Search

The US wood pellet exports—mostly to Europe, Japan, and South Korea—comprised 85 percent of its wood pellet production in the past 20+ years. In this article, we estimate the regional excess supply of wood pellets in the United States using annual data from 1996 to 2021.We use Seemingly Unrelated Regression (SUR) analysis in a profit maximization framework for the South, North, and West regions of the United States. We also include some clean energy policies of the European Union (EU) to examine their potential impact on US wood pellet export volume. Our results show that after application of the low-carbon energy target in the EU, the export volume of the United States has increased, especially in the South region. Elasticity of excess wood pellet supply from the United States was approximately 0.15 in all three regions, and interest rates and energy costs have negative impacts on regional excess supply. Therefore, the recent increase in interest rates and energy costs could slow down the growth of excess supply of wood pellets in the United States.

There has been a significant increase in the use of wood pellets for energy in the past decade due in large part to their climate mitigation potential. Because of this, the demand for wood pellets is largely driven by policy, as well as socioeconomic development, making projections of future wood energy markets highly uncertain. The aim of this study is to provide projections of future wood energy market trends under five distinct socioeconomic scenarios based on the assumed future evolution of gross domestic product, population, technological change, trade openness, and bioenergy preferences using the FOrest Resource Outlook Model. In four out of the five scenarios considered, it is projected that the use of roundwood and mill chips, particles, and residuals will rise in order to produce a growing output of wood pellets in the United States and globally. In terms of international markets, the global dominance of Europe’s demand for wood, to help that continent achieve its own climate goals, further explains the sustained and growing supply position of the U.S. South regions to meet that demand. Taken together, the projections suggest emerging bioenergy markets will drive increased competition for inputs with other manufacturers, particularly in the U.S. South regions.

We develop fully flexible, semi-parametric models of price linkages in spatially distinct regional US markets for plywood and lumber products. The models are developed within the framework of generalized additive models (GAM) which are estimated using penalized maximum likelihood estimation methods and generalized cross-validation techniques. Such nonlinear models are a natural extension to an extensive literature that has developed increasingly flexible models of the Law of One Price and spatial market integration. The GAM estimates and marginal responses to price changes exhibit substantial nonlinearities. Dynamic impulse responses demonstrate significant asymmetries in price responses to exogenous shocks. The results are largely consistent with efficiently linked regional markets for lumber and plywood products in that price transmission elasticities are generally close to unity and localized market shocks evoke equilibrating adjustments in regional markets.

This study aims to examine how organizations in the bio-based industry perceive risks and perform risk analysis within the capital investment decision-making process. More specifically, this study aims to assess sources of uncertainty commonly considered, identify tools and methods used for risk assessment, and understand how risk analysis is considered in capital budgeting. Eighty-six respondents were electronically surveyed on practices for capital investment risk analysis, including C-suite and upper management from different organization sizes and segments in the bio-based industry. It was found that some forms of risk analysis are utilized either in project assessment and/or for decision making by most respondents; however, qualitative and deterministic assessment practices dominate over probabilistic methods. In addition, risk assessment is most commonly performed in the later stages of a project, with less than 50% of adoption at the earlier stages. Overall, the main sources of uncertainties considered when performing risk assessment are financial, market and sales, and technology, with competition being considered mostly by upper management levels. Additionally, consistent with previous studies in other industry sectors, Internal Rate of Return, Return on Investment, and Net Present Value are the preferred financial indicators used to evaluate capital investments.

Wildfires have greater impacts on socially vulnerable communities. Identifying these vulnerable communities and enhancing understanding of what influences their susceptibility to wildfires can guide the design of spatially targeted strategies in preparedness, mitigation plans, and adaptation strategies. This paper investigates the heterogeneous spatial coincidence of social vulnerability and wildfire risk in Galicia (Spain) at the municipality level. Results show that socioeconomic status, rates of dependence on social programs, and household unit characteristics are factors that contribute the most to social vulnerability. In general, municipalities with the highest proportion of their area in the Wildland-Urban Interface (WUI) have the lowest social vulnerability. Within Galicia, locations with high social vulnerability and high wildfire risk are spatially concentrated in the south and tend to be low-population density communities, often in remote locations and with relatively high percentages of elderly people. Our findings provide an empirical foundation for wildfire management planning that accounts for the spatial distribution of vulnerable communities This research was supported by project ECO2017-89274-R (MINECO/AEI/FEDER, UE) SI

