Potter T.D., Colman B.R. (co-chief editors). The handbook of weather, climate, and water: dynamics, climate physical meteorology, weather systems, and measurements
Подождите немного. Документ загружается.
a broad community of users, primarily through a computerized bulletin board
system.
In 2001, St. John’s water management district in south Florida installed an even
faster radio telemetry network of 85 sites with multiple repeaters. The 85 sites are
polled every 15 min. The data are used to monitor levels and control gates and
pumps thus controlling surface water levels within an acceptable range. The benefit
of increased telemetry speed is real-time monitoring and control.
Spread-Spectrum RF
Spread-spectrum radio modems look and act like a wireless RS232 port. Constrained
by the FCC to less than 1 W effective radiated power, an FCC license is not required
to operate them. Transmitting and receiving on mul tiple frequencies in one of several
bands (902 to 928 MHz was the first), they work well in industrial areas where line-
of-sight is not available but distances are less than 5 km. With line-of-sight and yagi
antennas, distances up to 30 km are possible. Spread-spectrum radios utilize one of
two methods to transmit their data: frequency hopping or direct sequencing. The
frequency hopping method is preferred for stationary sites. It also allows
the frequency to be shared by more sites. Direct sequencing is preferred for
mobile sites. Current spread-spectrum radios require 100 mA in the receive mode
and 600 mA in the transmit mode. A spread-spectrum radio under design at CSI
should require less than 75 mA in the transmit mode.
Satellite
Satellite transmitters provide data transfer from very remote sites and networks with
wide spatial coverage. Most satellite links only provide one-way data transfer. Data
transfer via satellite is generally more costly due to the required infrastructure. At the
end of 1998, approximately 1300 satellites orbited Earth. Estimates put their number
at 2400 by the year 2008. Satellites are grouped into three main classes based on
height and type of orbit. They are: geosynchronous or geostationary Earth orbiting
satellites (GEOs), middle Earth orbit satellites (MEOs), and Low Earth orbiting
satellites (LEOs). The height of their orbits also determines the maximum geogra-
phical area they can cover.
Geosynchronous or Geostationary Earth Orbiting Satellites GEOs,
positioned 22,300 miles (36,000 km) above the equator, appear to be stationary in
the sky as they turn synchronously with Earth (hence the name geosynchronous or
geostationar y). This allows a ground station antenna to point at one place in the sky
to send and receive signals. GEOs are the farthest from Earth and cover the largest
surface area. Three GEOs are needed to cover the enti re planet below 70
latitude.
The Geostationary Operational Environmental Satellite (GOES) system was
installed and is maintained by the U.S. government. GOES channels (frequency
and time slice) are reserved for use by federal, state, and local U.S. governments
796 COMMERCIAL RESPONSE TO MEASUREMENT SYSTEM DESIGN
and some foreign governments. The recently introduced high data rate (HDR) GOES
data throughput is limited to:
300- or 1200-bit per second data rate
20-s transmission window
Transmissions inter vals of 1, 3, or 4 h
The highest elevation weather station in the Americas was installed at 6542 m on
the summit of Nevado Sajama in Bolivia in 1996 (Hardy et al., 1998). Installed to
better understand tropical ice core variations and global climatic change, the station
utilizes a GOES transmitter for near-real-time data and memory modules for on-site
storage and manual data retrieval. The remote, high-elevation location precludes
unscheduled ser vice or repair visits. Snow and ice buildup on the antenna reduced
near-real-time data recovery to 80% during the wettest month, but 100% of the on-
site data was recovered during the annual site visit. In addition to permitting the start
of the data analysis, the near-real-time data indicated that the failure of the lower
sensors midyear was due to their burial by an El Nin
˜
o induced larger than expected
snowfall.
Middle Earth Orbit Satellites MEOs, positioned between GEOs and LEOs,
orbit from 1000 to 22,300 miles. Depending on the altitude, as many as 10 or 12 are
needed to cover the entire planet. In addition to voice and data transmission, MEOs
are often used by surface navigation systems such as the Global Positioning System
(GPS).
Many environmental measurement applications benefit from being able to record
GPS position data. In addition, the GOES system now requires that transmitters use
GPS clock data to keep their transmissions accurately timed.
Low Earth Orbiting Satellites LEOs orbit at altitudes from 100 to 1000 miles.
Approximately 46 to 66 LEOs are needed to cover the entire planet. A LEO orbiting
satellite is above the local horizon for less than 20 min. Messages must be short or
LEO systems must support satellite-to-satellite hand-off to maintain communica-
tions.
