Delineation of forest/nonforest land use classes using nearest neighbor methods

The k-Nearest Neighbor (kNN) method of forest attribute estimation and mapping has become an integral part of national forest inventory methods in Finland in the last decade. This success of kNN method in facilitating multi-source inventory has encouraged trials of the method in the Great Lakes Region of the United States. Here we present results from applying the method to Landsat TM and ETM+ data and land cover data collected by the USDA Forest Service's Forest Inventory and Analysis (FIA) program. In 1999, the FIA program in the state of Minnesota moved to a new annual inventory design to reach its targeted full sampling intensity over a 5-year period. This inventory design also utilizes a new 4-subplot cluster plot configuration. Using this new plot design together with 1 year of field plot observations, the kNN classification of forest/nonforest/water achieved overall accuracies ranging from 87% to 91%. Our analysis revealed several important behavioral features associated with kNN classification using the new FIA sample plot design. Results demonstrate the simplicity and utility of using kNN to produce FIA defined forest/nonforest/water classifications.     

An effective assessment protocol for continuous geospatial datasets of forest characteristics using USFS Forest Inventory and Analysis (FIA) data

Geospatial datasets of forest characteristics are modeled representations of real populations on the ground. The continuous spatial character of such datasets provides an incredible source of information at the landscape level for ecosystem research, policy analysis, and planning applications, all of which are critical for addressing current challenges related to climate change, urbanization pressures, and data requirements for monitoring carbon sequestration. However, the effectiveness of these applications is dependent upon the accuracy of the geospatial input datasets. A comprehensive set of robust measures is necessary to provide sufficient information to effectively assess the accuracy of these modeled geospatial datasets being produced. Yet challenges in the availability of reference data, in the appropriateness of assessment methods to dataset use, and in the completeness of assessment methods available have continued to hamper the timely and consistent application of map assessments. In this study we present a suite of assessments that can be used to characterize the accuracy of geospatial datasets of modeled continuous variables—an increasingly common format for modeling such attributes as proportion or probability of forestland as well as more traditionally continuous attributes such as leaf area index and forest biomass. It is a comparative accuracy assessment, in which each modeled dataset is compared to a set of reference data, recognizing both the potential for error in reference data, and probable differences in spatial support between the datasets. When used together, this proposed suite of assessments provides essential information on the type, magnitude, frequency and location of errors in each dataset. The assessments presented depend upon reference data with large sample sizes. The U.S. Forest Service (USFS) Forest Inventory and Analysis (FIA) database is introduced as an available reference dataset of sufficient sampling intensity to take full advantage of these assessments and facilitate their prompt application after modeled datasets are developed. We illustrate the application of this suite of assessments with two modeled datasets of forest biomass, in Minnesota and New York. The information provided by this suite of assessments substantially improves a user's ability to apply modeled geospatial datasets effectively and to assess the relative strengths and weaknesses of multiple datasets depicting the same forest characteristic.     

Lidar remote sensing of forest biomass: A scale-invariant estimation approach using airborne lasers

Researchers in lidar (Light Detection And Ranging) strive to search for the most appropriate laser-based metrics as predictors in regression models for estimating forest structural variables. Many previously developed models are scale-dependent that need to be fitted and then applied both at the same scale or pixel size. The objective of this paper is to develop methods for scale-invariant estimation of forest biomass using lidar data. We proposed two scale-invariant models for biomass: a linear functional model and an equivalent nonlinear model that use lidar-derived canopy height distributions (CHD) and canopy height quantile functions (CHQ) as predictors, respectively. The two models are called functional regression models because the predictors CHD and CHQ are themselves functions or functional data. The model formulation was justified mathematically under moderate assumptions. We also created a fine-resolution biomass map by mapping individual tree component biomass in a temperate forest of eastern Texas with a lidar tree-delineation approach. The map was used as reference data to synthesize training and test datasets at multiple scales for validating the two scale-invariant models. Results suggest that the models can accurately predict biomass and yield consistent predictive performances across a variety of scales with an R² ranging from 0.80 to 0.95 (RMSE: from 14. 3 Mg/ha to 33.7 Mg/ha) among all the fitted models. Results also show that a training data size of around 50 plots or less was enough to guarantee a good fitting of the linear functional model. Our findings demonstrate the effectiveness of CHD and CHQ as lidar metrics for estimating biomass as well as the capability of lidar for mapping biomass at a range of scales. The functional regression models of this study are useful for lidar-based forest inventory tasks where the analysis units vary in size and shape. They also hold promise for estimating other forest characteristics such as below-ground biomass, timber volume, crown fuel weight, and Leaf Area Index.     

Development and validation of AVHRR-derived regional forest cover data for the north-eastern U.S.