At a fundamental level, smoke from wildland fire is of scientific concern because of its potential adverse effects on human health and social well-being. Although many impacts (e.g., evacuations, property loss) occur primarily in proximity to the actual fire, smoke can end up having a significant social impact far from the source. This dynamic, combined with lengthening fire seasons, suggests that understanding how wildland fire smoke affects diverse social values will be increasingly critical. This chapter reviews the existing scientific knowledge related to wildland fire smoke with respect to four topic areas: human health, economics, social acceptability, and risk communication. The broadest existing knowledge base, regarding the health effects attributed to wildland fire smoke exposure, stems from decades of research on the health effects of exposures to ambient fine particulate matter (PM2.5). Despite the potential consequences, scientific knowledge about chronic health effects, economic impacts, and effectiveness of protective actions in response to wildfire smoke risk communication is fairly limited. The chapter concludes with identification of (1) key areas where the need for more empirical information is most critical, and (2) challenges that inhibit an improved scientific understanding.

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

Climate warming has been implicated as a major driver of recent catastrophic wildfires worldwide but analyses of regional differences in US wildfires show that socioeconomic factors also play a large role. We previously leveraged statistical projections of annual areas burned (AAB) over the fast-growing southeastern US that include both climate and socioeconomic changes from 2011 to 2060 and developed wildfire emissions estimates over the region at 12 km × 12 km resolution to enable air quality (AQ) impact assessments for 2010 and selected future years. These estimates employed two AAB datasets, one using statistical downscaling (“statistical d-s”) and another using dynamical downscaling (“dynamical d-s”) of climate inputs from the same climate realization. This paper evaluates these wildfire emissions estimates against the U.S. National Emissions Inventory (NEI) as a benchmark in contemporary (2010) simulations with the Community Multiscale Air Quality (CMAQ) model and against network observations for ozone and particulate matter below 2.5 µm in diameter (PM2.5). We hypothesize that our emissions estimates will yield model results that meet acceptable performance criteria and are comparable to those using the NEI. The three simulations, which differ only in wildfire emissions, compare closely, with differences in ozone and PM2.5 below 1 % and 8 %, respectively, but have much larger maximum mean fractional biases (MFBs) against observations (25 % and 51 %, respectively). The largest biases for ozone are in the fire-free winter, indicating that modeling uncertainties other than wildfire emissions are mainly responsible. Statistical d-s, with the largest AAB domain-wide, is 7 % more positively biased and 4 % less negatively biased in PM2.5 on average than the other two cases, while dynamical d-s and the NEI differ only by 2 %–3 % partly because of their equally large summertime PM2.5 underpredictions. Primary species (elemental carbon and ammonium from ammonia) have good-to-acceptable results, especially for the downscaling cases, providing confidence in our emissions estimation methodology. Compensating biases in sulfate (positive) and in organic carbon and dust (negative) lead to acceptable PM2.5 performance to varying degrees (MFB between −14 % and 51 %) in all simulations. As these species are driven by secondary chemistry or non-wildfire sources, their production pathways can be fruitful avenues for CMAQ improvements. Overall, the downscaling methods match and sometimes exceed the NEI in simulating current wildfire AQ impacts, while enabling such assessments much farther into the future.

