Search

This study evaluated the hypothesis that urban-tree planting increases neighborhood gentrification in Portland, Oregon. We defined gentrification as an increase in the median sales price of single-family homes in a Census tract compared to other tracts in the city after accounting for differences in the housing stock such as house size and number of bathrooms. We used tree-planting data from the non-profit Friends of Trees, who have planted 57,985 yard and street trees in Portland (1990–2019). We estimated a mixed model of gentrification (30 years and 141 tracts) including random intercepts at the tract level and a first-order auto-regressive residual structure. Tract-level house prices and tree planting may be codetermined. Therefore, to address potential endogeneity of tree planting in statistical modeling, we lagged the number of trees planted by at least one year. We found that the number of trees planted in a tract was significantly associated with a higher tract-level median sales price, although it took at least six years for this relationship to emerge. Specifically, each tree was associated with a $131 (95% CI: $53-$210; p-value=0.001) increase in tract-level median sales price six years after planting. The magnitude of the association between the number of trees planted and median sales price generally increased as the time lag lengthened. After twelve years, each tree was associated with a $265 (95% CI: $151-$379; p-value

ObjectiveThis study proposes and evaluates the theory that people who are susceptible to injury in residential fires are not susceptible to death in residential fires and vice versa. It is proposed that the population vulnerable to death in residential fires can be proxied by ‘frailty’, which is measured as age–gender adjusted fatality rates due to natural causes.MethodsThis study uses an ecological approach and controls for exposure to estimate the vulnerability of different population groups to death and injury in residential fires. It allows fatalities and injuries to be estimated by different models.ResultsFrailty explains fire-related death in adults while not explaining injuries, which is consistent with the idea that deaths and injuries affect disjoint populations.ConclusionsDeaths and injuries in fire are drawn from different populations. People who are susceptible to dying in fires are unlikely to be injured in fires, and the people who are susceptible to injury are unlikely to die in fires.

Research to date has not examined how the impacts of arrests manifest across space and time in environmental crimes. We evaluate whether arrests reduce or merely spatiotemporally displace intentional illegal outdoor firesetting. Using municipality-level daily wildfire count data from Galicia, Spain, from 1999 to 2014, we develop daily spatiotemporal ignition count models of agricultural, non-agricultural and total intentional illegal wildfires as functions of spatiotemporally lagged arrests, the election cycle, seasonal and day indicators, meteorological factors and socioeconomic variables. We find evidence that arrests reduce future intentional illegal fires across space in subsequent time periods. This research was partly funded by Project ECO2017–89274-R MINECO/AEI/FEDER, UE SI

The wildland–urban interface (WUI) is defined in terms of housing density and proximity to wildlands, yet its relevance seems to be only in conjunction with wildland fire threats. The objective of this paper is to (1) identify the WUI areas threatened from wildfire during the 2000’s and (2) quantify the values that were threatened. We use 1 km fire detection data generated using MODIS satellite imagery over a 10-year period combined with population densities to identify threatened areas of the WUI. We then use data on structures, structure content, and population to identify the people and property threatened from identified fires within the WUI. We find that 6.3 % of the U.S. population (17.5 million) resided within these areas and that 2.1 % of the population lived in WUI areas where more than one fire has occurred. However, we find that only a third of the affected population was threatened during daytime hours, as most leave the threatened portion of the WUI during peak ignition hours. The threatened area comprised 4.1 % of the coterminous USA and 44.9 % of the WUI. Within these areas were 7.8 million residential, commercial, industrial, governmental, religious, and educational structures, with a building and building content value estimated at $1.9 trillion. Overall, 7.3 % of residential structures in the USA were found within the WUI with wildfire activity; however, for some states, this number was as high as 25.4 %.

Municipal fire departments responded to approximately 53,000 intentionally-set fires annually from 2003 to 2007, according to National Fire Protection Association figures. A disproportionate amount of these fires occur in spatio-temporal clusters, making them predictable and, perhaps, preventable. The objective of this research is to evaluate how the aggregation of data across space and target types (residential, non-residential, vehicle, outdoor and other) affects daily arson forecast accuracy for several target types of arson, and the ability to leverage information quantifying the autoregressive nature of intentional firesetting. To do this, we estimate, for the city of Detroit, Michigan, competing statistical models that differ in their ability to recognize potential temporal autoregressivity in the daily count of arson fires. Spatial units vary from Census tracts, police precincts, to citywide. We find that (1) the out-of-sample performance of prospective hotspot models for arson cannot usefully exploit the autoregressive properties of arson at fine spatial scales, even though autoregression is significant in-sample, hinting at a possible bias-variance tradeoff; (2) aggregation of arson across reported targets can yield a model that differs from by-target models; (3) spatial aggregation of data tends to increase forecast accuracy of arson due partly to the ability to account for temporally dynamic firesetting; and (4) arson forecast models that recognize temporal autoregression can be used to forecast daily arson fire activity at the Citywide scale in Detroit. These results suggest a tradeoff between the collection of high resolution spatial data and the use of more sophisticated modeling techniques that explicitly account for temporal correlation.

