Limits...
Identifying Heat Waves in Florida: Considerations of Missing Weather Data.

Leary E, Young LJ, DuClos C, Jordan MM - PLoS ONE (2015)

Bottom Line: In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors.The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September).A heat wave definition that incorporates information from all monitors is advised.

View Article: PubMed Central - PubMed

Affiliation: School of Natural Resources and Environment, University of Florida, PO Box 116455, Gainesville, FL, 32611, United States of America.

ABSTRACT

Background: Using current climate models, regional-scale changes for Florida over the next 100 years are predicted to include warming over terrestrial areas and very likely increases in the number of high temperature extremes. No uniform definition of a heat wave exists. Most past research on heat waves has focused on evaluating the aftermath of known heat waves, with minimal consideration of missing exposure information.

Objectives: To identify and discuss methods of handling and imputing missing weather data and how those methods can affect identified periods of extreme heat in Florida.

Methods: In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors. Calculated thresholds are used to define periods of extreme heat across Florida.

Results: Modeling of missing data and imputing missing values can affect the identified periods of extreme heat, through the missing data itself or through the computed thresholds. The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September).

Conclusions: Missing data considerations are important when defining periods of extreme heat. Spatio-temporal methods are recommended for data imputation. A heat wave definition that incorporates information from all monitors is advised.

Show MeSH

Related in: MedlinePlus

Imputation model results for the Gainesville weather monitor within the Jacksonville (JAX) region, for the years 1981 and 1985, using each imputation method.Imputed daily maximum heat indexes during the warm season are graphed (by day) in green with observed values dropped from the sample denoted by black dots. (A) Graph of the temporal model results for 1981. (B) Graph of the temporal model results for 1985. (C) Graph of the spatial model results for 1981. (D) Graph of the spatial model results for 1985. (E) Graph of the spatio-temporal model results for 1981. (F) Graph of the spatio-temporal model results for 1985.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4664249&req=5

pone.0143471.g002: Imputation model results for the Gainesville weather monitor within the Jacksonville (JAX) region, for the years 1981 and 1985, using each imputation method.Imputed daily maximum heat indexes during the warm season are graphed (by day) in green with observed values dropped from the sample denoted by black dots. (A) Graph of the temporal model results for 1981. (B) Graph of the temporal model results for 1985. (C) Graph of the spatial model results for 1981. (D) Graph of the spatial model results for 1985. (E) Graph of the spatio-temporal model results for 1981. (F) Graph of the spatio-temporal model results for 1985.

Mentions: To assess how well the temporal, spatial, and spatio-temporal models predicted missing daily maximum heat index values, a 10% stratified sample of daily maximum heat index measurements for all stations during the warm season for the 40 year period was taken (314,760 total observations where strata were the day, with 7320 possible days). The models for each method were fit without these sampled data, and the predicted values were compared with the observed values (Fig 2). The primary objective is to identify the method that is best able to predict missing values and, given that the observed values are available from the 10% sample, the root mean squared prediction error (RMSPE) for the sample was calculated for each model. In addition, because extreme heat is of primary interest, RMSPE was also calculated for data that were greater than 37.78°C (100°F). For extreme heat, the 97.5, 95, 90, and 80th percentiles are often considered (e.g. [3, 11–12]). The 97.5, 95, 90, and 80th percentiles for daily maximum heat index values during the warm season were estimated using the complete data (observed and imputed), derived from each of the four missing data approaches and all 40 years of data.


Identifying Heat Waves in Florida: Considerations of Missing Weather Data.

Leary E, Young LJ, DuClos C, Jordan MM - PLoS ONE (2015)

Imputation model results for the Gainesville weather monitor within the Jacksonville (JAX) region, for the years 1981 and 1985, using each imputation method.Imputed daily maximum heat indexes during the warm season are graphed (by day) in green with observed values dropped from the sample denoted by black dots. (A) Graph of the temporal model results for 1981. (B) Graph of the temporal model results for 1985. (C) Graph of the spatial model results for 1981. (D) Graph of the spatial model results for 1985. (E) Graph of the spatio-temporal model results for 1981. (F) Graph of the spatio-temporal model results for 1985.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4664249&req=5

pone.0143471.g002: Imputation model results for the Gainesville weather monitor within the Jacksonville (JAX) region, for the years 1981 and 1985, using each imputation method.Imputed daily maximum heat indexes during the warm season are graphed (by day) in green with observed values dropped from the sample denoted by black dots. (A) Graph of the temporal model results for 1981. (B) Graph of the temporal model results for 1985. (C) Graph of the spatial model results for 1981. (D) Graph of the spatial model results for 1985. (E) Graph of the spatio-temporal model results for 1981. (F) Graph of the spatio-temporal model results for 1985.
Mentions: To assess how well the temporal, spatial, and spatio-temporal models predicted missing daily maximum heat index values, a 10% stratified sample of daily maximum heat index measurements for all stations during the warm season for the 40 year period was taken (314,760 total observations where strata were the day, with 7320 possible days). The models for each method were fit without these sampled data, and the predicted values were compared with the observed values (Fig 2). The primary objective is to identify the method that is best able to predict missing values and, given that the observed values are available from the 10% sample, the root mean squared prediction error (RMSPE) for the sample was calculated for each model. In addition, because extreme heat is of primary interest, RMSPE was also calculated for data that were greater than 37.78°C (100°F). For extreme heat, the 97.5, 95, 90, and 80th percentiles are often considered (e.g. [3, 11–12]). The 97.5, 95, 90, and 80th percentiles for daily maximum heat index values during the warm season were estimated using the complete data (observed and imputed), derived from each of the four missing data approaches and all 40 years of data.

Bottom Line: In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors.The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September).A heat wave definition that incorporates information from all monitors is advised.

View Article: PubMed Central - PubMed

Affiliation: School of Natural Resources and Environment, University of Florida, PO Box 116455, Gainesville, FL, 32611, United States of America.

ABSTRACT

Background: Using current climate models, regional-scale changes for Florida over the next 100 years are predicted to include warming over terrestrial areas and very likely increases in the number of high temperature extremes. No uniform definition of a heat wave exists. Most past research on heat waves has focused on evaluating the aftermath of known heat waves, with minimal consideration of missing exposure information.

Objectives: To identify and discuss methods of handling and imputing missing weather data and how those methods can affect identified periods of extreme heat in Florida.

Methods: In addition to ignoring missing data, temporal, spatial, and spatio-temporal models are described and utilized to impute missing historical weather data from 1973 to 2012 from 43 Florida weather monitors. Calculated thresholds are used to define periods of extreme heat across Florida.

Results: Modeling of missing data and imputing missing values can affect the identified periods of extreme heat, through the missing data itself or through the computed thresholds. The differences observed are related to the amount of missingness during June, July, and August, the warmest months of the warm season (April through September).

Conclusions: Missing data considerations are important when defining periods of extreme heat. Spatio-temporal methods are recommended for data imputation. A heat wave definition that incorporates information from all monitors is advised.

Show MeSH
Related in: MedlinePlus