Michael J. Falkowski

Remote sensing techniques to assess active fire characteristics and post-fire effects

Space and airborne sensors have been used to map area burned, assess characteristics of active fires, and characterize post-fire ecological effects. Confusion about fire intensity, fire severity, burn severity, and related terms can result in the potential misuse of the inferred information by land managers and remote sensing practitioners who require unambiguous remote sensing products for fire management. The objective of the present paper is to provide a comprehensive review of current and potential remote sensing methods used to assess fire behavior and effects and ecological responses to fire. We clarify the terminology to facilitate development and interpretation of comprehensible and defensible remote sensing products, present the potential and limitations of a variety of approaches for remotely measuring active fires and their post-fire ecological effects, and discuss challenges and future directions of fire-related remote sensing research.

Automated estimation of individual conifer tree height and crown diameter via two-dimensional spatial wavelet analysis of lidar data

We describe and evaluate a new analysis technique, spatial wavelet analysis (SWA), to automatically estimate the location, height, and crown diameter of individual trees within mixed conifer open canopy stands from light detection and ranging (lidar) data. Two-dimensional Mexican hat wavelets, over a range of likely tree crown diameters, were convolved with lidar canopy height models. Identification of local maxima within the resultant wavelet transformation image then allowed determination of the location, height, and crown diameters of individual trees. In this analysis, which focused solely on individual trees within open canopy forests, 30 trees incorporating seven dominant North American tree species were assessed. Two-dimensional (2D) wavelet-derived estimates were well correlated with field measures of tree height (r = 0.97) and crown diameter (r = 0.86). The 2D wavelet-derived estimates compared favorably with estimates derived using an established method that uses variable window filters (VWF) to estimate the same variables but relies on a priori knowledge of the tree height – crown diameter relationship. The 2D spatial wavelet analysis presented herein could potentially allow automated, large-scale, remote estimation of timber board feet, foliar biomass, canopy volume, and aboveground carbon, although further research testing the limitations of the method in a variety of forest types with increasing canopy closures is warranted.

Regression modeling and mapping of coniferous forest basal area and tree density from discrete-return lidar and multispectral satellite data

We compared the utility of discrete-return light detection and ranging (lidar) data and multispectral satellite imagery, and their integration, for modeling and mapping basal area and tree density across two diverse coniferous forest landscapes in north-central Idaho. We applied multiple linear regression models subset from a suite of 26 predictor variables derived from discrete-return lidar data (2 m post spacing), advanced land imager (ALI) multispectral (30 m) and panchromatic (10 m) data, or geographic X, Y, and Z location. In general, the lidar-derived variables had greater utility than the ALI variables for predicting the response variables, especially basal area. The variables most useful for predicting basal area were lidar height variables, followed by lidar intensity; those most useful for predicting tree density were lidar canopy cover variables, again followed by lidar intensity. The best integrated models selected via a best-subsets procedure explained ~90% of variance in both response variables. Natural-logarithm-transformed response variables were modeled. Predictions were then transformed from the natural logarithm scale back to the natural scale, corrected for transformation bias, and mapped across the two study areas. This study demonstrates that fundamental forest structure attributes can be modeled to acceptable accuracy and mapped with currently available remote sensing technologies.

The influence of conifer forest canopy cover on the accuracy of two individual tree measurement algorithms using lidar data

Individual tree detection algorithms can provide accurate measurements of individual tree locations, crown diameters (from aerial photography and light detection and ranging (lidar) data), and tree heights (from lidar data). However, to be useful for forest management goals relating to timber harvest, carbon accounting, and ecological processes, there is a need to assess the performance of these image-based tree detection algorithms across a full range of canopy structure conditions. We evaluated the performance of two fundamentally different automated tree detection and measurement algorithms (spatial wavelet analysis (SWA) and variable window filters (VWF)) across a full range of canopy conditions in a mixed-species, structurally diverse conifer forest in northern Idaho, USA. Each algorithm performed well in low canopy cover conditions (<50% canopy cover), detecting over 80% of all trees with measurements, and producing tree height and crown diameter estimates that are well correlated with field measurements. However, increasing tree canopy cover significantly decreased the accuracy of both SWA and VWF tree measurements. Neither SWA or VWF produced tree measurements within 25% of field-based measurements in high canopy cover (i.e., canopy cover >50%) conditions. The results presented herein suggest that future algorithm development is required to improve individual tree detection in structurally complex forests. Furthermore, tree detection algorithms such as SWA and VWF may produce more accurate results when used in conjunction with higher density lidar data.

