Search

IntroductionDirect-to-family clinical trials efficiently provide data while reducing the participation burden for children and their families. Although these trials can offer significant advantages over traditional clinical trials, the process of designing and implementing direct-to-family studies is poorly defined, especially in children with rheumatic disease. This paper provides lessons learnt from the design and implementation of a self-controlled, direct-to-family pilot trial aimed to evaluate the effects of a medication management device on adherence to hydroxychloroquine in paediatric SLE.MethodsSeveral design features accommodate a direct-to-family approach. Participants meeting eligibility criteria from across the USA were identified a priori through a disease registry, and all outcome data are collected remotely. The primary outcome (medication adherence) is evaluated using electronic medication event-monitoring, plasma drug levels, patient questionnaires and pill counts. Secondary and exploratory endpoints include (1) lupus disease activity measured by a remote SLE Disease Activity Index examination and the Systemic Lupus Activity Questionnaire; and (2) hydroxychloroquine pharmacokinetics and pharmacodynamics. Recruitment of the initial target of 20 participants was achieved within 10 days. Due to initial recruitment success, enrolment was increased to 26 participants. Additional participants who were interested were placed on a waiting list in case of dropouts during the study.Discussion and disseminationDirect-to-family trials offer several advantages but present unique challenges. Lessons learnt from the protocol development, design, and implementation of this trial will inform future direct-to-family trials for children and adults with rheumatic diseases. Additionally, the data collected remotely in this trial will provide critical information regarding the accuracy of teleresearch in lupus, the impact of adherence to hydroxychloroquine on disease activity and a pharmacokinetic analysis to inform paediatric-specific dosing of hydroxychloroquine.Trial registration numberClinicalTrials.gov Registry (NCT04358302).

Reduced rainfall in tropical regions decreases the availability of food resources for overwintering songbirds, with negative impacts on their body condition. Increasingly dry conditions at tropical wintering sites as a result of climate change may impact the ability of temperate-breeding songbirds to prepare for and execute their spring migration. We studied the carry-over effects of natural climatic fluctuations created by the El Niño–Southern Oscillation (ENSO) in tropical overwintering areas on 7 long-distance migratory songbirds at a Canadian stopover site. We used the Normalized Difference Vegetation Index as a proxy for tropical habitat productivity and resource availability and a 34-year bird banding dataset from Long Point, Ontario, Canada to assess migration timing and stopover body condition. To increase our ability to detect potential carry-over effects, we employed a novel approach of using recent migratory connectivity studies to restrict the wintering ranges to areas most likely used by individuals passing through southern Ontario. Using linear models, we found that overwinter habitat productivity was significantly negatively influenced by dry ENSO events in the overwintering ranges in 3 of 7 species, with a fourth near-significant. Subregional differences in the effect of ENSO on precipitation patterns may explain why we did not detect a consistent effect of ENSO on overwinter habitat productivity for all species. Despite restricting the wintering range and using a robust dataset for species with diverse life histories, we detected only weak and often conflicting evidence of population-level carry-over effects from dry ENSO events and overwinter habitat productivity. Negative carry-over effects may be strongest and most evident during the earlier stages of migration because birds may be able to compensate to some extent for poor departure condition and late departure while en route.

A new method for measuring the porosity of individual 2D raster patches in a GIS for characterizing the combined complexity of a shape&rsquo, s edge in conjunction with its internal perforations is developed. The method is centered on comparing the number of cellular edge&ndash, edge joins relative to the theoretical maximum number of similar joins possible given a set number of cells comprising a landscape patch. As this porosity (&Phi, ) increases, the patch (or shape) can be viewed as deviating from a maximally compact form, comprising higher edge complexity and internal heterogeneity (inclusion of perforations). The approach is useful for characterizing shapes for which a simple perimeter- or area-based metric misses the internal complexity and where the porosity of the patch may provide insight into spatial processes leading to the development of the landscape fabric. I present theoretical results to illustrate the mechanics of the approach and a small case study of boreal wildfire residual vegetation patches in Ontario, where real resulting wildfire process-driven landscape patches are assessed for their porosity at five spatial resolutions. The results indicate that naturally occurring and unsuppressed boreal wildfires in the study area typically produce residual vegetation patches with an average porosity of 17.6%, although this value varies slightly with the spatial resolution of the data representation.

