DRAFT DRAFT DRAFT

THIS DOCUMENT IS UNDER CONSTRUCTION AND HAS NOT YET BEEN APPROVED BY THE LTER COORDINATING COMMITTEE. UNTIL APPROVAL IS GIVEN THE DOCUMENT IS NOT THE OFFICIAL POLICY OF THE LTER PROGRAM.

LAST UPDATED: 4/17/98

METEOROLOGICAL OBSERVATION STANDARDS FOR LTER SITES

October 1997

TABLE OF CONTENTS

INTRODUCTION

Objectives, Levels of Participation, Designated Primary Observing Station, Station Site Selection,

Observation Record, Intersite Exchange of Data

LEVEL O METEOROLOGY

Instrumentation, Measurements, Reporting

LEVEL 1 METEOROLOGY

Instrumentation, Measurements, Preprocessing

LEVEL 2 METEOROLOGY

Instrumentation, Measurements, Preprocessing

LEVEL 3 METEOROLOGY

Instrumentation, Measurements

LEVEL 4 METEOROLOGY

Standardization of Specialized Measurements

The National Atmospheric Deposition Program

CALIBRATION

LITERATURE CITED

Appendix 1 Measurement of Solar Radiation

Appendix 2 Components of a typical Level 2 Meteorological site

Appendix 3 Estimation of vapor pressure

Appendix 4 Measurement of Wind Speed and Direction

INTRODUCTION

Each LongTerm Ecological Research (LTER) site assumes an obligation to collect and make available data to characterize the ecosystem which the site represents. This document is the LTER guide for assembling meteorological data. All LTER sites and new LTER sites should consider following the procedures outlined here. This document defines, for the LTER program, the measurement and reporting standards for meteorological data. The standards are based on earlier LTER proceedures, the needs at existing LTER sites,

other standards existing in the literature, and the substantial experience of current LTER scientists.

Objectives

The objectives of standardized meteorological measurements are:

1) establish baseline meteorological measurements to characterize each LTER site and enable intersite comparisons,

2) document for LTER objectives both cyclic and long-term changes in the physical environment,

3) provide a climatic history for each site's core research program to correlate with bioecological phenomena and to provide data for modeling,

4) provide a basis for coordinating specialized or short term meteorological measurements at two or more sites when such measurements are required for specific research problems.

We recognize that the rate of implementation of attaining these objectives is largely driven by the availability and cost of technology. Future technological advances will make the attaining of higher levels of participation easier but the ecological basis for the above objectives will remain essentially the same.

Levels of Participation

The diversity of sites and their core research programs argue against a single inclusive set of standard measurements. Consequently, LTER meteorological measurements are grouped into five levels of standardized measurements, a plan which establishes degrees of uniformity for intersite comparative data yet allows flexibility for the site specific requirements of each core research program. This heirarchical principle has found considerable use in other areas of ecological work such as general classification studies. The committee recognizes that sites can have an observation program which falls between these levels. The levels are set up to facilitate interersite description. The five levels are:

Level 0: The entry level meteorological measurements of maximum and minimum temperature and precipitation amounts over 24 hour periods.

Level 1: A basic climatic station using standard measurements and instruments to measure temperature and precipitation on a continuous basis throughout the day. Data will be extracted for specific times or intervals to serve the climatological goals of objectives 1 and 2. All LTER sites must achieve Level 1.

Level 2: A research meteorological station having more intensive measurements in order to characterize in detail both long and short-term meteorological events affecting biological systems. These stations sense temperature, precipitation, and other variables on a continuous basis, may record observations digitally, and may have the capability to extract instantaneous observations or do integrations on a real time basis. Most LTER sites will seek to meet, and have met, Level 2 standards for some or all measurements.

Level 3: This includes a number of variables which we beleive it would be optimal for sites to record but for which funds may not be available and funding priorities must be set.

These variables include: Photosynthetically Active Radiation (PAR), Absorbed Photsynthetically Active Radiation (APAR), Soil temperature, Soil moisture, Atmospheric pressure, Vapor pressure.

Level 4: At various times, the research program at each LTER site may require additional specialized meteorological measurements directly related to local research needs. At the conceptual stage of a study, researchers will benefit from coordination, with other LTER groups where appropriate, in order to develop standardized techniques, identify mutual interests, and facilitate short term data collection and potential intersite comparisons.

The entry level, Level 0, is available for new sites to the LTER program. We regard the existence of this level as an interim measure only and sites are urged to proceed to the next levels as soon as possible and certainly within one year. Since technology is now easily available for level two observations, we consider level two to be the "standard" level for LTER sites.

Almost twenty years of experience has allowed us to understand how the hierarchy concept operates in practice and how we may use it better in the future. LTER sites make local decisions regarding levels of measurement implementations based upon individual site priorities. Productive site implementations relevant to the network of sites may be identified over time as emergent standards and can be considered when the level descriptions are reviewed. Periodic summary tables and reviews of all site implementations are beneficial and are recommended in the future and will be incorporated in future editions of this document and in other places when opportuinites arise. Such summaries and reviews provide a clear statement of the corporate priority decision making process of the network over time.

An initial summary of expectations at the levels mentioned above is as follows:

Table 1: Summary of Variables Measured at Different Levels.

