Modeling spatiotemporal forest health monitoring data

Nicole H. Augustin, Monica Musio, Klaus von Wilpert, Edgar Kublin, Simon N. Wood, Martin Schumacher

Research output: Contribution to journalArticle

56 Citations (Scopus)
96 Downloads (Pure)

Abstract

Forest health monitoring schemes were set Lip across Europe in the 1980s in response to concerns about air pollution-related forest dieback (Waldsterben) and have continued since then. Recent threats to forest health are climatic extremes likely due to global climate change and increased ground ozone levels and nitrogen deposition. We model yearly data on tree crown defoliation, an indicator of tree health, from a monitoring survey carried Out in Baden-Wurttemberg, Germany since 1983. On a changing irregular grid, defoliation and other sue-specific variables are recorded, In Baden-Wurttemberg, the temporal trend of defoliation differs among areas because of site characteristics and pollution levels, making it necessary to allow for space-time interaction in the model. For this purpose, we propose using generalized additive mixed models (GAMMs) incorporating scale-invariant tensor product smooths of the space-time dimensions. The space-time smoother allows separate smoothing parameters and penalties for the space and time dimensions and thus avoids the need to make arbitrary or ad hoc choices about the relative scaling of space and time. The approach of using a space-time smoother has intuitive appeal, making it easy to explain and interpret when communicating the results to nonstaticians, such as environmental policy makers. The model incorporates a nonlinear effect for mean tree age. the most important predictor. allowing the separation of trends in time. which may be pollution-related, from trends that relate purely to the aging of the survey population. In addition to a temporal trend due to site characteristics and other conditions modeled with the space-time Smooth, We account for random temporal correlation at site level by an autoregressive moving average (ARMA) process. Model selection is carried Out using the Bayes information criterion (BIC). and the adequacy of the assumed spatial and temporal error Structure is investigated with the empirical semivariogram and the empirical autocorrelation function.
Original languageEnglish
Pages (from-to)899-911
Number of pages13
JournalJournal of the American Statistical Association
Volume104
Issue number487
DOIs
Publication statusPublished - Sep 2009

Fingerprint

Spatio-temporal Modeling
Health Monitoring
Defoliation
Space-time
Pollution
Health
Bayes Information Criterion
Semivariogram
Irregular Grids
Moving Average Process
Additive Models
Temporal Correlation
Ozone
Air Pollution
Autoregressive Moving Average
Smoothing Parameter
Autoregressive Process
Appeal
Scale Invariant
Mixed Model

Cite this

Modeling spatiotemporal forest health monitoring data. / Augustin, Nicole H.; Musio, Monica; von Wilpert, Klaus; Kublin, Edgar; Wood, Simon N.; Schumacher, Martin.

In: Journal of the American Statistical Association, Vol. 104, No. 487, 09.2009, p. 899-911.

Research output: Contribution to journalArticle

Augustin, NH, Musio, M, von Wilpert, K, Kublin, E, Wood, SN & Schumacher, M 2009, 'Modeling spatiotemporal forest health monitoring data', Journal of the American Statistical Association, vol. 104, no. 487, pp. 899-911. https://doi.org/10.1198/jasa.2009.ap07058
Augustin, Nicole H. ; Musio, Monica ; von Wilpert, Klaus ; Kublin, Edgar ; Wood, Simon N. ; Schumacher, Martin. / Modeling spatiotemporal forest health monitoring data. In: Journal of the American Statistical Association. 2009 ; Vol. 104, No. 487. pp. 899-911.
@article{d41b5e768c5a44b991c8083f8b4656e5,
title = "Modeling spatiotemporal forest health monitoring data",
abstract = "Forest health monitoring schemes were set Lip across Europe in the 1980s in response to concerns about air pollution-related forest dieback (Waldsterben) and have continued since then. Recent threats to forest health are climatic extremes likely due to global climate change and increased ground ozone levels and nitrogen deposition. We model yearly data on tree crown defoliation, an indicator of tree health, from a monitoring survey carried Out in Baden-Wurttemberg, Germany since 1983. On a changing irregular grid, defoliation and other sue-specific variables are recorded, In Baden-Wurttemberg, the temporal trend of defoliation differs among areas because of site characteristics and pollution levels, making it necessary to allow for space-time interaction in the model. For this purpose, we propose using generalized additive mixed models (GAMMs) incorporating scale-invariant tensor product smooths of the space-time dimensions. The space-time smoother allows separate smoothing parameters and penalties for the space and time dimensions and thus avoids the need to make arbitrary or ad hoc choices about the relative scaling of space and time. The approach of using a space-time smoother has intuitive appeal, making it easy to explain and interpret when communicating the results to nonstaticians, such as environmental policy makers. The model incorporates a nonlinear effect for mean tree age. the most important predictor. allowing the separation of trends in time. which may be pollution-related, from trends that relate purely to the aging of the survey population. In addition to a temporal trend due to site characteristics and other conditions modeled with the space-time Smooth, We account for random temporal correlation at site level by an autoregressive moving average (ARMA) process. Model selection is carried Out using the Bayes information criterion (BIC). and the adequacy of the assumed spatial and temporal error Structure is investigated with the empirical semivariogram and the empirical autocorrelation function.",
author = "Augustin, {Nicole H.} and Monica Musio and {von Wilpert}, Klaus and Edgar Kublin and Wood, {Simon N.} and Martin Schumacher",
year = "2009",
month = "9",
doi = "10.1198/jasa.2009.ap07058",
language = "English",
volume = "104",
pages = "899--911",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis",
number = "487",