From Source to Sink: Past Changes and Model Projections of Carbon Sequestration in the Global Forest Sector

Developing Detailed Shared Socioeconomic Pathway (SSP) Narratives for the Global Forest Sector

Planted forests are a rising share of total forests globally and an increasingly important source of timber product output, affecting national and global markets. We estimated econometric models of planted forest area by OECD and non-OECD country groups that control for economic, institutional and environmental policies likely to influence future changes in planted forest area. The models are then used to project planted forest area over next 55 years for 180 countries under five alternative scenarios of global socio-economic changes, represented in shared socioeconomic pathways (SSPs), adjunct products emerging from the Fifth Assessment of the Intergovernmental Panel on Climate Change (IPCC). By embedding key features of the SSP projections into a global forest sector model, we evaluate how planted forests lead to different global forest product market outcomes for each SSP, compared to corresponding outcomes where planted forests are not considered separately. Projected global planted forest area in 2070 ranges from 379 million ha (Mha) for SSP3 (a relatively poor and unequal world) to 475 Mha under SSP5 (a relatively wealthier and more equal world), representing respective increases of 46% and 66% compared to 2015. SSPs with the highest planted forest area increases have the lowest product prices (down by 12% by 2070, compared to SSP5 without planted forests) and higher global forest products production and consumption quantities (by as much as 3.3% by 2070, compared to SSP5 without planted forests). However, production does not increase in all countries by similar amounts, due to changes in relative advantages in production brought about by reduced product prices.

There is rising global interest in growing more trees in order to meet growing population, climate change, and wood energy needs. Using recently published data on planted forests by country, we estimated relationships between per capita income and planted forest area that are useful for understanding prospective planted forest area futures through 2100 under various United Nations Intergovernmental Panel on Climate Change-inspired Shared Socio-economic Pathways (SSPs). Under all SSPs, projections indicate increasing global planted forest area trends for the next three to four decades and declining trends thereafter, commensurate with the quadratic functions employed. Our projections indicate somewhat less total future planted forest area than prior linear forecasts. Compared to 293 million ha (Mha) of planted forests globally in 2015, SSP5 (a vision of a wealthier world) projects the largest increase (to 334 Mha, a 14% gain) by 2055, followed by SSP2 (a continuation of historical socio-economic trends, to 327 Mha, or an 11% gain), and SSP3 (a vision of a poorer world, to 319 Mha, a 9% gain). The projected trends for major world regions differ from global trends, consistent with differing socio-economic development trajectories in those regions. Our projections based on empirical FAO data for the past 25 years, as well as those by other researchers, suggest that achieving the much more ambitious global planted forest targets proposed recently will require exceptional forest land and investment supply shifts.

The health benefits of exposure to trees and plants is a rapidly expanding field of study. Research has shown that exposure is associated with improvements in a wide range of health outcomes including cardiovascular disease, birth outcomes, respiratory disease, cancer, mental health and all-cause mortality

Renewable natural resources that have biological constraints on reproduction and are open-access may be subjected to intense harvest activity that limits regeneration, potentially leading to a backward-bending long-run supply curve. Empirical evidence of such supply abnormalities has been found for some open-access fish species but not yet for non-timber forest products (NTFPs). We describe the theory of the backward-bending long-run supply and how such a supply relationship could produce multiple market equilibria, affecting regulatory outcomes. An empirical example is provided to test the theory in the case of wild American ginseng (Panax quinquefolius), which has been subjected to habitat loss and harvest pressure since the 18th Century and now has its exports regulated. We find evidence that quantities supplied are negatively related to price in the long run, indicating that harvest pressure is restricting wild ginseng harvestable stocks. Also, we find that a federal regulation banning exports of roots from plants under five years old, in effect since 1999, coincided with a reduction of supply. This result could be due to the slow natural rate of population recovery.

The southern pine beetle (SPB) is among the leading biological agents killing southern pine species in the eastern United States. In light of recognized spatiotemporal autocorrelation in SPB outbreaks, we devise a spatiotemporal block bootstrapping method that can be applied to analyze spatiotemporally dependent infestations. We also identify the relevant risk determinants and evaluate their impacts on the frequency of SPB outbreaks. For example, we find forest type, climate, and natural disasters like storm and forest management are all significantly associated with SPB risks. Using the results of a statistical model, we design a county‐level group index insurance plan that generates estimates of actuarially fair premium rates for timber stands containing southern pine species. Given that no government‐provided compensation scheme for SPB epidemics currently exists, application of this new insurance product could reduce forest owners losses. Our study offers an approach to analyzing and protecting against risks of other destructive pests affecting the timber sector.