Delineation of geographic space partitions landscapes into regions that express relative homogeneity within those delineated regions and many efforts exist to analyze the resulting shapes that these delineations produce. There is a desire to establish means by which the shapes of planar polygons can be measured, characterized, and compared. I present the ShrinkShape2 toolbox, a completely redesigned set of end-user tools (an advancement of its precursor ShrinkShape) that is implemented seamlessly in the free and open-source R and SAGA GIS environments. This free, efficient, and accessible environment provides a multi-dimensional, rotation invariant, scale-controlled method for characterizing and comparing planar shapes and tools for visualizing results. ShrinkShape2 incrementally buffers polygons internally until the point of extinction; at each shrinking phase a series of summary metrics are computed to form spectra of index values that characterize general characteristics of shape structure and complexity. Demonstrations with simple synthetic shapes, political boundary maps, and a sample sub-Arctic pond demonstrate the ability to assess boundary complexity, pinch-points, and identify holes within larger shapes. For perfectly convex and compact shapes, perimeter decompositions will decrease linearly and area decompositions will decrease according to one limb of a second-order polynomial function relative to the cumulative shrinking distance. Resume La boite a outils ShrinkShape2 libre et a code source ouvert pour la caracterisation insensible a la rotation et la comparaison de forme des polygones La delimitation des espaces geographiques fragmente les paysages en regions qui manifestent une certaine homogeneite interne. De nombreux travaux ont ete realises afin d'analyser les formes que generent ces delimitations, ce qui requiert le developpement des moyens pour mesurer, caracteriser et comparer les formes de polygones planaires. J'ai donc mis au point la boite a outils ShrinkShape2 qui est constituee d'un ensemble d'outils entierement repenses en fonction de l'utilisateur (une amelioration de son predecesseur ShrinkShape) et est integree en langage R et dans l'environnement a code source ouvert du SIG SAGA . Cet environnement gratuit, efficace, et accessible fournit une methode multi-dimensionnelle, insensible a la rotation et a echelle controlee afin de caracteriser et de comparer des formes planes ainsi que des outils pour la visualisation des resultats. ShrinkShape2 construit des zones tampon a l'interieur des polygones de maniere recursive jusqu'au stade d'extinction. Un ensemble de parametres de synthese est calcule a chaque phase de reduction pour former des spectres d'indices qui caracterisent les composantes structurelles generales et la complexite de la forme. Des exemples de formes simples et synthetiques, de cartes des limites politiques, et d'un etang subarctique illustrent sa capacite pour evaluer la complexite des limites et des goulots d'etranglement, et pour identifier des trous dans les grandes regions. Lors de la decomposition de formes parfaitement convexes et compactes, les perimetres diminuent lineairement et les superficies diminuent selon une fonction polynomiale de second degre de la distance de contraction cumulee.

Background— The aim of this study is to examine the relationship between time in the therapeutic range (TTR) and clinical outcomes in heart failure patients in sinus rhythm treated with warfarin. Methods and Results— We used data from the Warfarin versus Aspirin in Reduced Cardiac Ejection Fraction (WARCEF) trial to assess the relationship of TTR with the WARCEF primary outcome (ischemic stroke, intracerebral hemorrhage, or death), with death alone, ischemic stroke alone, major hemorrhage alone, and net clinical benefit (primary outcome and major hemorrhage combined). Multivariable Cox models were used to examine how the event risk changed with TTR and to compare the high TTR, low TTR, and aspirin-treated patients, with TTR being treated as a time-dependent covariate. A total of 2217 patients were included in the analyses; among whom 1067 were randomized to warfarin and 1150 were randomized to aspirin. The median (interquartile range) follow-up duration was 3.6 (2.0–5.0) years. Mean (±SD) age was 61±11.3 years, with 80% being men. The mean (±SD) TTR was 57% (±28.5%). Increasing TTR was significantly associated with reduction in primary outcome (adjusted P P =0.001), and improved net clinical benefit (adjusted P P =0.082 for ischemic stroke and adjusted P =0.109 for major hemorrhage). Conclusions— In patients with heart failure in sinus rhythm, increasing TTR is associated with better outcome and improved net clinical benefit. Patients in whom good quality anticoagulation can be achieved may benefit from the use of anticoagulants. Clinical Trial Registration— URL: http://www.clinicaltrials.gov . Unique identifier: NCT00041938.

We present a rigorous and simple approach for the comparison of binary landscapes by class-focused metric values that complements the ease of computing these metrics for landscape ecology research. First, we assess whether a class-focused pattern metric value could have emerged due to random chance. Second, we compare two landscapes and assess whether class-focused pattern metrics computed for each landscape are significantly different or not. Our frameworks are based on conditional autoregressive simulations to derive empirical distributions for each metric where composition and configuration parameters are controlled. Our method permits the computation of probabilities that an observed metric value is either greater than or less than a given level of expectation. We also provide means for situating any landscape on a selected pattern metric-surface defined by parameters of composition and configuration. These surfaces illustrate which parameter would be most easily adjusted to effect a desired change in a selected class-focused pattern metric’s value. Implementation is fully within the R statistical computing environment.

Categorical recognition of a tree's genus is known to be valuable information for the effective management of forest inventories. In this paper, we present a method for learning a discriminative model using Random Forests to classify individual trees into three genera: pine, poplar, and maple. We believe that both internal and external geometric characteristics of the tree crown are related to tree form and therefore useful in classifying trees to the genus level. Our approach involves the extraction of both internal and external geometric features from a LiDAR point cloud as we believe that geometric features provide important information about the organization of the points inside the tree crown along with overall tree shape and form. We developed 24 geometric features and then reduced the number of features to increase efficiency. These geometric characteristics, computed for 160 sampled trees from eight field sites, were classified using Random Forests and achieved an 88.3% average accuracy rate by us...