One of the LEO systems called ARGOS, utilize the National Oceanic and Atmo-
spheric Administration’s (NOAA’s) polar orbiting satellites. These are frequently
used either in oceanography because station position can be tracked, or at latitudes
greater than 70
where GOES satellites cannot be seen. The ARGOS platform
provides good data transfer at the poles because the satellites cross the poles 24
times per day, which is three times more often than at the equator. ARGOS data
throughput is limited by:
The number of times per day the satellites pass overhead
One satellite pass overhead lasts from 10 to 14 min
4 DATA RETRIEVAL 797
Redundant 32-byte messages are transmitted every 90 or 200 s to ensure data
integrity
400-bit per second data rate
Satellite communication is changing rapidly. Several projects currently underway
will be launched in the near future. The satellite business is not easy due to the
numerous governments involved and the fight for frequencies. Many projects have
merged together so that the merged company has a better chance of seeing its project
to completion. The following is an incomplete list of websites that provide links to
companies that offer data transmission services via one of the GEO, MEO, or LEO
satellites:
GEO Systems Website
GOES http:==NOAA.WFF.NASA.GOV
Inmarsat-C http:==217.204.152.210=index.cfm
Qualcomm http:==www.qualcomm.com
MEO Systems Website
ICO http:==www.ico.com
ORBLINK http:==www.orbital.com
LEO Systems Website
Iridium http:==www.iridium.com
Orbcomm http:==www.orbcomm.com
Globalstar http:==www.globalstar.com
Argos http:==www.argosinc.com
5 SUMMARY
Commercial measurement system design requires careful attention to all detai ls.
Only the most important details were presented here. The experience gained through
the design of similar measurement systems that have been produced and supported
in quantity is valuable in the design of a new system. It is important to quickly utilize
the technological advances in electrical components, sensors, and telecommunica-
tion methods. There is an optimum high level of quality at which the total costs of
designing, producing, selling, and servicing a product are minimized. Careful, high-
quality records help maintain high-quality standards. Low turnover of key personnel
is important to keep expertise and continuity. Finally, the personal joy and satisfac-
tion that comes from service to customers with legitimate needs is the basis for our
existence as an organization (Campbell, 1987).
798 COMMERCIAL RESPONSE TO MEASUREMENT SYSTEM DESIGN
REFERENCES
Brock, F. V., and K. C. Crawford (1995). The Oklahoma mesonet: A technical overview,
J. Atmos. Oceanic Tech. 12, 5–19.
Campbell, E. C. (1987). Overview. ‘‘Good Morning, This is Campbell Scientific ...’’ (in-house
newsletter), March, 1.
Campbell, E. C. (1991). Overview, Independent Verification. The Campbell Update, A
Newsletter for the Customers of Campbell Scientific, Inc., 2(1), 2.
Hardy, D. R., M. Vuille, C. Braun, F. Keimig, and R. S. Bradley (1998). Annual and daily
meteorological cycles at high altitude on a tropical mountain, Bull. Am. Meteorolog. Soc.
79, 1899–1913.
Schimmelpfennig, H. G., B. D. Tanner, and E. C. Campbell (1979). Applications of a Minature,
Low Power, Computing Datalogger in Environmental Investigations. Fourteenth Confer-
ence on Agriculture & Forest Meteorology and Fourth Conference on Biometeorology,
April 2–6, 1979, Minneapolis, MN, American Meteorological Society, Preprint Volume,
pp 154–155.
Tanner, B. D. (1990). Automated weather stations, Remote Sensing Rev. 5(1), 73–98.
REFERENCES 799
CHAPTER 40
DESIGN, CALIBRATION, AND QUALITY
ASSURANCE NEEDS OF NETWORKS
SCOTT J. RICHARDSON AND FRED V. BROCK
1 INTRODUCTION
The calibration and quality assurance needs of meteorological networks vary signif-
icantly depending on factors such as end-user needs, data accuracy and resolution
requirements, site maintenance schedule, climatology, and, of course, the budget of
the project. This chapter descri bes general principles that can be used to ensure
quality data in meteorological networks.
The design of a meteorological observation network depends on many factors,
perhaps most importantly what the system is to meas ure and with what accuracy.
However, design of a measurement system is also powerfully affected by other
considerations such as choice of sensor and data logger. Selection of the measure-
ment platform, data communication system, and type of power system has a
profound affect on overall system design. Communication system limitations may
dictate the location of remote sites forcing compromises in site location. Power
limitations may prohibit the use of certain types of sensors.