Forest Monitoring Using Landsat Time Series Data: A Review

Abstract Unique among Earth observation programs, the Landsat program has provided continuous earth observation data for the past 41 years. Landsat data are systematically collected and archived following a global acquisition strategy. The provision of free, robust data products since 2008 has spurred a renaissance of interest in Landsat and resulted in an increasingly widespread use of Landsat time series (LTS) for multitemporal characterizations. The science and applications capacity has developed steadily since 1972, with the increase in sophistication offered over time incorporated into Landsat processing and analysis practices. With the successful launch of Landsat-8, the continuity of measures at scales of particular relevance to management and scientific activities is ensured in the short term. In particular, forest monitoring benefits from LTS, whereby a baseline of conditions can be interrogated for both abrupt and gradual changes and attributed to different drivers. Such benefits are enabled by data availability, analysis-ready image products, increased computing power and storage, as well as sophisticated image processing approaches. In this review, we present the status of remote sensing of forests and forest dynamics using LTS, including issues related to the sensors, data availability, data preprocessing, variables used in LTS, analysis approaches, and validation issues.

Supporting large-area, sample-based forest inventories with very high spatial resolution satellite imagery

Information needs associated with forest management and reporting requires data with a steadily increasing level of detail and temporal frequency. Remote sensing satellites commonly used for forest monitoring (eg, Landsat, SPOT) typically collect imagery with sufficient temporal frequency, but lack the requisite spatial and categorical detail for some forest inventory information needs. Aerial photography remains a principal data source for forest inventory; however, information extraction is primarily accomplished through manual processes. The spatial, categorical, and temporal information requirements of large-area forest inventories can be met through sample-based data collection. Opportunities exist for very high spatial resolution (VHSR; ie, <1 m) remotely sensed imagery to augment traditional data sources for large-area, sample-based forest inventories, especially for inventory update. In this paper, we synthesize the state-of-the-art in the use of VHSR remotely sensed imagery for forest inventory and monitoring. Based upon this review, we develop a framework for updating a sample-based, large-area forest inventory that incorporates VHSR imagery. Using the information needs of the Canadian National Forest Inventory (NFI) for context, we demonstrate the potential capabilities of VHSR imagery in four phases of the forest inventory update process: stand delineation, automated attribution, manual interpretation, and indirect attribute modelling. Although designed to support the information needs of the Canadian NFI, the framework presented herein could be adapted to support other sample-based, large-area forest monitoring initiatives.

A cross-comparison of field, spectral, and lidar estimates of forest canopy cover

A common challenge when comparing forest canopy cover and similar metrics across different ecosystems is that there are many field- and landscape-level measurement methods. This research conducts a cross-comparison and evaluation of forest canopy cover metrics produced using unmixing of reflective spectral satellite data, light detection and ranging (lidar) data, and data collected in the field with spherical densiometers. The coincident data were collected across a ~25 000 ha mixed conifer forest in northern Idaho. The primary objective is to evaluate whether the spectral and lidar canopy cover metrics are each statistically equivalent to the field-based metrics. The secondary objective is to evaluate whether the lidar data can elucidate the sources of error observed in the spectral-based canopy cover metrics. The statistical equivalence tests indicate that spectral and field data are not equivalent (slope region of equivalence = 43%). In contrast, the lidar and field data are within the acceptable error margin of most forest inventory assessments (slope region of equivalence = 13%). The results also show that in plots where the mean lidar plot heights are near zero, each of modeled remotely sensed estimates continues to report canopy cover >21% for lidar and >30% for all investigated spectral methods using near-infrared bands. This suggests these metrics are sensitive to the presence of herbaceous vegetation, shrubs, seedlings, saplings, and other subcanopy vegetation.