As part of developing the geographic information system (GIS) to support a north-eastern U.S. regional forest change modelling effort, we investigated the utility of several sources of AVHRR data in regional forest cover mapping. Single-date classified Advanced Very High Resolution Radiometer (AVHRR) imagery in combination with existing USGS Land Use/Land Cover data was used to create a forest cover database that encompassed eastern New York state and all of New England. The USGS EROS Data Center Conterminous U.S. Land Cover Characteristics database was also evaluated for comparison. Statistical analysis showed that the AVHRR-derived regional land cover datasets provided estimates of total forest area that were comparable to U.S. Forest Service county level estimates. The AVHRR imagery recorded after leaf fall appeared to enhance the discrimination of coniferous vs. deciduous forests.     

A technique for extrapolating and validating forest cover across large regions Calibrating AVHRR data with TM data

Abstract

An approach to extending high-resolution forest cover information across large regions is presented and validated. Landsat Thematic Mapper (TM) data were classified into forest and nonforest for a portion of Jackson County, Illinois. The classified TM image was then used to determine the relationship between forest cover and the spectral signature of Advanced Very High Resolution Radiometer (AVHRR) pixels covering the same location. Regression analysis was used to develop an empirical relationship between AVHRR spectral signatures and forest cover. The regression equation developed from data from the single county calibration area in southern Illinois was then applied to the entire AVHRR scene, which covered all or parts of ten states, to produce a regional map of forest cover. This map was used to derive estimates of forest cover, within a geographical information system (GIS), for each of the 428 counties located within the boundaries of the original AVHRR scene. The validity of the overall regional map was tested by comparing the AVHRR/TM-derived estimates of county forest cover with independent estimates of county forest cover developed by the U.S. Forest Service (USFS). The overall correlation coefficient of the AVHRR/TM and USFS county forest cover estimates was r=0-89 (n=428 counties). Not surpris0ingly, some individual states and the areas nearer to the southern Illinois calibration centre had higher correlation coefficients. Absolute estimates of forest cover percentages were also significantly well predicted. With the future inclusion of multiple calibration centres representing a number of physiographic regions, the method shows promise for predicting continental and global estimates of forest cover.     

Combining national forest inventory field plots and remote sensing data for forest databases

Information about forest cover is needed by all of the nine societal benefit areas identified by the Group of Earth Observation (GEO). In particular, the biodiversity and ecosystem areas need information on landscape composition, structure of forests, species richness, as well as their changes. Field sample plots from National Forest Inventories (NFI) are, in combination with satellite data, a tremendous resource for fulfilling these information needs. NFIs have a history of almost 100 years and have developed in parallel in several countries. For example, the NFIs in Finland and Sweden measure annually more than 10,000 field plots with approximately 200 variables per plot. The inventories are designed for five-year rotations. In Finland nationwide forest cover maps have been produced operationally since 1990 by using the k-NN algorithm to combine satellite data, field sample plot information, and other georeferenced digital data. A similar k-NN database has also been created for Sweden. The potentials of NFIs to fulfil diverse information needs are currently analyzed also in the COST Action E43 project of the European Union. In this article, we provide a review of how NFI field plot information has been used for parameterization of image data in Sweden and Finland, including pre-processing steps like haze correction, slope correction, and the optimization of the estimation variables. Furthermore, we review how the produced small-area statistics and forest cover data have been used in forestry, including forest biodiversity monitoring and habitat modelling. We also show how remote sensing data can be used for post-stratification to derive the sample plot based estimates, which cannot be directly estimated from the spectral data.     

Automatic detection of harvested trees and determination of forest growth using airborne laser scanning

This paper demonstrates the applicability of small footprint, high sampling density airborne laser scanners for boreal forest change detection, i.e. the estimation of forest growth and monitoring of harvested trees. Two laser acquisitions were carried out on a test site using a Toposys-1 laser scanner. Three-dimensional canopy height models were calculated for both data sets using raster-based algorithms. Object-oriented algorithms were developed for detecting harvested and fallen trees, and for measuring forest growth at plot and stand levels. Out of 83 field-checked harvested trees, 61 could be automatically and correctly detected. All mature harvested trees were detected; it was mainly the smaller trees that were not. Forest growth was demonstrated at plot and stand levels using an object-oriented tree-to-tree matching algorithm and statistical analysis. The precision of the estimated growth, based on field checking or statistical analysis, was about 5 cm at stand level and about 10–15 cm at plot level. The authors expect that the methods may be feasible in large area forest inventories that make use of permanent sample plots. Together with methods for detecting individual sample trees, the methods described may be used to replace a large number of permanent plots with laser scanning techniques.     