In investing in a new venture, companies aim to increase their competitiveness and generate value in scenarios where volatile markets, geopolitical instabilities, and disruptive technologies create uncertainty and risk. The biobased industry poses additional challenges as it competes in a mature, highly efficient market, dominated by petroleum-based companies, and faces significant feedstock availability and variability constraints, limited technological data, and uncertain market conditions for newly developed products. Thus, decision-making strategies and processes for these investment projects must consider solid risk estimation and mitigation measures. Focusing on the biobased industrial sector, this paper critically reviews state-of-the-art probabilistic and deterministic methodologies for assessing financial risk; discusses how a complete risk analysis should be performed; and addresses risk management, listing major risks and possible mitigation strategies. © 2017 Society of Chemical Industry and John Wiley & Sons, Ltd

This paper proposes an alternative and potentially novel approach to analyzing the law of one price in a nonlinear fashion. Copula-based models that consider the joint distribution of prices separated by space are developed and applied to weekly prices for lumber products. The copulas capture nonlinearities that arise in the extremes of the joint distributions of price differentials and suggest faster equilibrating adjustments when deviations from parity are extreme.

Forest resources are critical to environmental, economic, and social development, and there is substantial interest in understanding how global forest area will evolve in the future. Using an Environmental Kuznets Curve (EKC) model of total forest area that we updated using more recent data sets, we projected forest area through 2100 in 168 countries using variables including income, rural population density, and the size of the labor force under different world visions drawn from alternative Intergovernmental Panel on Climate Change socioeconomic pathways (SSPs). Results provided support for the existence of an EKC for total forest area, with rural population density negatively affecting forest area and labor force size positively affecting forest area. The projections showed modest and continuous increases in global forest area in all the SSPs, but varying trends for major world regions, which is consistent with the projected trends from the explanatory variables in each country. Aggregate global forest area is projected to increase by 7% as of 2100 relative to 2015 levels in SSP3, which predicts a future with the lowest rate of economic growth, and by 36% in SSP5, which is a future with the highest rate of economic growth and greater economic equality across countries. The results show how projections driven only by income produce biased results compared to the projections made with an EKC that includes rural population density and labor force variables.

Government agencies in the United States eradicated 10.3 million cannabis plants in 2010. Most (94%) of these plants were outdoor-grown, and 46% of those were discovered on federal lands, primarily on national forests in California, Oregon, and Washington. We developed models that reveal how drug markets, policies, and environmental conditions affect grow siting decisions. The models were built on a rational choice theoretical structure, and utilized data describing 2322 cannabis grow locations (2004–2012) and 9324 absence locations in the states' national forests. Predictor variables included cannabis market prices, law enforcement density, and socioeconomic, demographic, and environmental variables. We also used the models to construct regional maps of grow site likelihood. Significant predictors included marijuana street price and variables associated with grow site productivity (e.g., elevation and proximity to water), production costs, and risk of discovery. Overall, the pattern of grow site establishment on national forests is consistent with rational choice theory. In particular, growers consider cannabis prices and law enforcement when selecting sites. Ongoing adjustments in state cannabis laws could affect cultivation decisions on national forests. Any changes in cannabis policies can be reflected in our models to allow agencies to redirect interdiction resources and potentially increase discovery success.

The United States’ shares of global industrial roundwood production and derivative products have declined precipitously since the 1990s. We evaluate the extent of these declines compared with those of major producing countries from 1961 to 2013. We find that the US global share of industrial roundwood peaked at 28% in 1999 but by 2013 was at 17%, with the decline attributable to a combination of cyclical factors and long-run trends. Wood products output declines are linked to low construction levels, whereas paper’s decline is connected to the offshoring of US manufacturing and growth in electronic media. Prospects are for increased use of wood in construction as the housing market returns to long-run averages in the coming years. However, the paper sector is unlikely to recover to its 1990s levels of output, implying that it is unlikely that the United States will return to its historical highs in global market share of industrial roundwood production.

