-
The most accurate moisture measurements are obtained from a chilled-mirror hygrometer, generally aircraft-mounted and operated by a trained observer. No operational radiosonde is considered to be an adequate reference, but carbon and capacitive humidity sensors are becoming increasingly responsive and consistent.
-
As future generations of operational satellites are designed, most instruments are being upgraded to include more channels, better spatial resolution, and less instrument drift. Errors of individual moisture retrievals should decrease somewhat due to improved spatial resolution. The vertical resolution of moisture profiles is unlikely to improve greatly because additional spectral channels provide only a little independent information.
-
Experimental satellite results suggest that potentially more useful global estimates of temperature and humidity (derived from refractivity) profiles could be obtained through the use of GPS receivers on polar-orbiting satellites.
Recommendations for upper air network operation -
See recommendations for the radiosonde network as described under the temperature parameter
-
Stable operation of all GUAN stations should be encouraged, with as few relocations or instrument changes as possible. GUAN stations in China, India, and Russia should transition to modern humidity sensors quickly, and all stations using older sensor types should upgrade as soon as feasible.
-
WMO instrument intercomparisons and similar ongoing co-operative research programs should continue with approximately the current level of effort. It would be desirable to include Japanese instruments in an intercomparison.
-
All countries should report all station moves and instrument or processing algorithm changes. The WMO online instrument catalogue should be maintained as an ongoing project, and should include recent historical data, but CARDS should be the most comprehensive historical metadata archive. A current effort to infer missing metadata by detailed examination of station time series should be completed.
-
Humidity should be reported to the highest feasible level in each sounding, especially if the sounding uses Vaisala, VIZ (Sippican), or another modern-type instrument.
Recommendations for satellite moisture measurements -
Efforts should be encouraged to develop satellite moisture retrievals in all conditions (cloudy and non-cloudy, with and without precipitation, over land and oceans) so retrievals are not biased.
-
Long-period satellite retrievals should be published, including computed time variations, for research evaluation. As many sources as possible should be included in blended data bases, such as NVAP (NASA Water Vapour Project), with sources readily separable to perform intercomparisons.
-
When validated radiosonde observations, adjusted to compensate for instrument discontinuities, become available, satellite retrievals should be compared with as many collocated soundings as possible. Multiple intercomparisons between satellite and radiosonde data sets should detect both satellite and radiosonde discontinuities, and should gradually build confidence in computed water vapour trends. Bias-adjusted radiosonde data should allow checking of satellite retrievals even when there is little or no satellite overlap.
-
The incorporation of GPS receivers on polar-orbiting satellites should be encouraged.
Recommendations for ground-based moisture measurements -
The one single record of upper tropospheric and stratospheric water vapour (from Boulder, Colorado) is invaluable and should be supplemented with other records around the globe to create a network of reference sites in key regions.
-
The considerable promise for obtaining total column water vapour observations over land from ground-based GPS receivers should be exploited globally through international coordination.
Variable: Upper air wind (surface to 30 hPa)
Main climate application
Essential indicator of atmospheric dynamics (circulation), transport of heat, moisture, pollution, trace species, chemical constituents, etc; teleconnections and planetary waves, and for identification of changes due to climate change. Vital for the evaluation of climate models.
Contributing baseline GCOS observations
Global Upper Air Network (GUAN), a subset of around 150 of the global network of radiosonde stations that support weather forecasting. Chosen to provide the best available coarse resolution global network for detecting changes in temperatures, this network is also useful for upper air winds. See upper air temperature.
Other contributing observations
WWW upper air network of around 900 radiosonde stations. About 96% of the stations that make radiosonde soundings (Raobs) also make wind soundings. Nearly two thirds of the stations are designated to make observations at 0000 UTC and 1200 UTC and a few make soundings at 0600 and 1800 UTC. In addition to soundings taken in conjunction with Raobs, there are 100 to 200 wind soundings taken alone (Pibals) at all 4 times of day, with slightly larger numbers at 0600 and 1800 UTC (complementing the Raobs). Between 100 and 200 stations make observations once per day, while about 100 have 'temporarily' suspended operations. In ocean areas, radiosonde observations are taken by around 15 ships fitted with automated shipboard upper-air sounding facilities (ASAP). Most of these soundings are presently from the North Atlantic and North West Pacific Oceans, but the programme is expanding into other ocean basins.