Measuring forest structure and biomass in New England forest stands using Echidna ground-based lidar

A ground-based, upward-scanning, near-infrared lidar, the Echidna® validation instrument (EVI), built by CSIRO Australia, retrieves forest stand structural parameters, including mean diameter at breast height (DBH), stem count density (stems/area), basal area, and above-ground woody biomass with very good accuracy in six New England hardwood and conifer forest stands. Comparing forest structural parameters retrieved using EVI data with extensive ground measurements, we found excellent agreement at the site level using five EVI scans (plots) per site (R² = 0.94-0.99); very good agreement at the plot level for stem count density and biomass (R² = 0.90-0.85); and good agreement at the plot level for mean DBH and basal area (R² = 0.48-0.66). The observed variance at site and plot levels suggest that a sample area of at least 1 ha (10⁴ m²) is required to estimate these parameters accurately at the stand level using either lidar-based or conventional methods. The algorithms and procedures used to retrieve these structural parameters are dependent on the unique ability of the Echidna® lidar to digitize the full waveform of the scattered lidar pulse as it returns to the instrument, which allows consistent separation of scattering by trunks and large branches from scattering by leaves. This successful application of ground-based lidar technology opens the door to rapid and accurate measurement of biomass and timber volume in areal sampling scenarios and as a calibration and validation tool for mapping biomass using airborne or spaceborne remotely sensed data.     

Multi-sensor data fusion for estimating forest species composition and abundance in northern Minnesota

The magnitude, duration, and frequency of forest disturbance caused by the spruce budworm and forest tent caterpillar in northern Minnesota and neighboring Ontario, Canada have increased over the last century due to a shift in forest species composition linked to historical fire suppression, forest management, and pesticide application that has fostered increased dominance of host tree species. Modeling approaches are currently being used to understand and forecast potential management effects in changing insect disturbance trends. However, detailed forest composition data needed for these efforts is often lacking. We used partial least squares (PLS) regression to integrate different combinations of satellite sensor data including Landsat, Radarsat-1, and PALSAR, as well as pixel-wise forest structure information derived from SPOT-5 sensor data (Wolter et al., 2009), to determine the best combination of sensor data for estimating near species-level proportional forest composition (12 types: 10 species and 2 genera). Single-sensor and various multi-sensor PLS models showed distinct species-dependent sensitivities to relative basal area (BA), with Landsat variables showing greatest overall sensitivity. However, best results were achieved using a combination of data from all these sensors, with several C-band (Radarsat-1) and L-band (PALSAR) variables showing sensitivity to the composition and abundance of specific species. Pixel-level forest structure estimates derived from SPOT-5 data were generally more sensitive to conifer species abundance (especially white pine) than to hardwood species composition. Relative BA models accounted for 68% (jack pine) to 98% (maple spp.) of the variation in ground data with RMSE values between 2.46% and 5.65% relative BA, respectively. Receiver operating characteristic (ROC) curves were used to determine the effective lower limits of usefulness of species relative BA estimates which ranged from 5.94% (jack pine) to 39.41% (black ash). These estimates were then used to produce a dominant forest species map for the study region with an overall accuracy of 78%. Most notably, this approach facilitated discrimination of aspen from paper birch as well as spruce and fir from other conifer species which is crucial for the study of forest tent caterpillar and spruce budworm dynamics in the Upper Midwest. We also demonstrate that PLS regression is an effective data fusion strategy for mapping composition of heterogeneous forests using satellite sensor data.     

Using satellite imagery and the k-nearest neighbors technique as a bridge between strategic and management forest inventories

Strategic forest inventory programs produce forest resource estimates for large areas such as states and provinces using data collected for a large number of variables on a relatively sparse array of field plots. Management inventories produce stand-level estimates to guide management decisions using data obtained with sampling intensities much greater than for strategic inventories. The costs associated with these greater sampling intensities have motivated investigations of alternatives to traditional sample-based management inventories. This study focused on a relatively inexpensive alternative to management inventories that uses strategic forest inventory plot data, Landsat Thematic Mapper (TM) satellite imagery, and the k-Nearest Neighbors (k-NN) technique. The approach entailed constructing stem density and basal area per unit area maps from which stand-level means were estimated as averages of k-NN pixel predictions. The study included investigations of the benefits of selecting optimal combinations of k-NN feature space variables derived from the TM imagery and the benefits of modifying the k-NN technique to eliminate spurious nearest neighbors. For both the stem density and basal area per unit area training data, the selection of optimal feature space covariates produced less than 1.5% improvement in root mean square error relative to using all covariates. The k-NN modification improved the sum of mean squared deviations for stand-level stem density and basal area per unit area estimates by 7–20% depending on the k-NN feature space covariates. For the best combination of feature space covariates, estimates of stand-level means were within confidence intervals for validation estimates for 11 of 12 stands for stem density and for 10 of 12 stands for basal area per unit area.