Les travaux menés dans cette thèse concernent le développement d'un dispositif de vision permettant l'estimation de caractéristiques géométriques d'un feu de végétation en propagation. Ce dispositif est composé de plusieurs systèmes de stéréovision multimodaux générant des paires d'images stéréoscopiques à partir desquelles des points tridimensionnels sont calculés et les caractéristiques géométriques de feu tels que sa position, vitesse, hauteur, profondeur, inclinaison, surface et volume sont estimées. La première contribution importante de cette thèse est la détection de pixels de feu de végétation. Tous les algorithmes de détection de pixels de feu de la littérature ainsi que ceux développés dans le cadre de cette thèse ont été évalués sur une base de 500 images de feux de végétation acquises dans le domaine du visible et caractérisées en fonction des propriétés du feu dans l'image (couleur, fumée, luminosité). Cinq algorithmes de détection de pixels de feu de végétation basés sur la fusion de données issues d'images acquises dans le domaine du visible et du proche-infrarouge ont également été développés et évalués sur une autre base de données composée de 100 images multimodales caractérisées. La deuxième contribution importante de cette thèse concerne l'utilisation de méthodes de fusion d'images pour l'optimisation des points appariés entre les images multimodales stéréoscopiques.La troisième contribution importante de cette thèse est l'estimation des caractéristiques géométriques de feu à partir de points tridimensionnels obtenus depuis plusieurs paires d'images stéréoscopiques et recalés à l'aide de relevés GPS et d'inclinaison de tous les dispositifs de vision.Le dispositif d'estimation de caractéristiques géométriques à partir de systèmes de stéréovision a été évalué sur des objets rigides de dimensions connues et a permis d'obtenir les informations souhaitées avec une bonne précision. Les résultats des données obtenues pour des feux de végétation en propagation sont aussi présentés. / This thesis presents the geometrical characteristics measurement of spreading vegetation fires with multimodal stereovision systems. Image processing and 3D registration are used in order to obtain a three-dimensional modeling of the fire at each instant of image acquisition and then to compute fire front characteristics like its position, its rate of spread, its height, its width, its inclination, its surface and its volume. The first important contribution of this thesis is the fire pixel detection. A benchmark of fire pixel detection algorithms and of those that are developed in this thesis have been on a database of 500 vegetation fire images of the visible spectra which have been characterized according to the fire properties in the image (color, smoke, luminosity). Five fire pixel detection algorithms based on fusion of data from visible and near-infrared spectra images have also been developed and tested on another database of 100 multimodal images. The second important contribution of this thesis is about the use of images fusion for the optimization of the matching point’s number between the multimodal stereo images.The second important contribution of this thesis is the registration method of 3D fire points obtained with stereovision systems. It uses information collected from a housing containing a GPS and an IMU card which is positioned on each stereovision systems. With this registration, a method have been developed to extract the geometrical characteristics when the fire is spreading.The geometrical characteristics estimation device have been evaluated on a car of known dimensions and the results obtained confirm the good accuracy of the device. The results obtained from vegetation fires are also presented.

Degraded pollinator habitat may have far-reaching consequences for recovery of imperiled flowering plant populations. Studies indicate that broad-spectrum insecticides used in mosquito abatement are detrimental to non-target invertebrates, including pollinators. A decline in efficient pollinators can reduce plant fitness by decreasing reproductive output and constraining genetic diversity, a challenge for rare plants. In 2015, I monitored flower visitation and fruit set of three imperiled plant species throughout protected areas on three islands in The Lower Florida Keys. These islands consist of conservation land fragmented by intermittently dispersed residential neighborhoods that seasonally spray insecticides for mosquito control. Flowers open at treatment sites had decreased flower visitor activity following insecticide applications, but only species that require invertebrate agents for pollen transfer had significantly reduced fruit set. Implications of mosquito insecticides near conservation lands may pose immediate threats to invertebrate pollinators and flowering plants that require pollinators for reproduction, although long-term threats to genetic diversity are unknown for automatic self-pollinating species.

