Risk and damage of southern pine beetle outbreaks under global climate change
Introduction
The southern pine beetle (SPB) (Dendroctonus frontalis Zimmermann) is the most destructive insect to southern pine forests. From 1970 to 1996, trees killed by SPB were valued at nearly US$ 1.5 billion in the US south (Price et al., 1998). Among the factors that influence SPB infestations, climate is probably the most recognized and important one (Gagne et al., 1980, Ungerer et al., 1999). Due to increasing CO2 concentration in the atmosphere, global climate is predicted to change at an unprecedented rate over the next century (IPCC, 2003). Changes in temperature and precipitation would influence SPB populations directly, through the physiological process of the insect, and indirectly, through its host trees and natural predators. Consequently, climate change could have profound effects on the distribution and abundance of SPB (Ayres and Lombardero, 2000), and thus the range (area) and intensity (risk) of SPB infestations. SPB outbreak areas are projected to expand and generally shift northward as temperature increases. The expansion of the projected infestation areas would be primarily due to the shifts in the distribution of southern pines resulting from climate change (Williams and Liebhold, 2002). However, the impact of climate change on the intensity of SPB infestations, particularly in terms of economic damage, has been relatively unexplored.
While the links between climate and SPB populations have long been recognized (Craighead, 1925, Beal, 1927), the efforts to date to quantify climate-SPB outbreak relationships have met with only limited success. Many agree that extremes in temperature and precipitation affect SPB populations and host trees, influencing SPB infestations. Cold winters and hot summers generally reduce SPB populations (Beal, 1929, Thatcher, 1960) while moisture deficits and surpluses could both contribute to SPB outbreaks (Kalkstein, 1974, Warning and Cobb, 1989). Attempts also have been made to develop the correlations between climatic/weather conditions and SPB outbreaks. Several studies were conducted to relate a variety of climatic variables to SPB outbreaks using multiple regressions (King, 1972, Kroll and Reeves, 1978, Campbell and Smith, 1980). However, the potential multicollinearity among the independent variables might introduce biases to the results (Kalkstein, 1981). As a result, findings from these studies are often confusing and contradicting with each other (Turchin et al., 1991). To avoid the multicollinearity problem, principal component analysis (PCA) was employed to establish regression relationships between climatic/weather conditions and SPB outbreaks (Kalkstein, 1981, Michaels, 1984, Michaels et al., 1986). In this approach, a series of independent linear combinations (principal components) of climatic data is generated, and then used to fit the regression model of SPB outbreaks. While the introduction of PCA alleviates multicollinearity, exclusion of other nonclimatic/weather factors creates new limitations and lessened the predictive capability of these models (Martinat, 1987, Turchin et al., 1991, McNulty et al., 1997). For this reason, their results are mostly qualitative rather than quantitative (Michaels, 1984). In addition, it becomes very cumbersome to separate or identify the effects of individual climatic factors like temperature and precipitation in a specific time period, because each principal component is related to many climatic factors in various time periods included in the model.
Moreover, these earlier models developed using time series data cannot account for natural and human adaptation to climate change. Incorporating natural and human adaptation is important in assessing climate change impacts because the impacts are complex and interrelated (Mendelsohn et al., 1994, Chen and McCarl, 2001, Lindner et al., 2002). Because global climate change would take place gradually over a long time period, the insect, its natural enemies, and host trees may adapt to the change. Meanwhile, climate change is likely to alter stand composition as it affects species distribution and abundance and forest landscapes due to natural and human adaptation (Davis and Zabinski, 1992, Bachelet and Neilson, 2000), which also could affect SPB outbreaks. Also, human beings may respond to SPB outbreaks by altering forest management practices and salvation efforts according to the severity of damage and timber market conditions (Hedden, 1978, de Steiguer et al., 1987). Because existing models are inappropriate for quantifying the impact of climate change and fail to account for natural and human adaptation, more robust approaches and models are needed for evaluating the risk and damage of SPB infestations resulting from climate change.
To overcome these deficiencies, the panel data modeling approach is used to assess the impact of climate change on SPB outbreak risk and the value of the damage caused by SPB due to climate change. Compared to the models using time series or cross-sectional data, panel data models can alleviate collinearity and are better able to control for the effect of missing or unobserved variables (Baltagi, 2001, Hsiao, 2003). This is particularly useful for fitting regression models associated with climatic variables that are often highly correlated and in the case of data limitations on some relevant variables. Because panel data share the features of both cross-sectional and time series data, this approach also is able to account for human and natural adaptation to climate change. Therefore, panel data modeling is an appropriate approach to assessing the impacts of climate change on SPB outbreaks. The estimated risk and damage of SPB outbreaks would be helpful in better understanding the potential impact of climate change on southern pine forests.
