MS048
Measurement differences between air temperature instruments used at H.J. Andrews meteorological stations

CREATOR(S): Stephanie A Schmidt, Mark D Schulze, Christopher Daly

PRINCIPAL INVESTIGATOR(S): Mark D Schulze

ORIGINATOR(S): Christopher Daly

OTHER RESEARCHER(S): Sherri L. Johnson, Stephanie A Schmidt, Adam M Kennedy, Greg M Cohn

DATA SET CONTACT PERSON: Mark D Schulze

FORMER INVESTIGATOR: John Moreau, Frederick A. Bierlmaier

DATA SET CREDIT:

This work was funded by the Andrews LTER (National Science Foundation grant DEB 1440409) and USDA Forest Service, Pacific Northwest Research Station. Data and facilities were provided by the HJ Andrews Experimental Forest and Long Term Ecological Research program, administered cooperatively by the USDA Forest Service Pacific Northwest Research Station, Oregon State University, and the Willamette National Forest.

MOST RECENT METADATA REVIEW DATE:

29 Jan 2021

KEYWORDS:

air temperature, maximum temperature

PURPOSE:

To examine the measurement differences between air temperature instruments that are using the same sensor but different radiation shields. These measurement differences are examined for various temperature statistics (average, minimum, maximum) at varying temporal resolutions. This information will help researchers understand the magnitude of measurement bias associated with instruments that have been used to record air temperature at H.J. Andrews meteorological stations over time. This bias can have implications during analysis when air temperature data from different instruments is combined into one dataset without accounting for differences between air temperature methods.

METHODS:

Experimental Design - MS048:

Description: Instruments measuring air temperature, wind speed, incoming solar radiation and reflected solar radiation were placed side-by-side at 1.5m above the ground. There were four air temperature instruments, all using an identical sensor (Campbell Scientific 107 thermistor) but different radiation shields: 1) fan-aspirated (ASP or aspirated), 2) cotton region shelter (CRS), 3) Gill multi-plate (Gill), and 4) a custom-fabricated shield developed by staff at the research site (HJA). One propeller anemometer was used to measure wind speed. One upward-facing pyranometer measured incoming solar radiation (ISR), and one down-facing pyranometer measured reflected solar radiation (RSR). All sensors were connected to the same Campbell Scientific CR1000 data logger.

Instrumentation:

Four air temperature instruments were deployed, all using the same Campbell Scientific 107 thermistor sensor (Sensor accuracy: +/-0.2 deg C between 0 and 60; +/-0.4 deg C below 0 to -35) and four different radiation shields:

1) fan-aspirated (ASP or aspirated) - R.M. Young 43502-L compact aspirated shield (shield radiation error <0.2 deg C @ 1000 W/m2)

2) cotton region shelter (CRS) - NovaLynx 380-605 large instrument shelter (no accuracy information available)

3) Gill multi-plate (Gill) - R.M. Young 41303-5A 6-plate solar radiation shield (no accuracy information available)

4) a custom-fabricated shield developed by staff at the research site (HJA) - 8-inch long, 3.5 inch diameter, schedule 40 PVC pipe split in half lengthwise over PVC tubing containing sensor (no accuracy information available)

One propeller anemometer was used to measure wind speed - R.M. Young 05103 (Resolution: 1.0 m/s ; accuracy: +/-0.3 m/s or 1% of reading)

One upward-facing pyranometer measured incoming solar radiation (ISR), and one down-facing pyranometer measured reflected solar radiation – both CM3 Kipp & Zonen (Response time: 18s; spectral range 305-2800 nm; sensitivity: 10-35 micro V/W/m2; error: +/-25 W/m2)

All sensors were connected to the same data logger - Campbell Scientific CR1000

None of the instruments were calibrated during the course of the study.

Field Methods - MS048:

Description: Instruments measuring air temperature, wind speed, incoming solar radiation and reflected solar radiation were placed side-by-side at 1.5m above the ground. There were four air temperature instruments, all using an identical sensor (Campbell Scientific 107 thermistor) but different radiation shields:

1) fan-aspirated (ASP or aspirated),

2) cotton region shelter (CRS),

3) Gill multi-plate (Gill), and

4) a custom-fabricated shield developed by staff at the research site (HJA).

One propeller anemometer was used to measure wind speed.