Each year, wildland fires threaten structures and occupants of the wildland-urban interface (WUI). Currently, wildfire ignition estimates largely exclude ignitions originating within municipal jurisdictions, which contain the majority of the US population. The objective of this article is to provide national estimates and trends of the WUI fire problem; in particular, we analyze the US Fire Administration's National Fire Incident Reporting System fire incident data from 2002 to 2006 reported by local municipal fire departments across the United States. We estimate that, on average, the burning of wildland fuels is associated with 116,971 fires annually. Each year, these fires are responsible for, on average, 15 civilian (nonfire service) fatalities, 88 civilian injuries, and $160 million in direct property losses. These damages include losses to 599 residential structures, 649 nonresidential structures, and 829 vehicles (per year). Based on the value of a statistical life ($8.75 million) and statistical injury ($189 thousand), we find that the economic value of fatalities and injuries averaged $148 million annually. Thus, the losses associated with fires occurring on municipal lands total on average $308 million annually over this time period.

This paper investigated whether greater tree-canopy cover is associated with reduced risk of poor birth outcomes in Portland, Oregon. Residential addresses were geocoded and linked to classified-aerial imagery to calculate tree-canopy cover in 50, 100, and 200 m buffers around each home in our sample (n=5696). Detailed data on maternal characteristics and additional neighborhood variables were obtained from birth certificates and tax records. We found that a 10% increase in tree-canopy cover within 50 m of a house reduced the number of small for gestational age births by 1.42 per 1000 births (95% CI-0.11-2.72). Results suggest that the natural environment may affect pregnancy outcomes and should be evaluated in future research.

We estimated the effect of shade trees on the summertime electricity use of 460 single-family homes in Sacramento, California. Results show that trees on the west and south sides of a house reduce summertime electricity use, whereas trees on the north side of a house increase summertime electricity use. The current level of tree cover on the west and south sides of houses in our sample reduced summertime electricity use by 185 kWh (5.2%), whereas north-side trees increased electricity use by 55 kWh (1.5%). Results also show that a London plane tree, planted on the west side of a house, can reduce carbon emissions from summertime electricity use by an average of 31% over 100 years.

Our article considers the economic contributions of forest ecosystem services, using a case study from Flores, Indonesia, in which forest protection in upstream watersheds stabilize soil and hydrological flows in downstream farms. We focus on the demand for a weak complement to the ecosystem services—farm labor—and account for spatial dependence due to economic interactions, ecosystem processes, and data integration. The estimated models have theoretically expected properties across eight different specifications. We find strong evidence that forest ecosystem services provide economically substantive benefits to local people and that these services would be substantially undervalued if spatial dependence is ignored. Forests provide an array of ecosystem services by sequestering carbon, maintaining habitat and biodiversity, stabilizing hydrological flows, mitigating soil erosion, and improving microclimates. Deforestation and forest degradation can irreversibly and substantively impair these ecosystem functions. This raises the question of why society and governments would allow rapid or excessive deforestation. One reason is the failure to consider the full range of goods and services provided by the forests, particularly any latent and complex ecosystem services (Dasgupta). The economic contributions of forest ecosystem services are not well understood and rarely quantified. This article illustrates a method for estimating the value of watershed services from protected tropical forests in Flores, Indonesia. Specifically, we respond to three challenges posed in the literature. First, valuation studies have typically overlooked livelihood values of natural resources in developing countries, focusing largely on amenity values in developed countries (Deacon et al., Dasgupta). Second, a detailed consideration of the spatial aspects of ecosystems and ecological processes, such

Six Poisson autoregressive models of order p[PAR(p)] of daily wildland arson ignition counts are estimated for five locations in Florida (1994–2001). In addition, a fixed effects time-series Poisson model of annual arson counts is estimated for all Florida counties (1995–2001). PAR(p) model estimates reveal highly significant arson ignition autocorrelation, lasting up to eleven days, in addition to seasonality and links to law enforcement, wildland management, historical fire, and weather. The annual fixed effects model replicates many findings of the daily models but also detects the influence of wages and poverty on arson, in ways expected from theory. All findings support an economic model of crime. Copyright 2005, Oxford University Press.

The absence of a comprehensive and accurate database on fire occurrence means that some proportion of wildfires is not reported. This paper examines wildfires reported in the National Fire Incident Reporting System (NFIRS) database and compares it with the satellite-based Moderate Resolution Imaging Spectroradiometer (MODIS) fire detection data and the CAL FIRE Fire and Resource Assessment Program (FRAP) geodatabase to understand underreporting of wildland fires in the NFIRS system. The paper discusses a series of large wildland fires and uses a Generalised Linear Model to identify the conditions where large wildfires go unreported. This paper shows that the NFIRS database, which is the primary fire database in the US, significantly underreports wildland fires. Evidence from California suggests that 32% of fires within local jurisdictions detected via MODIS satellite were identified in the NFIRS database; thus, there could be a significant number of fires going unreported nationwide. Examining eight large fires in California, only 16% of the structures damaged or destroyed, as reported in CAL FIRE, were reported in NFIRS. Areas with underreporting tend to have a higher level of poverty, higher population density, higher level of people without vehicles, lower income and a higher level of single parents with children under 18, as well as a higher level of people under the age of 18.