Archiving of the data collected may be in any form at a site with the exception that sites are required to write their own filters to ensure that their data is compatible with the ClimDB data harvesting system described in other parts of this document (hyperlink). The concepts of filters and data harvesting are explained in the intersite data exchange part of this document.

Designated Primary Observing Station

It is common for LTER sites to develop more than one meteorological observing site. Extra sites may be used at any of the levels described here. However each LTER site should designate one observing station as its primary observing station. The primary observing station will be used for intersite studies unless there is some sound scientific reason for using another station at a site for such a study. In the latter case the reason for not using the designated station should be fully explained in the study.

Station Site Selection

The LTER meteorological station which is to be designated as the primary station should be located where surface measurements will record, as best as possible, representative conditions for the LTER site. A level area is more desirable than an unusual topographic setting. The station should not be on a slope, a ridge, or in a sheltered area unless such extreme positions are representative of the LTER site. Substations may be located to establish the range of conditions at a site.

The primary station should be located where surroundings are uniform. For example on a sod base at least 30 meters from hard surface areas such as asphalt or concrete and stations should be no closer to vertical obstructions (trees, buildings, etc.) than four times the height of the obstructions (USDC, 1989). At LTER sites with very tall trees this instruction may not be practical. Instead, the measurement site should be selected so that it has at least a 35 degree horizon i.e. no obstacles should be above 35 degrees on the local horizon. Similarly for lake and other aquatic sites it may be more expedient to record wind at a low level on or near the lake. In the case of the wind variable important local considerations will be permitted to take precedence over intersite standardization.

Exceptions to the all or parts of the above may be required by individual sites. The exceptions should be clearly outlined in the metadata provided with the LTER primary meteorological observing site.

The Observation Record

The original record of meteorological data for an LTER site will be retained. Entries in an Observation Record or log made when instruments are read, original chart recordings and printouts of electronic records, and in some cases, the records themselves, are examples of original records. Retention is important for verification of derived data because the Observation Record usually contains the comments necessary to establish the station history. Intersite reports are climatological summaries, and thus detailed data for onsite and intersite studies will be made only in the original record. Where possible, the original record will be available onsite to researchers from other sites for research activities requiring the primary record of meteorological data. It is recognized that it is not possible for all original electronic records to be retained since these data are often neccessarily transferred from one platform to another. In such cases the principle of keeping the original data should be maintained as much as possible. This, for example will require the provision of good metadata and data quality flags, including in some cases text notes, for data variables.

Intersite Exchange of Data

Each site must make available data from levels 0, 1, and 2 as soon as possible after collection and quality control has been done. LTER data exchange files may be a subset of the total data collected at a site. Information collected as part of a specific study at a single site may be reserved by the scientist until the results are analyzed and reported. However all data must meet the NSF mandated regulations for being made accessible to the public. Data will be processed and available for exchange between LTER sites through the ClimDB data base system described in a separate part of this document. Data should be used on a user beware basis. Although every effort has been made for quality control and the identification of questionable data, it is still possibe that errors of various kinds may be contained within the data. Users are especially encouraged to read the metadata files associated with the data of interest to them. If any possible errors are found users should contact the site contact person.

Return to Table of Contents

LEVEL 0 METEOROLOGY

Level 0 Meteorology is strictly manual daily reporting. A site may choose to initiate meteorological measurements with Level 0 as a temporary expedient. An existing Cooperative Observer station for the National Weather Service might be used as a proxy until the LTER site can establish its own station.

Instrumentation

Temperature instrumentation (Table 2) will consist of maximum and minimum thermometers mounted in a National Weather Service type instrument shelter. The installation of the instrument shelter and thermometers will follow the guidelines of the National Weather Service for Cooperative Observer Stations (USDC, 1989). When NWS standards for electronic temperature measurements at cooperative observer stations are established, these will also be acceptable guidelines for LTER sites.

The nonrecording precipitation gage (Table 2) should be located no nearer the instrument shelter than twice the height of the instrument shelter. At exposed windswept sites a windshield may be required for the precipitation gage. The gage must be elevated above maximum snow depth, and, if possible, operation should continue during freezing weather. These considerations are covered in the Observing Handbook No. 2 (USDC, 1989) which is the guideline document for the precipitation gage.

Measurements

Daily observation of precipitation and temperature is necessary at level 0 (Table 2). All daily meteorological measurements and comments on station operation will be entered into a Permanent Observation Record or log which is the official data source for calculated values appearing in LTER intersite reports. Observations made early in the morning are interpreted as representing conditions on the previous day. We suggest observations be made between 0500 and 0900 hours. The observation time should be as consistent as possible from day to day and should be noted in the observation log. This specification of observation time is important since daily, monthly or longer mean temperatures, calculated from daily maximum and minimum values, may be biased by the time of observation by as much as 2 or 3°C compared to the midnight to midnight reading (Baker, 1975, Karl et al. 1986).

Reporting

A monthly intersite report for Level 0 meteorology (Table 3) will consist of daily values for maximum, minimum and average daily temperature and total precipitation. Monthly means for maximum, minimum and average daily temperature will be calculated along with total monthly precipitation. Maximum and minimum temperatures will be reported in degrees Celsius. The daily mean air temperature will be the average of the maximum and minimum temperatures. Daily precipitation will be reported as mm of water and the water equivalent depth of snow will be recorded (USDC, 1989).