A Comparison of Two Open Source LiDAR Surface Classification Algorithms

With the progression of LiDAR (Light Detection and Ranging) towards a mainstream resource management tool, it has become necessary to understand how best to process and analyze the data. While most ground surface identification algorithms remain proprietary and have high purchase costs; a few are openly available, free to use, and are supported by published results. Two of the latter are the multiscale curvature classification and the Boise Center Aerospace Laboratory LiDAR (BCAL) algorithms. This study investigated the accuracy of these two algorithms (and a combination of the two) to create a digital terrain model from a raw LiDAR point cloud in a semi-arid landscape. Accuracy of each algorithm was assessed via comparison with >7,000 high precision survey points stratified across six different cover types. The overall performance of both algorithms differed by only 2%; however, within specific cover types significant differences were observed in accuracy. The results highlight the accuracy of both algorithms across a variety of vegetation types, and ultimately suggest specific scenarios where one approach may outperform the other. Each algorithm produced similar results except in the ceanothus and conifer cover types where BCAL produced lower errors.

Production of vegetation spatial-structure maps by per-object analysis of juniper encroachment in multitemporal aerial photographs

The remote sensing of vegetation, which has predominantly applied methods that analyze each image pixel as independent observations, has recently seen the development of several methods that identify groups of pixels that share similar spectral or structural properties as objects. The outputs of “per-object” rather than “per-pixel” methods represent characteristics of vegetation objects, such as location, size, and volume, in a spatially explicit manner. Before decisions can be influenced by data products derived from per-object remote sensing methods, it is first necessary to adopt methodologies that can quantify the spatial and temporal trends in vegetation structure in a quantitative manner. In this study, we present one such methodological framework where (i) marked point patterns of vegetation structure are produced from two per-object methods, (ii) new spatial-structural data layers are developed via moving-window statistics applied to the point patterns, (iii) the layers are differenced to highlight spatial-structural change over a 60 year period, and (iv) the resulting difference layers are evaluated within an ecological context to describe landscape-scale changes in vegetation structure. Results show that this framework potentially provides information on the population, growth, size association (nonspatial distribution of large and small objects), and dispersion. We present an objective methodological comparison of two common per-object approaches, namely image segmentation and classification using Definiens software and two-dimensional wavelet transformations.

Automatic detection of shrub location, crown area, and cover using spatial wavelet analysis and aerial photography

Characterizing shrub-steppe rangeland condition often requires fine-scale measurement of individual plants across broad areas. Advances in remote sensing to develop improved algorithms to census and monitor individual rangeland plants using image data are important for improving the efficiency with which these critical areas are monitored. Here, we performed and evaluated the first test of spatial wavelet analysis (SWA) to automatically detect the location and crown diameter of individuals of two species of shrubs (Artemisia tridentata and Purshia tridentata). Additionally, we quantified the aggregated cover of these shrubs at the plot scale. High spatial resolution (0.25 and 1 m) multispectral aerial imagery and field-based vegetation measurements were collected in both spring and fall 2005. We found that image- and field-based measures of individual shrubs and their crown areas were highly correlated in the fall imagery (r = 0.89). Image-based SWA prediction of shrub cover at the plot level correlated better with field-based measures (r = 0.91) than did a traditional, image texture-based measure (r = 0.71). Analyses of imagery acquired in spring resulted in poorer relationships due to the decreased phenological contrast between shrubs and understory grasses in spring relative to fall. Statistical equivalence tests demonstrated that individual shrub crown areas derived from field data and SWA were statistically equivalent and not biased, but the SWA- and field-based assessments of plot-level cover were not statistically equivalent. These results represent progress towards developing automatic methods to analyze shrubs at the landscape scale using remotely sensed imagery.