Characterizing landscape spatial structure can provide insights about the underlying mechanisms that generate pattern. Quantifying spatial structure enables analysis of landscape change over time as well as comparisons among different locations. Although numerous landscape metrics (LMs) exist to quantify spatial structure and characterize a landscape, how do we know when two landscapes significantly differ? As a single landscape represents only one replicate, its metrics are not statistics; thus, testing for differences between two landscapes becomes difficult. To address this problem, randomization procedures can help assess statistical significance using simulation approaches that assess whether the observed spatial structure could have occurred by chance alone. In this chapter, exercises will allow students to accomplish the following objectives.

This chapter provides an introduction and overview of using light detection and ranging (LiDAR) in forest applications. The first section explains the principles and basic terminology for LiDAR and introduces the use of LiDAR on three different platforms (spaceborne, airborne, and terrestrial) for forest applications. The second section discusses applications in relation to the primary measurements from a LiDAR point cloud, primarily information derived from distance (from the aircraft to the target). We cover concepts related to different representations of surfaces (e.g., digital surface model, digital terrain model, digital elevation model, and canopy height model). Typically, single trees can be identified from the canopy height model and there are two different ways to assign LiDAR points to individual trees, the surface-based method and the point-based method. The third section discusses forest applications in relation to secondary measurement from a LiDAR point cloud, information derived from point cloud geometry rather than direct distance measurements. This section covers tree genera classification; the use of allometric equations for deriving DBH, biomass, and other forest attributes; and the classification of vegetation types. Three ways of getting genera information are discussed, including the vertical profile method, methods relying on geometry derived from individual tree point clouds, and methods that incorporate spectral information. The fourth section provides a case study for identifying potential tree hazards along a powerline corridor in Ontario, Canada. We conclude by discussing the future of this technology.

In this chapter, we offer a primer that defines key terminology and that positions forest landscape mapping within a broad context that encompasses all readers of this book. We present the fundamentals of cartography, and emphasize the importance of data representations, map projections, scales, and data collection options and principles. We then formalize the term forest landscape, and clarify how to understand this term in the context of this book by considering the numerous characteristics that could be mapped and the influence of scale on this mapping. Next, we focus on mapping forest landscapes from the perspectives of both regions and boundaries between them by considering the fuzziness of both; we then extend these concepts beyond two dimensions. The chapter concludes with a long discussion of map utility and how maps can be interpreted, mined for information, and how scale affects these types of interpretations. We stress that data with higher spatial resolution is not always better and that multiple-scale and cross-scale analyses may yield more meaningful information than interpreting a landscape at only a single scale. We conclude with a summary and discussion of the assessment of accuracy, error sources, and overall map validation. Throughout, we draw attention to chapters in this book that advance the discussion of the topics introduced in this chapter.

Boreal wildfires are characterized by internal heterogeneity that arises from variations in fuel availability, fuel moisture content, and weather conditions during a fire event. This heterogeneity extends from an uneven burn intensity that affects the degree of forest disturbance to inconsistency in boundary abruptness at the fire perimeter, in spot fires associated with the main fire, and in areas internal to the fire where residual vegetation and unburnable land cover types are encountered. We begin with a brief discussion of wildfire anatomy and how fires burn to create new and complex landscape patterns. We then describe some of the common approaches that are used to map wildfires, paying particular attention to the importance of scale in the mapping process. We address the complexities and heterogeneities of wildfire boundaries and internal structures by consistently linking their characterization and interpretation to spatial scale and statistical characteristics of mapping. Having outlined the variability in the formation and mapping of wildfire complexity, we propose a standardized terminology for describing these phenomena and provide some thoughts on the future of efforts to map dynamic and complex landscape entities such as wildfires.

Significant advances in geospatial sciences are being made, aided by multibillion dollar technological industries that focus on remote sensing, geographic information systems (GIS), global positioning system (GPS), and computing. As a result of these improvements, forest landscape ecologists and forest managers now have access to increasingly comprehensive and accurate georeferenced information that is frequently updated. However, the knowledge of forest landscape patterns and processes and the ability to convey this knowledge for applications have not grown in proportion to the advances in mapping technology. Thus, there are vast opportunities to improve the knowledge of forest landscape patterns, especially in remote regions that are difficult to study in the field. There are numerous avenues to improve the efficacy and efficiency of map use. Here we highlight the importance of resisting the allure of excessive detail and instead focus on identifying the most appropriate scale, which may not be the most detailed resolution available, to support a user’s research or management goals. It is also crucial to improve our awareness of the many embedded assumptions that shape how information is mapped and the sources and magnitude of the errors that are inherent to any map, however advanced the science and technology that produced it. We emphasize that ongoing communication and interaction between the communities of map developers and map users will be essential to achieve wise use of forest landscape maps.