Table 2. Level 0 Meteorological Station Equipment

Table 3. Level 0 Meteorological Measurements

Table 4. Level 0 Meteorological Summaries.

Return to Table of Contents

LEVEL 1 METEOROLOGY

Level 1 Meteorology involves continuous recording by a mechanical recorder. Sometimes, an LTER site will begin its direct meteorological reporting at Level 1. Level 1 is still regarded as interim, however; all sites should eventually maintain a Level 2 meteorological station (next main section).

Instrumentation

The instruments required for a Level 1 station (Table 5) contain some which record continuously. Other instruments are identical to those used for level 0 Table 2.

Table 5. Level 1 Meteorological Station Equipment

Installation of the instrument shelter, air thermometers and preciptation gage should follow the NWS guidelines for Cooperative Observer Stations (USDC, 1989). When NWS standards for electronic temperature measurements at cooperative observer stations are established, these will also be acceptable guidelines for LTER sites. The level 1 site has much the same instrumentation as the level 0 site (Table 2).

Measurements

Maximum and mininum temperatures for the calendar day are extracted from a strip chart. Total precipitation is computed from a continous chart and checked against the totalizing precipitation gage.

Preprocessing

Relatively little preprocessing at the site is required for Level 1 (and Level 0) data.

Temperature

- Maximum: Directly measured, reported in degrees Celsius

- Minimum: Directly measured, reported in degrees Celsius

- Mean: computed as average of Maximum and Minimum

Precipitation

- Liquid: Directly measured, reported as mm of water

- Frozen: computed as water equivalent (USDC, 1989), reported as mm of water.

Return to Table of Contents

LEVEL 2 METEOROLOGY

Level 2 Meteorology entails hourly, or at least synoptic (four times daily at 0000, 0600, 1200, and 1800 hr GMT), reporting throughout a day. This reporting is necessarily automated. A day is defined as a 24-hour period from local midnight to midnight as measured by standard time in the time zone.

An established LTER site is expected to maintain at least one Level 2 meteorological station for intersite comparison and standardization purposes.

In addition to basic climatic parameters obtained at a Level 1 station, a Level 2 station obtains the more detailed meteorological data appropriate for a research site. Table 6 lists the variables to be recorded at Level 2. All variables except radiation should be recorded at least at synoptic times but preferably hourly.

Another important distinction of Level 2 meteorological systems is the capability for continuous, unattended operation, as required by the periodic (hourly or synoptic) measurements. The relatively low cost of so-called electronic data loggers makes automatic recording an especially attractive method for handling the additional recording requirements of a Level 2 station. Most LTER sites have already acheived this level of observation.

Instrumentation

Level 2 instrumentation makes periodic measurements of maximum, minimum and (separately) mean air temperature, precipitation (as water-equivalent), wind (speed and direction), relative humidty, and global solar radiation, as summarized in Table 6. The air temperature and precipitation sensors at a Level 2 station are often are more sensitive and need to be exposed in a different manner than those at a Level 1 station. Some "packaged" meteorological stations come with masts and equipment enclosures that obviate the need for the standard NWS-type shelters and other equipment described in sections 2 and 3 above.

Table 6. LEVEL 2 Meteorological Station

The nonrecording precipitation gage should be located no nearer the instrument shelter than twice the height of the instrument shelter. At exposed windswept sites, a windshield may be required for the precipitation gage. The gage must be elevated above maximum snow depth, and, if possible, operation should continue during freezing weather.

The recording precipitation gage may be either a weighing or tipping type gage. Gages should record to at least 0.5 mm (0.02inch) unless a NWS recommended Fisher Porter gage or gage from an elctronic data logger system, is used. Both standard and recording precipitation gages will be maintained at the same site. Recording gages will be impractical for some LTER sites in winter unless exposure and servicing can be provided in deep snow and the gage heated. It is a common acceptable practice to use a mixture or any combination of antifreeze, alchohol and oil, in the storage container of the raingage in order to use the instrument in the winter time. The water equivalent depth of snow will be recorded (USDC, 1989).

Both precipitation gages should not be closer to trees, buildings or the instrument shelter than twice the height of the obstruction. Standards for precipitation measurement given in for lower levels also apply.

The hygrothermograph will not be needed if an electronic relative humidity sensor is available. But note comments elsewhere concerning the calibraation of such sensors.

The anemometer should be located away from obstructions which would interfere with wind flow over the instrument. The anemometer will be mounted with the cups at 10 meters (Note this is a change from the first edition of the standards). Maintenance on the bearings and spindle will be performed twice yearly as recommended in the Observer Handbook No. 2 (USDC, 1989) or the instrument manufacturers manual. Wind travel may be accumulated by an internal counter or at a separate recorder. The optional wind direction variable will require a recording system. Level 2 stations require increased reliance upon recording instruments. The individual LTER site may elect to install a data logging system rather than separate recorders.

The global incoming radiation sensor must be fully exposed to the sky in all directions (not shaded by vegetation, buildings, or topography). An exception may be made if all of an LTER site is similarly shaded by topographic obstacles. A fully exposed sensor is preferable because the data have wider application and the effect of shading can be subtracted from full sky data. The sensor should be inspected daily, the glass kept clean, and the sensor and recorder recalibrated every 18 months.