Future changes in society and climate are expected to affect wildfire activity in the south-eastern United States. The objective of this research was to understand how changes in both climate and society may affect wildfire in the coming decades. We estimated a three-stage statistical model of wildfire area burned by ecoregion province for lightning and human causes (1992–2010) based on precipitation, temperature, potential evapotranspiration, forest land use, human population and personal income. Estimated parameters from the statistical models were used to project wildfire area burned from 2011 to 2060 under nine climate realisations, using a combination of three Intergovernmental Panel on Climate Change-based emissions scenarios (A1B, A2, B2) and three general circulation models. Monte Carlo simulation quantifies ranges in projected area burned by county by year, and in total for higher-level spatial aggregations. Projections indicated, overall in the Southeast, that median annual area burned by lightning-ignited wildfire increases by 34%, human-ignited wildfire decreases by 6%, and total wildfire increases by 4% by 2056–60 compared with 2016–20. Total wildfire changes vary widely by state (–47 to +30%) and ecoregion province (–73 to +79%). Our analyses could be used to generate projections of wildfire-generated air pollutant exposures, relevant to meeting the National Ambient Air Quality Standards.

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

Downscaled climate projections need to be linked to downscaled projections of population and economic growth to fully develop implications for land, natural resources, and ecosystems for future scenarios. We develop an empirical spatiotemporal approach for jointly projecting population and income at the county scale in the United States that is consistent with neoclassical economic growth theory and overlapping labor markets and that accounts for labor migration and spatial spillovers. Downscaled projections generated for the five Shared Socioeconomic Pathways used to support global scenario analysis generally show growth focused around relatively few centers especially in the southeast and western regions, with some areas in the Midwest and northeast experiencing population declines. Results are consistent with economic growth theory and with historical trends in population change and convergence of per capita personal income across US counties.

The recent rise of institutional timberland ownership has led to a significant change in the structure and conduct of the timber industry in the United States. In this study,we apply a two-period harvest model to assess the timber harvesting behavior of various landowners at the stand level by utilizing USDA Forest Service Forest Inventory and Analysis data for nine southern states. Forest industry and institutional timberland owners were found to be more likely to conduct partial and final harvests than nonindustrial private forest landowners. Aggregately, Timberland Investment Management Organizations were found to be most, and timberlandReal Estate InvestmentTrusts to be least, price-responsive among ownership groups

Development of efficient forest wildfire policies requires an understanding of the underlying reasons behind forest fire occurrences. Globally, there is a close relationship between forest wildfires and human activities; most wildfires are human events due to negligence (e.g., agricultural burning escapes) and deliberate actions (e.g., vandalism, pyromania, revenge, land use change attempts). We model the risk of wildfire as a function of the spatial pattern of urban development and the abandonment/intensity of agricultural and forestry activities, while controlling for biophysical and climatic factors. We use a count data approach to model deliberately set fires in Galicia, N.W. Spain, where wildfire is a significant threat to forest ecosystems, with nearly 100,000 wildfires recorded during a thirteen-year period (1999e2011). The spatial units of analysis are more than 3600 parishes. Data for the human influences are derived from fine-resolution maps of wildlandeurban interface (WUI), housing spatial arrangements, road density, forest ownership, and vegetation type. We found wildfire risk to be higher where there are human populations and development/urbanisation pressure, as well as in unattended forest areas due to both rural exodus and a fragmented forest ownership structure that complicates the profitability of forestry practices. To better help direct management efforts, parameter estimates from our model were used to predict wildfire counts under alternative scenarios that account for variation across space on future landuse conditions. Policies that incentivize cooperative forest management and that constrain urban development in wildlands at hotspot fire locations are shown to reduce wildfire risk. Our results highlight the need for spatially targeted fire management strategies.