A very important and sometimes overlooked aspect of system design is future
data requirements and network upgrades. A system designed to meet today’s require-
ments will not necessarily satisfy tomorrow’s needs and=or may not work well with
new technology. A data collection system should be designed with this in mind so
that upgrades can be made without the need to replace the entire infrastructure of the
network. Note that once a data collection system and=or vendor is chosen, it may be
necessary to use this same system for the duration of the project including upgrades
or modifications. This is because, for example, once a data logger is chosen it may
Handbook of Weather, Climate, and Water: Dynamics, Climate, Physical Meteorology, Weather Systems,
and Measurements, Edited by Thomas D. Potter and Bradley R. Colman.
ISBN 0-471-21490-6 # 2003 John Wiley & Sons, Inc.
801
not be possible or feasible to switch to a different vendor or system because of
compatibility problems. This is not necessarily bad but should be considered when
choosing a manufacturer. For example, one should consider if the manufacturer will
be in business when network upgrades are required and, if not, how much of the
system will require replacement.
The text draws from the experiences of the Oklahoma Mesonet (Brock et al.,
1995) as well as the Atmospheric Radiation Measurement (ARM) Program (Stokes
and Schwartz, 1994; DOE, 1990); a brief descri ption of both will be given to put the
subsequent writing in context. More details on network design and instrument
performance can be found in Meteorological Measurement Systems by Brock and
Richardson (1991). An Introduction to Meteorological Instrumentation and
Measurement by T. P. DeFelice (1998) is also a good reference.
The Oklahoma Mesonet
The Oklahoma Mesonet is a system of 115 remote surface observing stations across
the state of Oklahoma and is a joint effort between Oklahoma State University
and the University of Oklahoma. Parameters meas ured at all 115 sites include
pressure, wind speed and direction at 10 m above ground level (agl), air temperature
and relative humidity at 1.5 m agl, rainfall, global solar radiation, and ground
temperature at 10 cm under bare soil and native ground cover. Additional parameters
measured at about half the sites include wind speed at 2 m agl, air temperature at 9 m
agl, soil moisture at 5, 25, 60, and 75 cm below ground, and additional soil tempera-
ture measurements down to 30 cm. Surface flux measurement capabilities are being
added to the Oklahoma Mesonet through the OASIS (Oklahoma Atmospheric
Surface-layer Instrumentation System) Project (Brotzge et al., 1999). Once
completed in 2000, Okl ahoma Mesonet sites will measure net radiation, ground
heat flux, and sensible flux, and latent heat flux using a profile technique (at 90
sites) and eddy correlation techniques (at 9 sites).
Five-minute averaged data are collected from all 115 stations every 15 min and
transmitted through the Oklahoma Law Enforcement Telecommunication System
(OLETS) to an archival system at the University of Oklahoma. All stations are
solar-powered with battery backup; site maintenance is typically performed 3 to 4
times per year. The system was commissioned in 1994 and each year more than
99.9% of possible observations (approximately 15,537,000 of a possible 15,547,000
observations) are archived. The use of OLETS was important to the success of the
Oklahoma Mesonet, since it and allowed data collection and two-way communica-
tion between the base station and all remote sites without the need for expensive
phone lines.
Atmospheric Radiation Measurement Program
The ARM Program is a multilaboratory, interagency program that was created in
1989 with funding from the U.S. Department of Energy (DOE). The ARM Program
is part of the DOE effort to resolve scientific uncertainties about global climate
802 DESIGN, CALIBRATION, AND QUALITY ASSURANCE NEEDS OF NETWORKS
change with a specific focus on improving the performance of general circulation
models (GCMs) used for climate research and prediction. These improved models
will help scientists better understand the influences of human activities on Earth’s
climate.
In pursuit of its goal, the ARM Program establishes and operates field research
sites, called Cloud and Radiation Testbeds (CARTs), in several climatically signifi-
cant locations (the north slope of Alaska, the tropical western Pacific, and the U. S.
Southern Great Plains). Data are collected over extended periods of time (years)
from large arrays of instruments (both state-of-the-science and conventional instru-
mentation are used) to study the effects and interactions of sunlight, radiant energy,
and clouds on temperatures, weather, and climate. Specifically, ARM focuses on
cloud–radiation interactions.
The ARM Program has taken advantage of the advances in communications and
the World Wide Web. Data are ingested and available in near real time via the Web,
trouble reports are submitted via the Web, complete site and instrument history
information is archived and available via the Web, etc.