We present a comparative study between two different approaches for tree genera classification using descriptors derived from tree geometry and those derived from the vertical profile analysis of LiDAR point data. The different methods provide two perspectives for processing LiDAR point clouds for tree genera identification. The geometric perspective analyzes individual tree crowns in relation to valuable information related to characteristics of clusters and line segments derived within crowns and overall tree shapes to highlight the spatial distribution of LiDAR points within the crown. Conversely, analyzing vertical profiles retrieves information about the point distributions with respect to height percentiles; this perspective emphasizes of the importance that point distributions at specific heights express, accommodating for the decreased point density with respect to depth of canopy penetration by LiDAR pulses. The targeted species include white birch, maple, oak, poplar, white pine and jack pine at a study site northeast of Sault Ste. Marie, Ontario, Canada.

Geographic data collection, manipulation, analysis and visualization options have experienced substantial improvements during the past several decades, largely spurred by advancements in computing capabilities. While geographers are often credited with identifying and expanding many of the emerging application areas and innovations for the analysis of spatially (and sometimes temporally) referenced data, we are specifically interested in the role of Canadian geographers in the rapidly evolving domain of spatial science. We pose the following provocative question with the intent of not only summarizing the Canadian literature, but also to stimulate an informed discussion: ‘are Canadian geographers developers or users of spatial analytical methods?’ We review the refereed literature from 1980 to 2008 to describe the nature of contributions by Canadian geographers, beginning at about the time of widely accessible computing (1980s). Our summary broadly classifies subdisciplinary contribution areas as being best described as GIS, remote sensing, or spatial statistics, while each contribution area may take the type of algorithm development, advancement and synthesis of theory, or the application of existing methods. We paint a picture of the current contribution landscape and reflect on significant achievements while commenting on some potential weakness that with increased resources and focus might become future realms of advancement. La recherche en sciences d'analyse spatiales : quel est l’apport des geographes canadiens? Les demarches de collecte, de traitement, d’analyse et de representation visuelle des donnees geographiques ont connu des ameliorations remarquables au cours des dernieres decennies, essentiellement grâce aux progres realises en informatique. Si de maniere generale on attribue aux geographes la conception et le developpement de nouveaux logiciels et la mise au point de methodes d’analyse spatiale (et dans certains cas, temporelle), nous nous demandons a quel point les geographes canadiens font figure de modeles en matiere de recherche spatiale. La question susceptible de soulever la controverse que nous nous posons est de savoir si les geographes canadiens sont des inventeurs ou simplement des utilisateurs des methodes d’analyse spatiale. L’objectif vise est double : dresser une synthese de la litterature canadienne et susciter une reflexion eclairee. Les articles scientifiques publies par les pairs et parus entre 1980 et 2008 sont passes en revue. Cette analyse porte sur le corpus des textes publies par des geographes canadiens depuis les annees 1980 a l’epoque de l'emploi des outils informatiques. Elle permet de repertorier les articles en fonction d’un ensemble de sous-disciplines qui renvoient aux SIG, a la teledetection, ou aux statistiques spatiales. De plus, chacun de ces champs de recherche s’inscrit dans la cadre du developpement d'algorithmes, de l’avancement et de la demonstration theoriques, ou de l’application de methodes existantes. Nous esquissons un portrait de la situation actuelle du champ d’etude et nous nous penchons sur les realisations marquantes. Enfin, nous formulons des observations sur les faiblesses potentielles qui, en engageant plus de ressources et en leur accordant la priorite, pourraient eventuellement contribuer a l’avancement du domaine.

A 4-month controlled laboratory experiment was conducted to determine the impact of spatial configurational change on spectral characteristics of grasses. Eight different planting configurations were examined of two grass seed mixtures (sun and shade grass and Kentucky bluegrass) of equal composition (50% each). Spectral properties were quantified using (i) leaf area index (LAI), (ii) normalized difference vegetation index (NDVI) and soil-adjusted vegetation index (SAVI), and (iii) an unsupervised iterative self-organizing data analysis (ISODATA) classification. Analysis of variance (ANOVA) tests were performed for LAI, NDVI, and SAVI; Tukey’s post-hoc tests showed significant differences between random and all other treatments for LAI means, no significant differences for NDVI means, and significant differences between irregularly patchy and regularly patchy treatments for SAVI means. Chi-square tests on the four-class aggregation results for ISODATA indicated that configuration does influence the spectr...