Wildfire hazard is increasing in much of the United States, posing a threat to human communities and natural ecosystem services, especially in areas at the wildland-urban interface. There are steps people can take to reduce wildfire hazard, but often they are not used. Understanding and addressing human perceptions of wildfire risk and of risk-mitigating behaviors requires knowledge of both social and ecological systems. To better understand this complex issue, three types of factors must be addressed: social, cultural, demographic, and biophysical. This dissertation incorporates these three essential factors to intensively investigate the risk perception and behaviors of residents living in wildland-urban interface communities in three states (Arizona, California, and New Mexico). The first study examines the effect that individual risk perceptions have on intention to mitigate wildfire risk by integrating two social-psychological theories, Theory of Planned Behavior and Cultural Theory, to investigate the causal relationship and motivational factors that influence the intention to mitigate wildfire hazard. Results suggest that attitudes toward wildfire mitigation practices and perceived behavioral control play a significant role in the decision process. The effect of an individual’s orientation toward nature is mediated by attitude and perceived behavioral control. It is important that these orientations are taken into consideration when designing strategies to increase incentives to mitigate fire risk. The second study explores the linkage between property owners’ perception of risk and scientifically measurable wildfire risks that vary across hazard zones in the three study locations. Individuals’ perceptions of wildfire can be substantially different from each other and from reality. This study proposes that the perception of risk is formed in a multistage process (individual and community level). Results show that homeowners’ worldview with respect to nature, length of residency, place-based influence, and attitudes about risk factors all are significant predictors for how residents of fire-prone areas perceive their risks. The variance in social and physical vulnerability associated with wildfire can explain, to a certain extent, the variation in individual perceptions of wildfire risk. The perception of risk is consistent with the level of exposure to fire hazards. The third study investigates spatial relationships among social and ecological factors on private property. The biophysical characteristics of individual properties were extracted to observe wildfire risk and incorporated with information about social context from mail surveys. Results demonstrate that mitigation behaviors in the three study communities illustrate a spatial clustering pattern. Moreover, orientations toward nature and physical attributes of property had an impact on decisions to undertake mitigation behaviors.

In designing inventories of wildland vegetation, two of the many basic methodological choices are: 1) whether data are collected, reduced, and stored in discrete classes or as continuous variables, and 2) whether data are gathered as general purpose variables to bear upon many questions, or as specific purpose variables optimized for only one type of prediction. The effects of these two choices on accuracy of vegetation inventories to predict plant community production were examined by comparing regression models built upon differing sets of independent variables "inventoried" from a common data base. Contrary to expectations, discrete variables of classified community types were better predictors of plant community production than the same vegetation data reduced as continuous variables by three ordination techniques. Substitution of specific purpose soil and vegetation variables thought to be especially relevant to production did not improve correlations from those of analogous general purpose variables. These results do not show the anticipated accuracy loss of general purpose inventory variables, but such findings cannot yet be generalized to other situations. Implications for the design of practical, extensive survey methods for wildland vegetation are briefly discussed.

In conducting this research, I wanted to explore whether political incentives have a significant effect on wildfire management in the United States. I attempt to answer this question by proving a theoretical justification for why wildfires may become more expensive to fight and severe to manage because of political institutions. I then attempt to provide some hard evidence to support this theory by using regression analysis. My analysis suggests that political factors do matter for wildfire suppression funding, although I was unable to find strong enough evidence to suggest that these political factors are actually driving more severe wildfires. This research contributes to the literature on public choice theory, a branch of political economy that looks at government from the individual decision makers’ level. Additionally, this research contributes to the literature on what affects wildfire suppression effectiveness and funding. This research may contribute to future analyses of the institutions that make U.S. wildland firefighters more or less capable of effectively managing wildfires to protect human lives, property, and forests.

The Puerto Rican coqui frog, Eleutherodactylus coqui, was introduced to Hawaii in the late 1980s via the commercial horticulture trade. Previous research has shown that coquis can change invertebrate communities, but these studies were conducted at small scales using controlled, manipulative experiments. The objective of this research was to determine whether coqui invasions change invertebrate communities at the landscape scale across the island of Hawaii. At each invasion front, we measured environmental variability on either side of the front and removed sites that were too variable across the front to ensure that the impacts we measured were the result of the invasion. After doing this, there remained 15 sites for which we compared invertebrate communities in 30 m x 30 m plots situated on either side of coqui invasion fronts. In each plot, we collected invertebrate samples from three invertebrate communities, the leaf litter, foliage, and flying invertebrate communities. Multivariate analyses show that coqui frogs change leaf litter communities, by reducing microbivore and herbivore abundances. Coqui also change flying community composition, but have no measurable effect on foliage communities. Across sites, we found that coquis reduced the number of leaf litter invertebrates by 27%, and specifically abundant Acari by 36%. We also found that coquis increased the abundance of flying Diptera by 19% across sites. We suggest that the leaf litter community is altered through direct coqui predation and that Diptera increase because of increased frog carcasses and excrement in invaded plots. Results support previous studies conducted in more controlled settings, but add to our understanding of the invasion by demonstrating that coqui effects on invertebrate communities are measurable at the landscape scale.