Section snippets
Model specification
A model for panel data can generally be expressed aswhere i=1,2,…,N; t=1,2,…,T; y the dependent variable; x a k×1 vector of explanatory variables; β a k×1 vector of constants; α the individual effect; γ the time effect and ε the error term.
If for some reason, some relevant variable zit is omitted in Eq. (1), the effect of omitting zit can be accounted for by the individual or/and time effects. For instance, if zit=zi for all t, its effect will be incorporated into αi.
Risk and spatial patterns of SPB infestations
Table 1 shows the average infestation rate and volume killed by SPB annually across states from 1973 to 1996. The volume includes both pulpwood and sawtimber. Again, the infestation rate or risk represents the proportion of the volume killed against the total pine growing stock. Of the 11 states, Alabama had the highest infestation rate and volume killed, while Florida had the lowest infestation rate and volume killed. Tennessee had the second higher SPB infestation rate, followed by South
Conclusions
This article assesses the impact of climate change on the risk and damage of SPB infestations via panel data modeling. This approach alleviates the collinearity problem usually associated with climatic variables and accounts for the effect of omitted or unobserved variables. As a result, it is more appropriate for estimating the effect of individual climatic variables on SPB outbreaks. In addition, the panel data model incorporates natural and human adaptation to climate change into the impact
Acknowledgements
The author would like to thank two anonymous reviewers and a co-editor in chief for their valuable comments and suggestions. This study was financially supported by the Texas Agricultural Experiment Station. However, opinions and errors remain to be the author’s responsibility.
References (47)
- et al.
Assessing the consequences of global change for forest disturbance from herbivores and pathogens
Sci. Total Environ.
(2000) - et al.
Testing AR(1) and MA(1) disturbances in an error component model
J. Economet.
(1995) - et al.
Integrated forestry assessment for climate change impacts
For. Ecol. Manage.
(2002) - Bachelet, D., Neilson, R., 2000. Biome redistribution under climate change. In: Joyce, L., Birdsey, R. (Eds.), The...
- Baltagi, B., 2001. Econometric Analysis of Panel Data, 2nd ed. Wiley, New...
Weather as a factor in southern pine beetle control
J. For.
(1927)Temperature extremes as a factor in the ecology of the southern pine beetle
J. For.
(1929)- et al.
Analysis of transformation
J. Roy. Stat. Soc.
(1964) - et al.
Climatological forecasts of southern pine beetle infestations
S. E. Geogr.
(1980) - et al.
Southern pine beetle impacts and control policy in the 1982–1986 Texas epidemic
S. J. Appl. For.
(1991)
An investigation of the relationship between pesticide usage and climate change
Clim. Change
Bark beetle epidemics and rainfall deficiency
J. Econ. Entomol.
Attack and survival of Dendroctonus frontalis in relation to weather during three years in east Texas
Environ. Entomol.
Logistic regression for southern pine beetle outbreaks with spatial and temporal autocorrelation
For. Sci.
Efficient three dimensional global models for climate studies: Models I and II
Mon. Weather Rev.
The need for intensive forest management to reduce southern pine beetle activity in east Texas
S. J. Appl. For.
Cited by (90)
Decomposition of bark beetle-attacked trees after mortality varies across forests
2024, Forest Ecology and ManagementA new approach to evaluate the risk of bark beetle outbreaks using multi-step machine learning methods
2022, Forest Ecology and ManagementA comparison of presence-only analytical techniques and their application in forest pest modeling
2022, Ecological InformaticsCitation Excerpt :However, they may inhibit ecosystem services (e.g., carbon sequestration, timber values, and wildlife habitat) at higher population levels (at least in the short term) and may facilitate invasion by non-native species (Rosenberger et al., 2012). Negative effects of insect pests on forest health and sustainability have fueled research assessing pest outbreak risk, pest distribution and range fluctuations, and the probability of invasion or spread of non-native species (Aoki et al., 2018; Bright et al., 2020; Cudmore et al., 2010; Gan, 2004; Lantschner et al., 2014; Munro et al., 2021; Senf et al., 2017; Tribe and Cillié, 2004). However, insect pest data, especially for non-native species, are often collected as presence-only data due to the nature of sampling (i.e., collection devices have limited coverage) and cryptic nature of most insect pests at various life-stages.
Plant regeneration from seeds: Tibet Plateau in China
2022, Plant Regeneration from Seeds: A Global Warming PerspectiveThrough space and time: Predicting numbers of an eruptive pine tree pest and its predator under changing climate conditions
2021, Forest Ecology and ManagementResponses and modeling of southern pine beetle and its host pines to climate change
2021, Bark Beetle Management, Ecology, and Climate Change