Measurements

Level 2 measurements are to be made hourly, or at least synoptically, and reported daily. Synoptic observation times are well defined by the WMO as being 0000, 0600, 1200, annd 1800 GMT. Hourly times should always be referenced to local standard time, not daylight-savings time or other special times which may be in effect at a site.

Table 7 LEVEL 2 Meteorological Measurements

Preprocessing

Substantial preprocessing, including quality control, at the site is required for Level 2 data.

Assuming hourly data are summaries of short time interval observations, air temperature and relative hmidity data should include the instantaneous maximum and minimum and 60 minute average for each variable where available. Precipitation and solar radiation should be hourly totals. These hourly measurements are a minimum and additional parameters may need to be recorded at some sites.

Wind

The preprocessing of the wind data is complex. Observers using elctronic data processing eqipment should follow the suggestion provided in Appendix 4. Total wind travel, observed for a 24 hour period, will be converted to mean daily wind speed in meters per second. Where calm wind conditions are the rule, listing of minimum wind speed may be omitted. At sites using data loggers the proceedure for obtaining daily mean wind values is outlined in Appendix 4. At sites experiencing diurnal wind shifts, the report may list day and night means in addition to a single 24 hour mean. Wind direction may be recorded as an instantaneous observation once an hour (or as the most common direction in a five minute interval at times when the direction is highly variable) or summarized as the mean direction for each hour. As a minimum, direction will be listed for eight points plus the calm condition. Where measured, wind direction will be reported as the number of hourly observations in each of 8 directions, plus calm, for each 24 hour period, for example:

Where data loggers are used vector mean wind direction may be computed and reported following the guidlines suggested in Appendix 4. Whichever method is chosen should be noted in the metadata for the variable.

Precipitation

Because the standard precipitation gage is considered the more accurate, the recording gage values are adjusted to equal the standard gage total. Daily precipitation will be reported in millimeters for the LTER site reports. Hourly precipitation totals will be tabulated and available at each site but not included in intersite reports.

Relative Humdity

The electronic sensor or hygrothermograph will provide a continuous record of temperature and relative humidity. The hygrothermograph record should be adjusted to read within 1 deg C of the maxmin thermometers. The accuracy of the hygrograph response to relative humidity will be verified using a portable psychrometer which draws a constant air stream over wet and dry bulb thermometers. The relative humidity reading of hygrograph and psychrometer will be compared at high and low relative humidities. The use and adjustment of the hygrothermograph are discussed in Field Manual for Research in Agricultural Hydrology Chapter 3 (Brakensiek, Osborn and Rawls, 1979) which serves as the guideline document for the hygrothermograph.

Global Solar Radiation

Global solar radiation may be integrated by the data logger. Total global incoming solar radiation will be reported in MJ/sq. m/day. More on this topic is provided in Appendix 1.

Additional Considerations

Because of their complexity, sensors in Level 2 systems typically require periodic calibration. Suggested methods of calibration are provided below.

No specific Level 2 meteorological equipment is recommended by the committee since each site should be free to select its own system. However, the selected system should be able to, at a minimum:

1) make the indicated measurements at hourly (or at least synoptic) intervals,

2) record the measurements for periodic collection, either by direct media transfer or by telemetry

3) translate and relay the recorded measurements on command in suitable form to an to an external computer.

The following are some criteria that LTER sites may use for selecting automated meteorological systems:

1) Components of a system must be physically and electrically compatible. Response speed and signal level of sensor and recorder must match to avoid degrading raw data. Recorder and translator must match to avoid losing data. Unless the user is prepared to assume system design responsibility, components should be bought from a single supplier who guarantees system compatibility.

2) The recorded data should be accessible in the field for checks and calibration as well as being easily translatable to a record which can be read and processed by computer for summary reporting, and further analysis.

3) The system should be able to operate during expected environmental conditions. Estimate climatic extremes at your site and specify that the equipment will operate within these limits. Components that seem particularly susceptible to cold (less than minus 10 deg.C) and moisture (RH greater than 90%) conditions include hard copy paper printers and cassette recorders. Modems are more reliable if a phone line can reach a station. Cellular phones and telemetering systems have also proved reliable in extreme conditions.

4) Be sure that the manufacturer can provide fast service and backup for your equipment and/or have two or more compatible systems to interchange components. If components are interchanged make sure intercalibration factors are available. Budget for repairs and recalibrations.

Although the LTER program does not endorse any particular manufacturer of electronic data sensing and recording instrumentation, it is noted that a large number of LTER sites use equipment made by Campbell Scientific inc, of Logan Utah; accordingly, Apppendix 2 includes a typical configuration from this manufacturer. Other manufacturers may be able to provide similar, acceptable systems

All original data records (tapes, charts, etc) should be kept, and the earliest listing of raw electronic data should become part of the permanent record for the site.

Return to Table of Contents

LEVEL 3 METEOROLOGY

Level 3 includes a number of variables which we beleive it would be optimal for sites to record but for which funds may not be available and for which funding priorities must be set.

These variables include: PAR, APAR, Soil temperature, Soil moisture, [for soil moisture and temp data take guidance from new LTER book on soil mesurements - Phil Roberstson (KBS) Editor] Atmospheric pressure, Vapor pressure.