The simple and intuitive nature of the coincidence matrix has not only made it the current “gold standard” for accuracy assessment (based on a sample of map pixels), but also a common tool for describing difference between two categorical maps (when all pixels are enumerated). It is this latter case of map comparison that this article explores. Coincidence matrices, although providing significant information regarding thematic agreement between two categorical maps (composition), can lack significantly in terms of conveying information about differences or similarities in the spatial arrangement (configuration) of those map categories in geographic space. This article introduces means for distilling the available configuration information from a coincidence matrix while demonstrating some simple categorical map comparisons. Specifically, while the coincidence matrix summarizes per-pixel compositional persistence or change, the introduced technique further quantifies the global and local configurational uncertainty between compared maps. I demonstrate how this quantification of configurational uncertainty can be used to gauge which thematic mismatch types are most significant and how to measure/present local configurational uncertainty in a spatial context. Implementation is through a straightforward mathematical algorithm in R that is illustrated by several examples. La sencillez y caracteristicas las matrices de confusion (tablas de contingencia o de error) no solo la ha convertido en (1) el estandar por excelencia para la evaluacion de confiabilidad (veracidad o validez) de mapas tematicos en escala nominal (en base de una muestra de pixeles en un mapa), sino tambien en (2) una herramienta comun para describir la diferencia entre dichos mapas (cuando todos los pixeles son tomados en cuenta). Este segundo uso de las matrices de confusion es el tema explorado por el presente articulo. A pesar de proporcionar informacion importante acerca de la confiabilidad tematica entre dos mapas de categorias nominales (composicion), las matrices de confusion sufren de limitaciones importantes cuando el objetivo es extraer informacion acerca del patron o arreglo espacial (configuracion) de las categorias del mapa. El presente articulo presenta herramientas para destilar la informacion de configuracion espacial disponible a partir de una matriz de confusion, y al mismo tiempo, ilustra algunas comparaciones entre mapas de categorias nominales. En terminos mas concretos, la matriz de confusion resume la persistencia o cambio en la composicion a nivel de cada pixel. La nueva tecnica presentada en este estudio incluye ademas la cuantificacion de la incertidumbre en la configuracion de los mapas comparados. El autor demuestra como esta cuantificacion puede ser utilizada para darse una idea de cuales tipos de errores tematicos son mas importantes, y como se puede medir e ilustrar la incertidumbre de la configuracion local en el contexto espacial. La implementacion del metodo se realiza a traves de un algoritmo matematico sencillo en lenguaje R, el mismo que es ilustrado con varios ejemplos. 一致性矩阵具有的简单性和直观性，不但使其成为当前准确度评估（基于图像像素样本的）的“黄金标准”，而且也是描述两类图像差异（即枚举所有像素时）的通用工具。本文即是探索后一种情况下的图像比较。尽管一致性矩阵提供了关于两类图像（组成）之间主题一致性的重要信息，但明显缺乏针对地理空间中的不同类别图像的空间排列（构型）所传达的相似性和差异性信息。本文介绍了从一个一致性矩阵中提取出有效的构型信息的方法，同时也展示了一些简单的分类图像的对比。具体而言，当一致性矩阵概括出每个像素成份的连续性或变化时，所引入的方法可以进一步量化对比图像之间的全局和局域构型的不确定性。笔者论证了如何将这种构形不确定性的量化方法用于测度哪些是最显著的错配专题类型；以及如何来测度/表达某个空间范围内局域构形的不确定性。本方法采用R语言中的一个简易数学算法加以实现，并通过几个实例予以展示。

The current provincial-extent digital elevation model (DEM) and corresponding hydrological maps for Ontario have been produced using traditional photogrammetry and aerial photograph interpretation. This process is labour-intensive and requires visual interpretation of stereo image pairs. The ground surface and small hydrological features may be inaccurately delineated in areas where vegetation is dense or the ground is otherwise shielded from aerial view. In an effort to improve and automate delineation of hydrological features, we examined the behaviour and final products of the D8 flowrouting algorithm in 2 software environments (TAS and TauDEM for ArcGIS) operating on a high spatial resolution DEM derived using canopy-penetrating light detection and ranging (LiDAR) technology in a pilot study in the Romeo Malette Forest (41.25°N, 81.50°W). Filtered LiDAR data points (5-m spacing) were interpolated using IDW, TIN, and splines, each resulting in a 2.5-m spatial resolution DEM. Results demonstrate improved realism in the characterization of surficial hydrology by LIDAR derived products as compared to applying identical algorithms on existing coarser provincial data. Benefits include the ability to represent streams of lower Strahler order to define crisp watershed boundaries, and the more accurate identification of local depressions that form potentially wet sites. This approach identifies wet sites that should be avoided during forest operations (e.g., skidder traffic) and can provide additional information for trail layout, road planning, and water crossings. By increasing the number of uses of LiDAR, the capital investment in these data becomes increasingly palatable for forest companies interested in obtaining detailed plans of their forest holdings. Key words: LiDAR, DEM, OBM, spatial resolution, interpolation, Strahler stream order, flow routing, topographic wetness

Studies of urban travel behaviour typically focus on weekday activities and commuting. This is surprising given the rising contribution of discretionary activities to daily travel that has occurred during the last few decades. Moreover, current understanding of the relationship between travel behaviour and land use remains incomplete, with little research carried out to explore spatial properties of activity-travel behaviour during the off-peak and weekend time periods. Weekend behaviours, for example, influenced by the availability of time and the spatiotemporal distribution of “weekend” destinations, likely produce spatially and temporally distinct activity-travel patterns. Using data from the first wave of the Toronto Travel-Activity Panel Survey (TTAPS), this paper examines an area of research that has received little attention; namely, the presence of spatial variety in activity-travel behaviour. The paper begins by looking at the extent to which individuals engage in spatially repetitive location choices during the course of a single week. Area-based measures of geographical extent and activity dispersion are then used to expose differences in weekday-to-weekend and day-to-day activity-travel patterns. Examination of unclassified activities carried out over a 1 week period reveals a level of spatial repetition that does not materialise across activities classified by type, travel mode, and planning strategy. Despite the inherent spatial flexibility offered by the personal automobile, spatial repetition is also found to be surprisingly similar across travel modes. The results also indicate weekday-to-weekend, and day-to-day fluctuations in spatial properties of individual activity-travel behaviour. These findings challenge the utility of the short-run survey as an instrument for capturing archetypal patterns of spatial behaviour. In addition, the presence of a weekday-to-weekend differential in spatial behaviour suggests that policies targeting weekday travel reduction could have little impact on travel associated with weekend activities.

