Development of relationships between lidar metrics and forest volume are of particular importance for P. radiata, as this species constitutes approximately 90% of the 1.8 Mha plantation resource. Using lidar mean height and the percentage of lidar ground returns, the initial model (model 1) accounted for 85% of the variance in V. Addition of stand stocking (number of stems ha-1), measured within the plots, to the model (model 2) significantly (p < 0.001) improved predictions, with R 2 increasing to 0.86 and the root mean square error declining from 80.1 m3 ha-1 to 71.6 m3 ha-1. For both models, partial responses show V to be most sensitive to lidar mean height, which was included in the model as a second-order polynomial. Although environmental variables are established determinants for V, their inclusion did not significantly improve either model 1 or 2. Residual values for both models showed little apparent bias when plotted against stand-level information or a wide array of environmental variables, supporting the general applicability of these relationships. © 2013 Copyright 2013 Taylor & Francis.

Loading Indufor Asia Pacific Ltd collaborators
Loading Indufor Asia Pacific Ltd collaborators