Instrumentation

PAR and APAR are two variables commonly used in ecological models. The instruments are essentially radiometers.

The measurement of soil temperature and moisture will be described by an LTER publication on soil measurement by Robertson et al. (Details of the exact citation will be added when they are available). Soil temperature should be measured at depths copatible with the Robertson work

Atmospheric pressure can be measured by electronic sensors which will attach to a data logger. Vapor pressure may be derived using tables (e.g. Marvin, 1941) and values of the air temperature, relative humidity, and atmospheric presure.

Measurements

Photosynthetically active radiation (PAR) and Absorbed Photosynthetically active radiation (APAR) should be measured in a similar way to radiation. electronic integrators can accumulate the energy input and display daily or hourly period totals or provide input to a data logger. We reccomend 15 minute integrations of these variables.

Soil moisture and temperature should be measured hourly but this instruction may be changed pending the Robertson LTER book on soil measurement procedures.

Vapor pressure may be derived at hourly intervals using hourly measurements of relative humidity and temperature. A single instantaneous observation of air pressure will be appropriate for this calculation. The computation is explained in Appendix 3

Return to Table of Contents

LEVEL 4 METEOROLOGY

General

Level 4 meteorology concerns additional specialized measurements directly related to local site specific research needs. Because a wide variety of such needs exist, the topic can be treated in only a general manner here. Persons interested undertaking specialized studies should contact investigators already performimg them to obtain advice and coordinate methodology. Climate Committee members may be able to help with advice on investigators to contact.

If funds are available we recommend that sites consider making obervations of wet deposition, atmospheric transmissivity and Ulltra Violet B (UVB) radiation. UVB measurements may be important in detecting long-term trends in this variable. Such changes might be expected in association with changes in the amount of stratospheric ozone and excessive UVB can have potentially harmful effects on plants and animals If sites are interested in this or other questions related to UVB then it should be measured. However this is an expensive task. Broad band UVB measuring systems currently cost about $5000 while multispectral instrument systems may cost $25,000. Persons interested in such measurements will get useful advice from the USDA UVB Monitoring Program at http://uvb.nrel.colostate.edu/UVB/

Sun photometer observations. The National Aeronautical and Space Administration (NASA) has developed an instrument called a sun photometer. This instrument tracks the sun during the day and measures the components of the transmissivity of the atmosphere. The measurements are designed to be used to give greater accuracy in the interpretation of satellite imagery. The sun photometer measurements are potentially useful in obtaining some elements of the surface radiation balance.

Wet deposition observations should be patterned after the wet deposition measurements made in the National Atmospheric Deposition Program (NADP). The NADP program uses an instrument to measure separately both wet and dry deposition. We recommend only the former be used since the latter may be unreliable. Wet deposition is distinguished from bulk deposition because the latter measures both wet and dry deposition in the same receptacle.

The National Atmospheric Deposition Program (NADP)

The National Atmospheric Deposition Program is a cooperative research program of the state agricultural experiment stations and other federal, state, and private research organizations. Its aim is to determine both the composition and amount of atmospheric deposition and its distribution on a national scale in order to assess the magnitude of the effects... (NADP, 1984). The NADP program is now called the National Atmospheric Deposition Program / National Trends Network (NADP/NTN). Look for more information at http://nadp.sws.uiuc.edu and see also http://btdqs.usgs.gov/acidrain/ .

Some LTER sites already participate in the NADP. The LTER climate committee endorses all the practices already in place for the standardization of NADP measurements. These practices are reported in the documents NADP Site Selection and Installation Manual, NADP Instruction Manual On Site Operations, and Field Operations Manual.

The NADP is now closed to the establishment of any new sites. Any LTER sites that are interested in making atmospheric deposition measurements should use the same sample collectors (Aerochemetrics Model 201) used in the NADP program. These are available from Aerochemetrics Ltd. 6832 SW 81 Street, Miami, Florida 33143. Investigators should also follow, as closely as possible, the same laboratory analytical methods used in the NADP program. Details of these may be obtained from the NADP documents NADP Quality Assurance Plan : Deposition Monitoring and CAL Analytical Methods Manual. Copies are available from Dr. Van Bowersox, Illinois Water Survey, 2204 Griffith Drive, Champaign, IL 61820 which is the location of the Central Analytical Laboratory (CAL) of the NADP.

Standardization of Specialized Measurements.

Owing to the large variety of specialized measurements that might be undertaken at LTER sites the Climatology Committee does not specify procedures but does make two recommendations:

First, where any specific future intersite study is anticipated the investigators are strongly urged to plan the experiment in such a way that instrumentation and methods are identical at the sites involved. The climate committee would be able to give advice on experimental design.

Second, investigators using specialized meteorological measurements must pay special attention to reporting the accuracy and precision of their observations. This will enable future studies to determine whether intersite comparisons fall within or outside of measurement error. Statements of accuracy and precision should address all parts of the following definitions which have been provided by the National Atmospheric Deposition Program (NADP, 1984. p38):

Accuracy: A measure of the degree of conformity of the mean value, obtained by using a specific method or procedure, with the true value. The concept of accuracy includes both bias (systematic error) and precision (random error).