The Lacey Act of 1900 was amended on May 22, 2008, to prohibit the import of illegally sourced plant materials and products manufactured from them into the United States and its territories, and to similarly ban their interstate transport. Trade theory suggests that the effect of the new law would be to reduce the flow of illegally sourced fiber into the United States, increasing prices. Monthly U.S. import data on tropical lumber (January 1989–June 2013) and hardwood plywood (January 1996–June 2013) quantity and unit value were used to estimate alternative statistical models that quantify the impact of the 2008 Lacey Act Amendment on import prices and import quantities of products from potential source countries. Results show that the Amendment's quantity effects are generally negative and double in magnitude in percentage terms than the price effects, consistent with expectations of the effects of a backwards shift in foreign supply against an elastic import demand. Models indicate that there have been double-digit percentage increases in prices and decreases in quantities of tropical lumber imports from Bolivia, Brazil, Indonesia, Malaysia, and Peru. Similarly large changes in hardwood plywood import prices and quantities from Brazil, Indonesia, and Malaysia have occurred, while smaller, and in some cases statistically insignificant, changes have been observed for hardwood plywood imports from China, Ecuador, and Taiwan.

In this study we demonstrate how the notion of diversification can be used in broad-scale resource allocation for surveillance of invasive species. We consider the problem of short-term surveillance for an invasive species in a geographical environment. We find the optimal allocation of surveillance resources among multiple geographical subdivisions via application of a classical portfolio framework, which allocates investments among multiple financial asset types with uncertain returns in a portfolio that maximizes the performance and, by meeting the desired diversification targets, protects against errors in estimating the portfolio's performance. We illustrate the approach with a case study that applies a spatial transmission model to assess the risk of spread of the emerald ash borer (EAB), a significant pest in North America, with infested firewood that may be carried by visitors to campground facilities in central Canada. Adding the diversification objective yields an expected survey performance that is comparable with undiversified optimal allocation, but more importantly, makes the geographical distribution of survey priorities less subject to possible errors in the spread rate estimates. Overall, diversification of pest surveillance can be viewed as a viable short-term strategy for hedging against uncertainty in expert- and model-based assessments of pest invasion risk.

We assess exchange rate pass-through (ERPT) for U.S. and Canadian prices for oriented strand board (OSB), a structural wood panel product used extensively in U.S. residential construction. Because of its prominence in construction and international trade, OSB markets are likely sensitive to general economic conditions. In keeping with recent research, we examine regime-specific ERPT effects; we use a smooth transition vector error correction model. We also consider ERPT asymmetries associated with a measure of general macroeconomic activity. Our results indicate that during expansionary periods ERPT is modest, but during downturns, ERPT effects are larger. USDA-USFS [SRS 09-CA-11330143-023] Published version This research has been supported by USDA-USFS Cooperative Agreement No. SRS 09-CA-11330143-023. We appreciate the helpful comments provided by an anonymous referee and by seminar participants at the International Conference of Agricultural Economists in Vancouver, Canada, August 2018. Authors names appear in alphabetical order - senior authorship is shared. Public domain – authored by a U.S. government employee

Ten US states and the District of Columbia have legalized cannabis as of November 2018, and have adopted other policies regarding production, consumption, and the penalties associated with it. These policy changes may have affected illegal growing operations on national forests of the United States. Using data on the number of cannabis grow sites reported on 111 national forests between 2004 and 2016 together with information about state cannabis laws and when they were implemented, we find that recreational cannabis legalization is associated with decreased reports of illegal grow operations on national forests. Laws mandating minimum sentences for illegal cannabis possession or sales are associated with fewer reported grows, as is strict regulation of cannabinoid products. Taxes on sales have positive impacts on illegal growing, while law enforcement presence has a negative effect. Counterfactual simulations for 2016 quantify the magnitudes of these policy effects.

In the U.S. forest products industry, wildfire is one of the leading causes of damage and economic losses. While individual wildfire behavior is well studied, new literature is emerging on broad‐scale (e.g., county‐level) wildfire risks. Our paper studies wildfire risks using crucial informational variables across both spatial units and time periods. Several statistical models are used to quantify the risks. We develop several maximum likelihood estimation methods to account for spatio‐temporal auto‐correlation in conditional risks. A group index insurance scheme is proposed, and its associated actuarially fair premium rates are estimated and presented. Implications for wildfire management policies are also discussed.

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.