}

TY - JOUR

T1 - Modeling spatiotemporal forest health monitoring data

AU - Augustin, Nicole H.

AU - Musio, Monica

AU - von Wilpert, Klaus

AU - Kublin, Edgar

AU - Wood, Simon N.

AU - Schumacher, Martin

PY - 2009/9

Y1 - 2009/9

N2 - Forest health monitoring schemes were set Lip across Europe in the 1980s in response to concerns about air pollution-related forest dieback (Waldsterben) and have continued since then. Recent threats to forest health are climatic extremes likely due to global climate change and increased ground ozone levels and nitrogen deposition. We model yearly data on tree crown defoliation, an indicator of tree health, from a monitoring survey carried Out in Baden-Wurttemberg, Germany since 1983. On a changing irregular grid, defoliation and other sue-specific variables are recorded, In Baden-Wurttemberg, the temporal trend of defoliation differs among areas because of site characteristics and pollution levels, making it necessary to allow for space-time interaction in the model. For this purpose, we propose using generalized additive mixed models (GAMMs) incorporating scale-invariant tensor product smooths of the space-time dimensions. The space-time smoother allows separate smoothing parameters and penalties for the space and time dimensions and thus avoids the need to make arbitrary or ad hoc choices about the relative scaling of space and time. The approach of using a space-time smoother has intuitive appeal, making it easy to explain and interpret when communicating the results to nonstaticians, such as environmental policy makers. The model incorporates a nonlinear effect for mean tree age. the most important predictor. allowing the separation of trends in time. which may be pollution-related, from trends that relate purely to the aging of the survey population. In addition to a temporal trend due to site characteristics and other conditions modeled with the space-time Smooth, We account for random temporal correlation at site level by an autoregressive moving average (ARMA) process. Model selection is carried Out using the Bayes information criterion (BIC). and the adequacy of the assumed spatial and temporal error Structure is investigated with the empirical semivariogram and the empirical autocorrelation function.

AB - Forest health monitoring schemes were set Lip across Europe in the 1980s in response to concerns about air pollution-related forest dieback (Waldsterben) and have continued since then. Recent threats to forest health are climatic extremes likely due to global climate change and increased ground ozone levels and nitrogen deposition. We model yearly data on tree crown defoliation, an indicator of tree health, from a monitoring survey carried Out in Baden-Wurttemberg, Germany since 1983. On a changing irregular grid, defoliation and other sue-specific variables are recorded, In Baden-Wurttemberg, the temporal trend of defoliation differs among areas because of site characteristics and pollution levels, making it necessary to allow for space-time interaction in the model. For this purpose, we propose using generalized additive mixed models (GAMMs) incorporating scale-invariant tensor product smooths of the space-time dimensions. The space-time smoother allows separate smoothing parameters and penalties for the space and time dimensions and thus avoids the need to make arbitrary or ad hoc choices about the relative scaling of space and time. The approach of using a space-time smoother has intuitive appeal, making it easy to explain and interpret when communicating the results to nonstaticians, such as environmental policy makers. The model incorporates a nonlinear effect for mean tree age. the most important predictor. allowing the separation of trends in time. which may be pollution-related, from trends that relate purely to the aging of the survey population. In addition to a temporal trend due to site characteristics and other conditions modeled with the space-time Smooth, We account for random temporal correlation at site level by an autoregressive moving average (ARMA) process. Model selection is carried Out using the Bayes information criterion (BIC). and the adequacy of the assumed spatial and temporal error Structure is investigated with the empirical semivariogram and the empirical autocorrelation function.

UR - http://www.scopus.com/inward/record.url?scp=70349777916&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1198/jasa.2009.ap07058

U2 - 10.1198/jasa.2009.ap07058

DO - 10.1198/jasa.2009.ap07058

M3 - Article

VL - 104

SP - 899

EP - 911

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 487

ER -