Precision: The degree of agreement of repeated measurements of a homogeneous sample by a specific procedure, expressed in terms of dispersion of the individual values about the mean value.

The LTER Climate Committee endorses the precision and accuracy values suggested by the North Central Region of the Agricultural Climate Committee and advises LTER sites to attempt to meet these standards. They are as follows:

Table 7. Precision and Accuracy of Measurements.

Return to Table of Contents

CALIBRATION

In order to maintain a high degree of accuracy in the measurements of climate parameters it is essential that instruments be checked and calibrated on a regular basis. The frequency and need for calibration depend upon the variable being measured and the type of sensor and method of recording being used. The use of electronic data loggers that can run for weeks or months without needing any technician attention make it especially important that a regular schedule of checking data and a set calibration schedule be followed. We recommend the following procedures for the various levels of measurement.

Level 0

Air temperature

It is expected that at this level mercury and alcohol glass thermometers will be used. These thermometers are calibrated and tested by the manufacturer and have proved to be reliable over long periods of time. They should be used for checking and calibration of both chart and electronically gathered data at higher levels.

Precipitation

Standard rain gages that are measured manually with a measuring stick are calibrated by the manufacturer and need no further calibration.

Level 1

Air temperature

At this level it is expected that air temperature will be recorded by either a thermograph or by an electronic data logger.

Thermograph:

For the thermograph chart temperature should be checked with a glass thermometer at each visit to the station and the maximum and minimum temperatures should be checked between chart and thermometer readings whenever the chart is replaced. When necessary adjustments can be made manually to the thermograph.

Data logger:

The data logger temperature reading should be checked against the mercury or alcohol thermometers at each visit to the site. Any sensor that deviates from the thermometer reading should be replaced and sent back to the manufacturer for calibration.

Precipitation:

It is expected that a manual precipitation gage will be used at this level (see level 0)

Level 2

Temperature: Same as for level 1

Precipitation:

At this level it is expected that precipitation will be recorded with a data logger with either a tipping bucket (using an event recorder) or by a weighing bucket (using a pressure transducer) precipitation gage.

Tipping bucket:

Manufacturer’s instructions for calibrating this instrument should be followed before this instrument is placed in the field. The instrument should be recalibrated on a seasonal or more frequent schedule.

As an additional check weekly or monthly cumulative amounts should be checked against a manual precipitation gage at the site.

Weighing bucket:

A series of calibration weights and the procedure for calibrating this instrument are provided by the manufacturer. The instrument should be calibrated seasonally.

As an additional check this instrument should also be checked against a manually read precipitation gage on a regular (weekly or monthly) basis.

Relative Humidity:

It is assumed that at this level RH will be recorded electronically, usually in a unit that includes both temperature and relative humidity sensors.

Relative humidity readings should be checked on a regular basis using an aspirated pschycrometer. Replacement units for the RH chip are usually available from the manufacturer.

Windspeed:

Windspeed calibration is made by the manufacturer. Manufacturer’s instruction for recalibration frequency and bearing replacement should be followed. Where available windspeed can be checked locally in a wind tunnel.

Wind direction:

Initially wind direction must be established using a compass and following the manufacturer’s instructions. Wind direction should be checked on a regular schedule by comparing instant logger readings with the actual direction shown by the sensor’s vane.

Solar radiation:

It is expected that a LI-COR pyranometer or similar sensor will be used at this level for recording global radiation. Each sensor is given a calibration factor by the manufacturer which needs to be programmed in to the data logger. Manufacturers recommend recalibration on an annual or biannual schedule. it is important that this recommended schedule be followed as these sensors do need frequent recalibration. This calibration can usually be provided by the manufacturer.

Level 3 and Level 4

The specialized sensors used at these more complex levels of climate parameter recordings will all need some level of calibration on a wide variety of schedules A few examples are given below.

Wet deposition:

The NADP (National Atmospheric Deposition Program) has a rigorous program for calibration of instruments and sample analysis. Sites recording wet deposition on their own should try to follow the NADP standards.

Sun Photometer:

Sites participating in the NASA sun photometer program should have the instruments calibrated by NASA on a regular schedule provided by NASA.

Soil Temperature sensors:

Soil temperature sensors should be calibrated against known temperatures before being placed in the soil. It is recommended that these sensors be tested again when they are removed from the soil. Experience has shown that these sensors may fail or drift significantly over long periods. The data should be carefully monitored and any irregularity noted. A regular schedule of tasting and replacement of the soil temperature sensors should be established for long term monitoring sites. [In the future this information will be made compatible with the Robertson et al. book on soil observation processes].

Return to Table of Contents

LITERATURE CITED

Baker, D.G., 1975. Effects of Observation Time on Mean Temperature Estimation. Journal of Applied Meteorology. 14:471476.

Brakensiek, D.L., Osborn, H.B., and Rawls, W.J., 1979. Field Manual for Research in Agricultural Hydrology. US Dept. Agri. Handbook 224, 547 pp.

Buck, A.L., 1981 New equations for computing vapor pressure and enhancement factor. Journal of Applied Meteorology 20:1527 1532.

Karl, T.R., C. N. Williams Jr., and P. J. Young. 1986. A model to estimate the time of observation bais associated with mean monthly maximum, mininum, and mean tempeartures for the United States. Journal of Climate and Applied Meteorology. 25:145-160.