No description available.

Current operational wildland fire models are based on numerous correlations from experiments performed on dry (dead) fuel beds. However, experience has shown distinct differences in burning behaviors between dry and moist (live) fuels. To better understand these fundamental differences, an experiment was designed to use a flat-flame burner to simulate a moving fire front which heated and ignited a stationary, individual fuel sample. Samples included various U.S. species from the California chaparral, the intermountain west, and the southeastern regions. Temperature, mass, and video images were recorded throughout each experimental run from which numerous data values were obtained such as time to ignition, ignition temperature, flame height, time of flame duration, and mass release rates. Qualitative results showed various phenomena such as color change, bubbling, bursting, brand formation, and bending; these phenomena were species-dependent. Quantitative results showed differences in the ignition values (time, temperature, and mass) among species. It was observed that all moisture did not leave the interior of the sample at the time of ignition. Also, from the temperature history profiles, no plateau was observed at 100°C, but instead at 200-300°C. This indicates a need to treat evaporation differently than the classical combustion model. Samples were treated with solvents in attempt to extract the cuticle from the surface. These treated samples were compared to non-treated samples, though no significant combustion characteristics were observed. The time of color change for the treated samples varied significantly, indicating that the cuticle was indeed removed from the surface. Two-leaf configurations were developed and compared to determine combustion interactions between leaves. A second leaf was placed directly above the original leaf. Results showed that the time of flame duration of the upper leaf was significantly affected by the presence of the lower leaf. Causes for the prolonged flame were found to be the consumption of O2 by the lower leaf and the obstruction provided by the lower leaf, creating a wake effect which displaced hot gases from the flat-flame burner as well as entrained surrounding room temperature gas. A semi-physical model based on fluid dynamics and heat and mass transfer was developed that included the observed plateau at 200-300°C, rather than at 100°C; this was done for both the single- and two-leaf configurations. Another model using a statistical approach was produced which described the combustion of a bush that incorporated data obtained from the experimental results. Overall burning times and percentage of fuel consumption were obtained for various fuel loadings using this statistical model.

The recent surge in fire activity and the extent of displaced communities as a result of wildfire has increased awareness of wildfire issues nationwide (Syphard et al., 2017). Climate change, population growth, and continued development in the wildland urban interface (WUI) has contributed to a growing body of research into the underlying causes of this continued destruction (Kramer et al., 2019). There is no doubt that statewide policies, such as defensible space or building regulations, are associated with home survival (Keeley & Syphard, 2019). However, the relative effectiveness of wildfire mitigation depends on a myriad of factors specific to individual communities impacted by wildfire. This study focuses on factors that contributed to structure loss as a result of the 2017 Thomas Fire in Ventura, CA. Through spatial analysis utilizing GIS software, we were able to determine that defensible space played a minimal role in structural survivability during the Thomas Fire. Our research shows that fence type (noncombustible, combustible, or none) is a more significant factor at decreasing the odds of structure loss for homes experiencing wildfire under similar conditions. Effective wildfire mitigation relies on multiple factors, and government agencies must take a holistic approach rather than singular, “one size fits all” approaches to reduce the impact of future catastrophic wildfire.

Greater numbers of people are moving into wildland-urban interface (WUI) areas, increasing the number of people at risk to large wildfires. To mitigate the hazard, emphasis is often placed on fuel treatments used to reduce fuel loads and subsequent fire behavior. This approach overlooks the additional benefits provided by vegetation, including carbon storage and sequestration along with air pollutant removal. This study aimed to calculate and compare differences in representative values by examining a study site in the Klamath Mountains of Northern California. Fire behavior simulations were done under various weather scenarios to illustrate both the impact of weather on fire intensity as well as the limitations of various fuel treatments. Ecosystem services were modeled using the I-tree Eco software (formerly the Urban Forest Effects model). Results showed a reduction in surface and an increase in canopy base height from the treatments and subsequent reductions in fire intensity under moderate and high conditions with the largest difference occurring in the Thin + Fire treatment. Under extreme weather conditions, the effectiveness of all fuel treatments was reduced. Ecosystem services showed a reduction of carbon sequestration in the fuel treatments corresponding to the reduction of smaller diameter trees from the fuel treatments. The greatest difference occurred in the Thin + Fire treatment. These results and the methods used to acquire them show the impacts from fuel treatments can be characterized and compared. This information will allow land managers to make decisions that account for a variety of considerations, while also providing them with tools that can facilitate the cooperation and collaboration of multiple stakeholders.