Natural large (catastrophic) disturbances are important because of their potential long-lasting impact on their surroundings, but underlying differences between frequent small and less common large disturbances are not well understood (Turner and Dale 1997, Romme et al. 1998, Turner and Dale 1998, chapter 4 of this book). Smaller disturbances may be better understood given their relative abundance, which lends itself more readily for study, but it is, perhaps, more useful to understand the forces driving damaging, catastrophic events. Wildland fires represent a perfect example. Nationwide, over 130,000 wildfires burn more than 4 million acres annually (1960-2002), these fires costing Federal agencies in excess of $768 million a year (1994-2002) in suppression alone (National Interagency Fire Center, http://www.nifc.gov/stats/index.html). Average wildfire size was 31 acres, with a suppression cost of $4800 per fire. The average wildfire does not appear a catastrophic threat, however this ignores the spatial distribution of these fires in relation to values at risk (an averaged size fire in a heavily populated area poses a different risk than a similarly size fire far removed from people and items of value). Catastrophic fire events, while relatively infrequent, do occur with some regularity—the 2000 Cerro Grande fire in New Mexico devastated 47,650 acres, two fire complexes in California in fall of 1999 each burned for three months and consumed a total of 227,647 acres, and during the 1998 Florida summer wildfire season, two fire complexes accounted for 205,786 acres or 9 percent of all wildfire acres, nationwide, in that year (National Interagency Fire Center, http://www.nifc.gov/stats/index.html). Do the largest fires account for a disproportional amount of the area burned and damage? Is it possible that the largest 1 percent of fires account for 99 percent of the area, as Strauss et al. (1989) explored? For the state of Florida, the largest 1 percent accounted for 67 percent of total area burned with an average fire size of 2,641 acres versus 13 acres for the smallest 99 percent (1981-2001). Understanding the differences between small and large wildfires, including the exogenous factors influencing each, may provide decision-makers with better tools to mitigate future large-scale fire events. It is not necessarily true that large

Research shows that some categories of human-ignited wildfires may be forecastable, owing to their temporal clustering, with the possibility that resources could be predeployed to help reduce the incidence of such wildfires. We estimated several kinds of incendiary and other human-ignited wildfire forecast models at the weekly time step for tribal land units in the United States, evaluating their forecast skill out of sample. Analyses show that an autoregressive conditional Poisson model of both incendiary and non-incendiary human-ignited wildfires is more accurate out of sample compared with alternatives, and the simplest of the autoregressive conditional Poisson models performed the best. Additionally, an ensemble of these and simpler, less analytically intensive approaches performed even better. Wildfire hotspot forecast models using all model types were evaluated in a simulation mode to assess the net benefits of forecasts in the context of law-enforcement resource reallocations. Our analyses show that such hotspot tools could yield large positive net benefits for the tribes in terms of suppression expenditures averted for incendiary wildfires but that the hotspot tools were less likely to be beneficial for addressing outbreaks of non-incendiary human-ignited wildfires.

Humans cause more than 55% of wildfires on lands managed by the USDA Forest Service and US Department of the Interior, contributing to both suppression expenditures and damages. One means to reduce the expenditures and damages associated with these wildfires is through fire prevention activities, which can include burn permits, public service programs or announcements, outreach efforts to schools, youth groups and equipment operators, and law enforcement. Using data from 17 US Bureau of Indian Affairs tribal units, we modelled the effect of prevention programs and law enforcement on the number of human-caused ignitions. We also included weather and lagged burned area in our estimation of fixed-effects count models. The results show that prevention activities led to significant reductions in wildfires caused by escaped campfires, juveniles, fire-use (e.g. escaped debris burns) and equipment. Increased law enforcement resulted in fewer incendiary- and equipment-caused wildfires. Using average suppression expenditures by wildfire and our estimate of avoided wildfires per additional year of prevention, we estimate partial benefit–cost ratios of greater than 4.5 for all Bureau of Indian Affairs regions for the continuation of the prevention program.

We analyse the spatio-temporal structure of wildfire ignitions in the St Johns River Water Management District in north-eastern Florida. We show, using tools to analyse point patterns (e.g. the L-function), that wildfire events occur in clusters. Clustering of these events correlates with irregular distribution of fire ignitions, including lightning and human sources, and fuels on the landscape. In addition, we define a relative clustering index that summarizes the amount of clustering over various spatial scales. We carry our analysis in three steps: purely temporal, purely spatial, and spatio-temporal. Our results show that arson and lightning are the leading causes of wildfires in this region and that ignitions by railroad, lightning, and arson are spatially more clustered than ignitions by other accidental causes.