Marvin, C. F. 1941. Psychrometric Tables for obtaining the vapor pressure, relative humidity, and temperature of the dew point. U. S. Department of Commerce. Weather Bureau. U. S. Government Printing Office. Washington. DC. 87 pp.

NADP. 1984. NADP Quality Assurance Plan: Deposition Monitoring. NADP Quality Assurance Steering Committee. Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, Colo. 39 pp.

Swift, L.W. Jr., and Ragsdale, H.L. 1985. Meteorological Data Stations at LongTerm Ecological Research Sites. Proceedings of Forest Environmental Measurements Conference; 1983 October 2328; Oak Ridge, TN. Reidel, Holland. pp. 2537.

TIE, 1979. LongTerm Ecological Research Concept Statement and Measurement Needs. Summary of a Workshop. Sponsored by National Science Foundation. Grant DEB 792043. Indianapolis, Indiana. June 2527, 1979.49 pp.

USDC, 1989. Cooperative Station Observations, Weather Bureau Observing Handbook No. 2. National Oceanic and Atmospheric Administration, US Dept. of Commerce, 83 pp.

Waring, R.H., Holbo, H.R., Bueb, R.P., and Fredriksen, R.L. 1978. Documentation of Meteorological Data from the Coniferous Forest Biome Primary Station in Oregon. General Technical Report PNW73. Pacific Northwest Forest and Range Experiment Station. U.S. Department of Agriculture. Portland, Oregon. 23pp.

WMO, 1971. Guide to Meteorological Instrument and Observing Practices, 4th Edition, WMONo. 8, TP3. World Meteorological Organization, Geneva.

WMO, 1970. Guide to Hydrometeorological Practices, 2nd Edition, WMONo. 168. TP.82. World Meteorological Organization, Geneva.

Other Useful and Related References:

Allen, R.G., Howell, T. A., Pruitt, W.O., Walter, I.A., and Jensen, M.E. (eds) 1991. Lysimeters for evapotranspiration and environmental measurements. American Society of Civil Engineers. New York. NY 10010-2398. 4

Fitter, A. H., and Hay. R.K.M. 1987. Environmental Physiology of Plants. 2nd Ed. Academic Press. London. .

Gates, D.M. 1993. Climate Change and its Biological Consequences. Sinauer Associates. Sunderland. Mass.

Goel, N. S. and J. M. Norman. 1990. Instrumentation for Studying Canopies for Remote Sensing in Optical and Thermal Infrared Regions. Remote Sensing Reviews. 5:360 pp.

Griffiths, J. F. (Ed.)1994. Handbook of Agricultural Climatology. Oxford University Press. New York and Oxford. 320 pp.

Monteith, J.L. 1973. Principles of environmental physics. Edward Arnold. London.

Oke, T.R. 1987. Boundary Layer Climatology. 2nd ed. Methuen. London.

Rosenberg, N.J., Blad, B.L., and Verma, S. 1983. Microlimate:the biological environment. 2nd ed. Wiley. New York.

Tuhkanen, S. 1980. Climatic parameters and indices in plant geography. Acta Phytogeographica Suecica. 67. Svenska Vaxgeografiska Sallskapet. Uppsala 105. pp.

WMO/UNEP 1997. Global Climate Observing System. GCOS/GTOS Plan for Terrestrial Climate-Related Observations. Version 2.0. WMO/TD-No. 796. UNEP/DEIA/TR.97-7. 130 pp.

Return to Table of Contents

Appendix 1 - Measurement of Solar Radiation

The Climate Committee does not advocate the use of any particular sensor for the measurement of solar radiation. Solar radiation is considered here to be both direct and diffuse radiation originating from the sun and having wavelengths between 0.15 and 3.0 micrometers. This radiation is commonly called shortwave radiation as oppose to longwave radiation (or infrared radiation 3.0 to 100 micrometers) which is emitted by the Earth and the atmosphere. We present examples of the types of instrument which are used at LTER site. Some sites use a LI-COR Silicon Pyranometer for solar radiation measurement. Other sites use Kipp and Zonen radiometers with a CM5 thermopile or a CM3 pyranometer while still other sites use radiometers from the Eppley Company. There are also other manufacturers of solar radiometers. As in other situations the more expensive equipment generally provides greater accuracy. Our experience has shown that most of the radiometers mentioned above are reliable. Users should pay attention to the need for careful siting (preferably with no obstruction of the horizon) and leveling of the sensors and make every effort to keep the surface of the sensor free from dust, ice and other forms of water. We choose not to suggest the measurement of other components of the radiation balance at level 2, such as longwave radiation flows, because such measurements are either expensive or labor intensive or both.

Return to Table of Contents

Appendix 2 - Components of a Typical Level 2 Meteorological Site

The following is one example of the basic components of a data logger system from Campbell Scientific, Inc. and their approximate 1997 costs that would comprise an electronic data sensing and logging system acceptable for level 2 meteorology. The list is given as an example only. It includes some items that will not be needed at certain sites and does not include extra items that may be needed to meet the specialized needs of other sites.

