Comparing daily temperature averaging methods: the role of surface and atmosphere variables in determining spatial and seasonal variability

Jase Bernhardt, Andrew Mark Carleton

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The two main methods for determining the average daily near-surface air temperature, twice-daily averaging (i.e., [Tmax+Tmin]/2) and hourly averaging (i.e., the average of 24 hourly temperature measurements), typically show differences associated with the asymmetry of the daily temperature curve. To quantify the relative influence of several land surface and atmosphere variables on the two temperature averaging methods, we correlate data for 215 weather stations across the Contiguous United States (CONUS) for the period 1981–2010 with the differences between the two temperature-averaging methods. The variables are land use-land cover (LULC) type, soil moisture, snow cover, cloud cover, atmospheric moisture (i.e., specific humidity, dew point temperature), and precipitation. Multiple linear regression models explain the spatial and monthly variations in the difference between the two temperature-averaging methods. We find statistically significant correlations between both the land surface and atmosphere variables studied with the difference between temperature-averaging methods, especially for the extreme (i.e., summer, winter) seasons (adjusted R2 > 0.50). Models considering stations with certain LULC types, particularly forest and developed land, have adjusted R2 values > 0.70, indicating that both surface and atmosphere variables control the daily temperature curve and its asymmetry. This study improves our understanding of the role of surface and near-surface conditions in modifying thermal climates of the CONUS for a wide range of environments, and their likely importance as anthropogenic forcings—notably LULC changes and greenhouse gas emissions—continues.

Original languageEnglish (US)
Pages (from-to)499-512
Number of pages14
JournalTheoretical and Applied Climatology
Volume136
Issue number1-2
DOIs
StatePublished - Apr 12 2019

Fingerprint

atmosphere
temperature
land cover
land use
land surface
asymmetry
atmospheric moisture
dew point
method
weather station
cloud cover
snow cover
humidity
greenhouse gas
surface temperature
air temperature
soil moisture
winter
climate
summer

All Science Journal Classification (ASJC) codes

  • Atmospheric Science

Cite this

@article{5729359decf44a94aba513f443b71725,
title = "Comparing daily temperature averaging methods: the role of surface and atmosphere variables in determining spatial and seasonal variability",
abstract = "The two main methods for determining the average daily near-surface air temperature, twice-daily averaging (i.e., [Tmax+Tmin]/2) and hourly averaging (i.e., the average of 24 hourly temperature measurements), typically show differences associated with the asymmetry of the daily temperature curve. To quantify the relative influence of several land surface and atmosphere variables on the two temperature averaging methods, we correlate data for 215 weather stations across the Contiguous United States (CONUS) for the period 1981–2010 with the differences between the two temperature-averaging methods. The variables are land use-land cover (LULC) type, soil moisture, snow cover, cloud cover, atmospheric moisture (i.e., specific humidity, dew point temperature), and precipitation. Multiple linear regression models explain the spatial and monthly variations in the difference between the two temperature-averaging methods. We find statistically significant correlations between both the land surface and atmosphere variables studied with the difference between temperature-averaging methods, especially for the extreme (i.e., summer, winter) seasons (adjusted R2 > 0.50). Models considering stations with certain LULC types, particularly forest and developed land, have adjusted R2 values > 0.70, indicating that both surface and atmosphere variables control the daily temperature curve and its asymmetry. This study improves our understanding of the role of surface and near-surface conditions in modifying thermal climates of the CONUS for a wide range of environments, and their likely importance as anthropogenic forcings—notably LULC changes and greenhouse gas emissions—continues.",
author = "Jase Bernhardt and Carleton, {Andrew Mark}",
year = "2019",
month = "4",
day = "12",
doi = "10.1007/s00704-018-2504-7",
language = "English (US)",
volume = "136",
pages = "499--512",
journal = "Theorectical and Applied Climatology",
issn = "0177-798X",
publisher = "Springer Wien",
number = "1-2",

}

TY - JOUR

T1 - Comparing daily temperature averaging methods

T2 - the role of surface and atmosphere variables in determining spatial and seasonal variability

AU - Bernhardt, Jase

AU - Carleton, Andrew Mark

PY - 2019/4/12

Y1 - 2019/4/12

N2 - The two main methods for determining the average daily near-surface air temperature, twice-daily averaging (i.e., [Tmax+Tmin]/2) and hourly averaging (i.e., the average of 24 hourly temperature measurements), typically show differences associated with the asymmetry of the daily temperature curve. To quantify the relative influence of several land surface and atmosphere variables on the two temperature averaging methods, we correlate data for 215 weather stations across the Contiguous United States (CONUS) for the period 1981–2010 with the differences between the two temperature-averaging methods. The variables are land use-land cover (LULC) type, soil moisture, snow cover, cloud cover, atmospheric moisture (i.e., specific humidity, dew point temperature), and precipitation. Multiple linear regression models explain the spatial and monthly variations in the difference between the two temperature-averaging methods. We find statistically significant correlations between both the land surface and atmosphere variables studied with the difference between temperature-averaging methods, especially for the extreme (i.e., summer, winter) seasons (adjusted R2 > 0.50). Models considering stations with certain LULC types, particularly forest and developed land, have adjusted R2 values > 0.70, indicating that both surface and atmosphere variables control the daily temperature curve and its asymmetry. This study improves our understanding of the role of surface and near-surface conditions in modifying thermal climates of the CONUS for a wide range of environments, and their likely importance as anthropogenic forcings—notably LULC changes and greenhouse gas emissions—continues.

AB - The two main methods for determining the average daily near-surface air temperature, twice-daily averaging (i.e., [Tmax+Tmin]/2) and hourly averaging (i.e., the average of 24 hourly temperature measurements), typically show differences associated with the asymmetry of the daily temperature curve. To quantify the relative influence of several land surface and atmosphere variables on the two temperature averaging methods, we correlate data for 215 weather stations across the Contiguous United States (CONUS) for the period 1981–2010 with the differences between the two temperature-averaging methods. The variables are land use-land cover (LULC) type, soil moisture, snow cover, cloud cover, atmospheric moisture (i.e., specific humidity, dew point temperature), and precipitation. Multiple linear regression models explain the spatial and monthly variations in the difference between the two temperature-averaging methods. We find statistically significant correlations between both the land surface and atmosphere variables studied with the difference between temperature-averaging methods, especially for the extreme (i.e., summer, winter) seasons (adjusted R2 > 0.50). Models considering stations with certain LULC types, particularly forest and developed land, have adjusted R2 values > 0.70, indicating that both surface and atmosphere variables control the daily temperature curve and its asymmetry. This study improves our understanding of the role of surface and near-surface conditions in modifying thermal climates of the CONUS for a wide range of environments, and their likely importance as anthropogenic forcings—notably LULC changes and greenhouse gas emissions—continues.

UR - http://www.scopus.com/inward/record.url?scp=85046806879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046806879&partnerID=8YFLogxK

U2 - 10.1007/s00704-018-2504-7

DO - 10.1007/s00704-018-2504-7

M3 - Article

AN - SCOPUS:85046806879

VL - 136

SP - 499

EP - 512

JO - Theorectical and Applied Climatology

JF - Theorectical and Applied Climatology

SN - 0177-798X

IS - 1-2

ER -