Most image classification algorithms rely on computing the distance between the unique spectral signature of a given pixel and a set of possible clusters within an n-dimensional feature space that represents discrete land cover categories. Each scrutinized pixel will ultimately be closest to one of the predefined clusters; different classification algorithms differ in the details of which cluster is considered as closest or most likely, but in general the selected algorithm will label each pixel with the label of the closest cluster. However, pixels expressing virtually identical distances to two or more clusters identify a limitation of this typical classification approach. Conditions for limitations to distance based classification algorithms include when distances are long and the pixel may not clearly belong to any single category, may represent mixed land cover, or can be easily confused spectrally between two or more categories. We propose that retention of the distance to the second closest cluster can shed light on the confidence with which label assignment proceeds and present several examples of how such additional information might enhance accuracy assessments and improve classification confidence. The method was developed with simplicity as a goal, assuming the classification has already been performed, and standard clustering reports are available. Over a test site in central British Columbia, Canada, we illustrate the described technique using classified image data from a nation-wide land cover mapping project. Calculation of multi-spectral Euclidean distances to cluster centroids, standardized by cluster variance, allows comparison of all potential class assignments within a unified framework. The variable distances provide a measure of relative confidence in the actual classification at the level of individual pixels.

GIScience 2016 Short Paper Proceedings Shape and resolution: quantifying feature morphology change due to coarsening spatial resolution using UAV- based images from vertical transects S. W. Mitchell 1 , T. K. Remmel 2 Department of Geography & Environmental Studies, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario, K1S 5B6 Email: Scott.Mitchell@carleton.ca Department of Geography, York University, 4700 Keele Street, Toronto, Ontario, M3J 1P3 Email: remmelt@yorku.ca Abstract Aside from tone or colour, shape is one of the most readily utilized characteristics to interpret remotely sensed imagery. As spatial resolution is coarsened, the level of observable spatial detail diminishes, thus the shapes of features become less distinct. This study controlled the observation platform, a Tetracam ADC snap multispectral camera mounted on a Phantom 2 quad-copter, to collect images along vertical transects. This design simulates continuously coarsening spatial resolution as the aircraft ascends, while maintaining consistent imaging parameters. Vegetation visible on all scenes was analysed using common shape characterizations. Scaling relationships were assessed between original shape complexity and spatial resolution (a proxy for scale). Initial results show strong scaling relationships at ultra- fine scales but an interesting area of stability which, if repeatable, could have important implications for the use of UAVs in environmental research. 1. Introduction The characterization and comparison of spatial pattern characteristics or landscape morphological elements are sectors of GIScience that intersect with mathematics and landscape ecology, and have been developing for several decades (Gardner et al. 1987; Haines-Young and Chopping 1996; Uuemaa 2009). Numerous studies span the breadth of such inquiries, ranging from studies of land cover change (Linke et al. 2009) to assessing habitat suitability (McAlpine and Eyre 2002). While computing a suite of metrics is facilitated by readily accessible software (Baker and Cai 1992; McGarigal and Marks 1995; Vogt et al. 2007), many redundancies have been identified (Riitters et al. 1995) and limitations explicitly characterized that limit the inference that can be drawn from such metrics (Remmel and Csillag 2003). Scale effects have been tested by manipulating landscape extent and spatial resolution, but these generally rely on resampling or the use of multiple platforms of data (e.g. Saura 2002; Baldwin et al. 2004). This study focuses on the characterization of planar shape, a single aspect of the broader collection of pattern metrics, with respect to scale. Shape provides the ability to characterize and discriminate individual landscape elements (Zhang and Atkinson 2016), rather than summarize landscape-wide patterns. Scaling analysis was applied to data obtained from a consistent platform, such that resampling or sensor substitution did not introduce spurious relationships and uncertainty that make interpretation more complex. By manipulating a single variable (spatial resolution) of a digital image while maintaining all camera, timing, and environmental conditions constant, we explored the effect of spatial resolution on the quantification of patch shape directly. This differs from Saura and Carballal (2004) where

We revisit one of the classical problems in geography and cartography where multiple observations on a lattice (N) need to be grouped into many fewer regions (G), especially when this number of desired regions is unknown a priori. Since an optimization through all possible aggregations is not feasible, a hierarchical classification scheme is proposed with an objective function sensitive to spatial pattern. The objective function to be minimized during the assignment of observations to regions (classification) consists of two terms: the first characterizes accuracy and the second, model complexity. For the latter, we introduce a spatial measure that characterizes the number of homogeneous patches rather than the usual number of classes. A simulation study shows that such a classification procedure is less sensitive to random and spatially correlated error (noise) than non-spatial classification. We also show that for conditional autoregressive error (noise) fields the optimal partitioning is the one that has the highest within-units generalized Moran coefficient. The classifier is implemented in ArcView to demonstrate both a socio-economic and an environmental application to illustrate some potential applications.