Return to Table of Contents

Appendix 3 - Estimationof Vapor Pressure

Actual vapor pressure in millibars may be calculated for each of the two daily observation points as:

Actual vapor pressure = {Relative humidity/100}*saturation vapor pressure at air temperature.

The latter may be obtained from tables (Marvin, 1941).

or

Actual vp = saturation vapor pressure at dewpoint temperature, where saturation vapor pressure in millibars:

e = 6.1121 exp (17.368 T / (238.88 + T))

Return to Table of Contents

Appendix 4. Measurement of Wind Speed and Direction

Wind Speed and Wind Direction [txt with sub&superscripts in same line]

Wind speed is sensed by an anemometer and customarily measured by counting pulses representing total wind travel during the sampling scan interval. Thus, speed is total travel over scan duration and is a mean, not instantaneous value, for that interval of time.

si = wind speed [m/s] = wind travel [distance] / sampling interval [time]

The scalar mean wind speed (i.e., without regard to direction), Smean, produced by a data logger at the end of the output period is:

Smean = sum(si / N) (1)

where N = number of sampling intervals in an output period. Note that any maximum wind speed selected by a data logger from all si in an output period is NOT an instantaneous peak wind speed but an average for the duration of the scan interval. If peak winds are an important parameter to be measured, then the logger should be programmed for short scan intervals, i.e. an interval less than typical wind gust duration. The maximum sampling rate of the data logger or the response time of the anemometer will determine the shortest sampling interval over which speed can be determined.

Anemometers are available which produce a continuous voltage signal proportional to wind speed. Cost and maintenance of sensitive models of these instruments may limit their use in climate stations. Valid peak wind speed can be defined by this type of anemometer if the signal is recorded continuously or sampled frequently.

Wind direction is customarily measured by a wind vane which is decoded by the logger into azimuth (az) from north so that N is 0 degrees (or 360 degrees), E is 90 degrees, S is 180 degrees, and W is 270 degrees.

Alternatively, wind speed and direction can be sensed by a two-propeller anemometer, usually facing North and East, measuring two components of wind. The resultant wind speed is:

si = (ni2 + ei2)1/2 (2)

where ni and ei are the instantaneous north and east components of wind speed (if ni or ei are negative they reprsent wind flux from the opposite direction, south or west). If this calculation is done at scan time, the scalar mean wind speed for a logger output period is as in equ. (1). The instantaneous wind direction from a two-propeller anemometer is:

azi = arctan(ei / ni) (3)

Obviously, ni and ei must be checked for the zero condition. Note that a simple mean of wind directions will yield invalid results. As an extreme example, the mean of four observations of wind directions near North at 357 degrees, 3 degrees, 358 degrees, and 2 degrees would compute as South or 180 degrees. The best way to obtain mean wind direction is therefore to compute the mean wind direction from the sums of the components:

Azmean = arctan(sum[ei] / sum[ni]) (4)

Some data loggers have the capability to compute vector mean wind velocity and vector mean wind direction over an output period from the scan interval observations of wind speed and direction. In addition to the scalar mean wind speed, the logger internally calculates the vector mean wind velocity from computed orthogonal components of the observed wind vectors. The vector mean wind velocity is always less than or equal to the scalar mean wind speed. The orthogonal components are calculated by the logger using equ. (5) and (6) from the scan interval observations of speed and direction. The logger averages direction over the output period as in equ. (4) and averages velocity as in equ. (7).

If the data logger does not have a routine for averaging vectors from anemometer and vane sensors, the orthogonal components for each scan interval may be calculated as:

ni = si cos(azi) and (5)

ei = si sin(azi) . (6)

The vector mean wind velocity would be computed for an output period, similar to equ. (2):

Vmean = (sum[ni2] + sum[ei2])1/2 (7)

and the vector mean wind direction computed by equ. (4).

If a data logger is not used or one is used that can not compute wind speed and direction vectors, mean wind direction might be approximated by recording the prevailing wind direction at observation time or by recording the wind direction at the time of maximum wind speed within each sampling period.

Wind direction is meaningless during calm periods, i. e. when wind speed is recorded as zero or within the sensitivity error band of the anemometer. For data systems using the wind vector calculation, the calm period wind directions are reduced to insignificant fractions of the output period vector mean direction. However, the negligible wind speeds for calm periods should and do contribute to both the scalar mean wind speed and vector mean wind velocity.

A useful option available with some data loggers is to output the standard deviation of the wind directions observed over the sampling output period. A climatic station representing a site or time period characterized by variable and non-persistent wind directions will record a larger standard deviation than the case where wind blows from a consistent direction. Thus, the standard deviation provides another quantifiable measure of the climate characteristics of a site in addition to its statistical values.

Aggregation rules:

When wind speed is aggregated to represent longer time periods, e. g. averaging hourly into daily periods or daily into monthly means, computing the mean of the scalar mean wind speeds will be valid. However, to average the wind vector, all vector mean wind velocity and direction pairs must be converted to rectangular coordinates using calculations similar to equ.(5) and (6) and the longer time period vector means of velocity and direction computed from the orthogonal values similar to equ. (7) and (4). This is the only valid method to compute the vector means. The daily mean of the hourly standard deviations would be the square root of the mean of the hourly deviations squared:

SDmean = (sum[SD2] / N)1/2 (8)

Return to Table of Contents