One upward-facing pyranometer measured incoming solar radiation (ISR), and one down-facing pyranometer measured reflected solar radiation (RSR).

All sensors were connected to the same Campbell Scientific CR1000 data logger.

Air temperature, incoming and reflected solar radiation, and wind speed data were collected continuously between July 2010 and April 2017. Measurements were taken every 15 seconds, and the average for a 15-minute period (60 observations) was recorded (referred to hereafter as “15-minute average observations”). Additionally, from April 2015 to April 2017, measurements were taken every 15 seconds, and for each 5-minute period (20 observations), the maximum and minimum air temperature and maximum incoming and reflected solar radiation and wind speed were recorded (referred to hereafter as “5-minute observations” with the relevant statistic specified).

Instrumentation:

Four air temperature instruments all used the same Campbell Scientific 107 thermistor sensor (Sensor accuracy: +/-0.2 deg C between 0 and 60; +/-0.4 deg C below 0 to -35) and four different radiation shields:

1) fan-aspirated (ASP or aspirated) - R.M. Young 43502-L compact aspirated shield (shield radiation error <0.2 deg C @ 1000 W/m2)

2) cotton region shelter (CRS) - NovaLynx 380-605 large instrument shelter (no accuracy information available)

3) Gill multi-plate (Gill) - R.M. Young 41303-5A 6-plate solar radiation shield (no accuracy information available)

4) a custom-fabricated shield developed by staff at the research site (HJA) - 8-inch long, 3.5 inch diameter, schedule 40 PVC pipe split in half lengthwise over PVC tubing containing sensor (no accuracy information available)

One propeller anemometer was used to measure wind speed - R.M. Young 05103 (Resolution: 1.0 m/s ; accuracy: +/-0.3 m/s or 1% of reading)

One upward-facing pyranometer measured incoming solar radiation (ISR), and one down-facing pyranometer measured reflected solar radiation – both CM3 Kipp & Zonen (Response time: 18s; spectral range 305-2800 nm; sensitivity: 10-35 micro V/W/m2; error: +/-25 W/m2)

All sensors were connected to the same data logger - Campbell Scientific CR1000

None of the instruments were calibrated during the course of the study.

Quality Assurance - MS048:

Description:

Air temperature, incoming and reflected solar radiation, and wind speed data were collected continuously between July 2010 and April 2017. Measurements were taken every 15 seconds, and the average for a 15-minute period (60 observations) was recorded (referred to hereafter as “15-minute average observations”). Additionally, from April 2015 to April 2017, measurements were taken every 15 seconds, and for each 5-minute period (20 observations), the maximum and minimum air temperature and maximum incoming and reflected solar radiation and wind speed were recorded (referred to hereafter as “5-minute observations” with the relevant statistic specified).

All data were reviewed for quality control using an automated process (GCE Data Toolbox Version 3.9.4b - https://gce-lter.marsci.uga.edu/public/im/tools/data_toolbox.htm). Rows with missing observations for aspirated air temperature, incoming solar radiation, reflected solar radiation, or wind speed were removed, with a total of 6,097 rows (out of 210,504) removed from the 15-minute data and 8,358 rows (out of 210,792) removed from the 5-minute data. Data logger issues from February 27-March 7, 2013 and August 2-31 in 2016 caused most missing values.

SITE DESCRIPTION:

The study was located at the Primary Meteorological Station (PRIMET) at H.J. Andrews Experimental Forest, located in Oregon’s Western Cascades (44.21, -122.26, elevation 430m). PRIMET is sited in a relatively flat clearing approximately 29 x 23 m2 in size, with conifer forests outside the perimeter. It is located in the bottom of a valley with steep ridges to the southeast and northwest. Annual precipitation averages 2200 mm and occurs primarily November through May. Snow can accumulate at this site in winter, but it does not remain for long periods. Summers are generally dry and warm, with cool nights and occasional thunderstorms.

TAXONOMIC SYSTEM:

None

GEOGRAPHIC EXTENT:

H.J. Andrews Experimental Forest, Western Cascades, Oregon, Primary Meteorological Station

MEASUREMENT FREQUENCY:

15 and 5 minute

PROGRESS DESCRIPTION:

Complete

UPDATE FREQUENCY DESCRIPTION:

notPlanned

CURRENTNESS REFERENCE:

Observed