The continual accumulation of categorical data sets, presented as nominal categories mapped onto regular grids, provides for the increased desire to compare the patterns observed between these maps. We present a measurement scheme for the comparison of categorical maps that decomposes the differences in multidimensional nested coincidence tables according to variables that record occurrence frequencies of categories (Z), at levels of spatial aggregation (Y), on specific maps (X). Sequences of conditional entropies computed according to the specific questions asked (e.g. is there coincidence between colours and locations), characterize the correspondence between the three types of variables in common units (bits) measured by mutual information. The form of these sequences, as a variable runs from coarse to fine detail, referred to as spectra, provide meaningful characterizations of the similarities/differences between categorical maps, including their spatial configuration. We introduce the information the...

We present a simple simulation scheme to derive confidence intervals for measures computed based on a coincidence matrix. Our approach is based on the conditional distributions between two categorical maps, and is a direct interpretation of how much information one map (usually the classified image) carries about the other map (usually the reference image). The simulation algorithm creates a realization of the map created by visiting each pixel and drawing a random sample from the conditional distribution of reference categories (conditioned on the category of the pixel of the classified image). Confidence intervals can be derived by repeating the simulation many times and computing the coincidence measure(s). Handling the coincidence matrix as a set of conditional distributions also allows the comparison of maps with different numbers of categories. This approach is an extension of the traditional methodology widely used in accuracy assessment of data derived from remotely sensed images. We illustrate th...

Identifying appropriate validation sources for large-area land cover products is a challenge, with logistical constraints frequently necessitating the use of preexisting data sources. Several issues exist when comparing polygon (vector-based) datasets to raster imagery: geolocational mismatches, differences in features or classes mapped, disparity between the scale of polygon delineation and the spatial resolution of the image, and temporal discrepancies. To evaluate the potential impact of using vector coverages to assess the accuracy of pixel-based land cover maps, five evaluation protocols are applied to test sites located in British Columbia and Newfoundland and Labrador, Canada. One protocol directly compared the land cover of the sample unit to the land cover of the forest inventory polygon within which the sample unit fell, two protocols used different regions around the sample unit to define the land cover class, and two protocols were based on homogeneity criteria that restricted the selection of...

Canada's ability to sustainably manage approximately 10% of the global forest cover is a critical environmental and economic issue. The capacity to meet such demands and to deliver on national and international commitments regarding forest management is enabled through collaboration between federal, provincial, and territorial agencies. A principal collaborator is the National Forest Inventory (NFI); a systematic photo-plot based monitoring system designed specifically for reporting purposes and as an important input for scientific models. Satellite imagery is illustrated here as a support data set to ensure the quality of the NFI, for auditing the photo-plot contents, and to detect spatial biases. The Canadian Forest Service, in collaboration with the Canadian Space Agency and other federal and provincial agencies, is producing a national land cover database of the forested area of Canada (Earth Observation for Sustainable Development of Forests (EOSD)) using Landsat-7 ETM+ data for circa 2000 conditions. The integration between the plot-based NFI with classified EOSD data is presented for central British Columbia, an area comprising 6 Landsat scenes and 324 2 km × 2 km photo-plots. Traditional accuracy assessment measures based on the analysis of coincidence matrices are reported as levels of agreement for hierarchically aggregated land cover categories (overall agreements of 91%, 79%, 64% and 26% for 3, 4, 6 and 20 classes respectively) to demonstrate coincidence between the different data products. Local agreement between NFI and EOSD is demonstrated as a means of photo-plot auditing while spatial biases are detected through investigations of geographic pattern in the coincidence values. The illustrated approaches may be expanded or applied to different mapped attributes (e.g., biomass) that are of utility to those attempting to characterize large areas in a consistent and rigorous fashion.

Spatial stochastic models play an important role in understanding and predicting the behaviour of complex systems. Such models may be implemented with explicit knowledge of only a limited number of parameters relating to spatial relationships among locations. Consequently, they are often used instead of deterministic-mechanistic models, which may potentially require an unrealistically large number of parameters. Currently, in contrast to spatial stochastic models, the parameterization of the joint spatial distribution of objects in landscape models is more often implicit than explicit. Here, we investigate the similarities and differences between bona fide spatial stochastic models and landscape models by focusing mostly on the relationships between processes, their realizations (patterns), representation and measurement, and their use in exploratory as well as confirmatory data analysis. One of the most important outcomes of recognizing the importance of stochastic processes is the acknowledgement that the spatial pattern observed in a landscape is only one realization of that process. Hence, while ecologists have been using landscape pattern indices (LPIs) to characterize landscape heterogeneity and/or make inferences about processes shaping the landscape, no stochastic modelling framework has been developed for their proper statistical elucidation. Consequently, several (mis)uses of LPIs draw conclusions about landscapes which are suspect. We show that several reports about sensitivities of LPIs to measurements have common roots that can be made explicitly manageable by adopting stochastic models of spatial structure. The key parameters of these stochastic models are composition and configuration, which, in general, cannot be estimated independently from each other. We outline how to develop the stochastic framework to interpret observations and make some recommendations to practitioners about everyday usage. The conceptual linkages between patterns and processes are particularly important in light of recent efforts to bridge the static-structural and the dynamic-analytic traditions of ecology.