This chapter will emphasize design, calibration, and quality assurance needs of
smaller networks such as the Oklahoma Mesonet or smaller.
2 SYSTEM DESIGN
Sensors are typically mounted on a stationary platform (a simple mast or tall tower)
or on a moving platform (balloons, planes , ships, etc.). Ideally, data are commu-
nicated in real time from the measurement site or platform to a central archiving
facility. In some cases, real-time communication is not possible but, instead, data are
manually collected at periodic intervals, usually in some electronic form. Availability
of electrical power, or the lack of it, may seriously affect the system desig n.
Instrument Platforms
It is not surprising that virtual ly every type of instrument platform is used in
meteorology because the atmosphere is so extensive and because most of it is
quite inaccessible. These platforms include masts, instrument shelters, tall towers,
balloons, kites, cars, ships, buoys, airplanes, rockets, and satellites. Synoptic data
platforms include balloons and satellites supplemented by buoys and ships over the
ocean. In addition, aircraft are used for hurricane observation and some data are
collected from commercial flights to fill in gaps in the observation networks. Aircraft
are extensively used for research investigations around thunderstorms or wherever
high-density upper air data are needed.
When selecting a plat form, consideration should be given to where the measure-
ment is to be made and whether the platform can be permanently fixed or is moving,
cost, and exposure. To some extent, any platfor m, even a simple tower for surface
measurements, interacts with the atmosphere and affects instrument exposure.
A simple 10-m tower, shown in Figure 1, has a wind sensor at 10 m and temperature
2 SYSTEM DESIGN 803
and a relative humidity (T&RH) sensor at 1.5 m in addition to a radio antenna for
data transmission, a solar panel and battery for power, a barometer, and a data
logger. These sensors must be mounted with due consideration for exposure to
prevailing winds to minimize tower effects.
Communication Systems
A comm unications network is a vital part of almost every meteorological measure-
ment system at all scales. Historically, meteorological communications have relied
primarily on land-line and radio links. More recently, polar orbiting and geostation-
ary satellites are used for data communications in macroscale or synoptic measure-
ment systems and even in many mesoscale systems. Commercial satellites are used
to broadcast data from central points, with sophisticated uplinks, to users equipped
with fairly simple antennas and receivers (inexpensive downlinks).
The ideal co mmunications system would reliably transmit data from the remote
instrument platform to a central facility and in the reverse direction with little or no
time delay, without limiting the volume of data to be transmitted. Communication
both to and from the remo te site is used to synchronize local clocks in the data
loggers, to load operating programs into the data loggers, and to make special
data requests, to name a few. Two-way communications is not essential but highly
desirable.
Figure 1 A 10-m tower for surface measurements. At the top of the tower is a propeller-vane
anemometer that measures wind speed and direction. Below the anemometer is an antenna for
two-way data communications, and power is provided by a solar panel. The pyranometer is
mounted to the south where tower and guy-wire shadows will not affect the data.
804
DESIGN, CALIBRATION, AND QUALITY ASSURANCE NEEDS OF NETWORKS
Telephone Commercial telephone systems provide adequate signal bandwidth,
are generally reliable, and cover most land areas. The cost is prohibitive if one must
pay for running lines to each station, especially for a short-term project. Even for
long-term projects like the Oklahoma Mesonet and the ARM Program, phone lines
were either avoided entirely or used very sparingly due to the expense involved.
Recent advances in cellular telephone technology coupled with decreasing
airtime charges means cellular data communications has become a viable alternative
to traditional phone lines.
Direct Radio Direct radio links from the remote stations to a central base station
are desirable because they offer flexibility but Earth curvature limits line-of-sight
links. Figure 2 shows the maximum line-of-sight distance between two stations if the
remote station antenna is at a height of 10 m. For a base station or repeater antenna
height of 200 m, the line-of-sight link is only a little more than 60 km. Direct radio,
even when augmented with repeaters, severely limits the size of a network and
causes immense difficulties in complex terrain. For example, if the path of the signal
from a remote station to the repeater or to the base station is too close to the ground,
the signal could be trapped in an inversion layer and ducted away from the intended
destination, thereby losing the connection between the sensor and the base station.
Satellite The first communications satellite that permitted an inexpensive uplink
(low-power transmitter and simple antenna) was the Geostationary Operational
Environmental Satellite (GOES). An inexpensive uplin k is essential when a large
Figure 2 Line-of-sight distance over a smooth earth as a function of height of one end of the
link when the other end is fixed at 10 m.
2 SYSTEM DESIGN 805