Understanding natural fire regimes is crucial to developing harvesting scenarios and conducting sustainable resource management in the boreal forest. To gain this understanding, resource professionals need efficient and cost-effective data collection methods that can operate over vast and isolated landscapes. We compared three Advanced Very High Resolution Radiometer (AVHRR)/Normalized Difference Vegetation Index (NDVI) methods of fire detection and mapping for a case study in northern Ontario, Canada, of the 1992, 1993, and 1995 fire seasons. Fire mapping accuracy was assessed by the spatial coincidence between mapped fires and ground-truthed information using a decision-tree approach and by testing the hypothesis that various calculated accuracy components were equal within an ANOVA design. Ground-truthed fire sizes and shapes were correlated with the AVHRR/NDVI-mapped areas; however, light cloud contamination increased false detection of fires due to NDVI suppression. The multiple threshold technique of change detection provided better estimates of fire areas than did single threshold methods.

Maps of tree genera are useful in applications including forest inventory, urban planning, and the maintenance of utility transmission line infrastructure. We present a case study of using high density airborne LiDAR data for tree genera mapping along the right of way (ROW) of a utility transmission line corridor. Our goal was to identify single trees that showed or posed potential threats to transmission line infrastructure. Using the three dimensional mapping capability of LiDAR, we derived tree metrics that are related to the geometry of the trees (tree forms). For example, the dominant growth direction of trees is useful in identifying trees that are leaning towards transmission lines. We also derived other geometric indices that are useful in determining tree genera; these metrics included their height, crown shape, size, and branching structures. Our pilot study was situated north of Thessalon, Ontario, Canada along a major utility corridor ROW and surrounding woodlots. The geometric features used for general classification could be categorized into five broad categories related to: 1) lines, 2) clusters, 3) volumes, 4) 3D buffers of points, and 5) overall tree shape that provide parameters as an input for the Random Forest classifier., Los mapas de géneros de árboles son útiles para el inventario forestal, planificación urbana y el mantenimiento de la infraestructura de líneas de transmisión. Se presenta un estudio de caso de uso de datos LiDAR de alta densidad para el mapeo de géneros de árboles a lo largo del derecho de paso (ROW) de un corredor de línea de transmisión. El objetivo de la investigación fue identificar árboles individuales que mostraban o poseían una amenaza potencial a la infraestructura de la línea de transmisión. Mediante el uso de mapas tridimensionales de LiDAR se derivaron métricas de árboles que están relacionadas con la geometría de éstos (formas del árbol). Por ejemplo, la dirección del crecimiento dominante de los árboles es útil para identificar árboles que crecen inclinados hacia las líneas de transmisión. También se derivaron otras métricas geométricas que son útiles para determinar los géneros de los árboles, tales como altura, forma de la copa, tamaño y estructura de ramas. El área de estudio se ubicó al norte de Thessalon, Ontario, Canadá, a lo largo de los principales corredores de ROW y en los bosques aledaños. Los atributos geométricos usados para la clasificación de los géneros fueron categorizados en cinco amplias clases: 1) líneas, 2) agrupamiento, 3) volúmenes, 4) amortiguamiento en 3D de puntos, y 5) forma general del árbol que provee parámetros como una entrada para el clasificador forestal aleatorio.

Background: Previous studies have demonstrated the importance of hypercholesterolemia and hyperhomocysteinemia as risk factors for stroke. Hypothyroidism has been shown to be associated with hypercholesterolemia and homocysteinemia, but has not been elucidated as a risk factor for stroke. Therefore we studied the association of hypothyroidism in transient ischemic attack (TIA) and ischemic stroke patients. Methods: Patients admitted to the University of Louisville with a diagnosis of TIA or ischemic stroke (IS) between January 2001 and September 2003 were evaluated for stroke risk factors. Subjects underwent stroke work-up which included laboratory evaluations of thyroid stimulating hormone (TSH), lipid panals, and homocysteine (hCys). If TSH was abnormal, further thyroid profiles were obtained. Results: Out of 406 subjects (age range 18 - 95, mean age 68, 205 men) admitted with TIA or ischemic stroke, 305 (75%) had TSH levels recorded. Thirty-three (10.8%) were diagnosed with hThy. Eighteen (54.5%) of these hThy subjects had hyperhCys. HCys level was obtained in 227 (55.9%) of the subjects. Three (9%)of the hThy subjects had hyperhCys. Conclusion: Our data clearly support the association between hThy and TIA or IS. A systematic investigation of stroke risk, comparing hThy to non-hThy subjects, contolling for other known risk factors is needed to further elucidate this relationship.