Several (about 6) upper tropospheric profilers were established on islands in the tropical Pacific, and a few were complemented with low troposphere profilers. The measurements of wind, including the vertical component, proved very useful for evaluating global analyses and led to some useful research results, but the failures to adequately distribute and process the data, and the logistical difficulties and costs led to the demise of this project in 2001.
In addition to balloon soundings, which have declined in numbers by perhaps 20% in recent years (relative to the early 1990s), many wind observations are made by aircraft, both while on route at altitude at 300 to 150 hPa, and as profile 'soundings' on ascent or descent into airports. The latter have increased enormously in volume since they began in 1993 but are most abundant within the United States. Other observations, from about 50 °N to 45 °S, are made by tracking cloud elements along with an altitude for the observation, to provide 'satellite winds' (SATOBS). In the lower troposphere these are from 1,000 to 700 hPa, and in the upper troposphere they are in the 400 to 150 hPa layer. They have generally improved in recent years, especially in altitude assignment; however, an order of magnitude increase in volume in 1998 has led to more redundancy and perhaps lower quality. Most of these winds are over the Pacific-Americas sector east of 180 ° to about 20 °W, where there are about twice as many in the upper vs. the lower layer, whereas elsewhere the numbers in the upper and lower layers are similar.
In all, in the year 2000 on average for each day, there were 1,226 Raobs, an additional 563 wind soundings, an average of over 22,000 aircraft wind reports at altitude (most over North America), over 42,000 reports as aircraft soundings, and over 168,000 satellite winds.
Significant data management issues
Key quality checks occur as part of four-dimensional data assimilation. Isolated wind observations can not be trusted. Data are archived and have been assembled for reanalysis projects by NCAR and the operational centres.
Analysis products
Most winds are not analysed separately, but are incorporated as part of four-dimensional data assimilation into multivariate analyses of the atmospheric state. Wind speeds and directions are given special scrutiny for aircraft operations. As well as operational products, more consistent sets of analyses result from the reanalyses from NCEP and ECMWF using a fixed state-of-the-art system. Such analyses also allow the rotational and divergent parts of the wind to be separately analysed (or equivalently vorticity and divergence). The divergent component is especially important because of its link to precipitation and diabatic processes.
Current capability
As Raobs have switched from radar and other tracking methods to methods using geostationary positioning system (GPS) sondes, the accuracy of wind measurements has increased. However, this has come at a cost and has led to the drop out of some stations and soundings. Aircraft observations and profiles in some locations show the potential for greatly enhancing the network in other areas. However, aircraft observations are only in the air corridors and are missing in many parts of the world; in part this is simply because they are not gathered.
In the Tropics, wind information is much more critical than temperature, and the dynamical constraints and ability to infer winds from other observations is less. Moreover, aircraft winds are few and profiler winds have ceased, decreasing capabilities in recent years.
Current products are useful for detailing means and interannual variability, but decadal variability and trends are lost in the noise of the changing observing system.
Issues and priorities
-
For some time there has been discussion of a satellite mission to measure winds directly using lidar. Because this involves an active laser radar instrument, the power requirements and costs are high, and progress in this area seems to have abated (see the article by Baker et al., 1996: Bull. Amer. Met. Soc., 76, 869-888) but the potential exists.
-
Prospects for improved products also exist from improved sounders of temperature (see upper air temperature), which will translate into improved winds through the assimilation process. Nevertheless, these changes are further perturbations to the observing system and are not necessarily helpful for tracking small changes over many years.
-
Reanalyses that address issues of the changing observing system and the best products for tracking climate are greatly needed. Observing System Experiments (OSEs) are a useful tool to detail the impact of changes in observing systems.
Recommendations are to:
-
Increase the profiles from aircraft as they take off and land.
-
Increase the recovery of aircraft winds.
-
Decrease costs of GPS sondes and increase their use by upgrading stations that are currently unable to perform.
-
Reprocess satellite winds using the latest techniques for use in reanalyses.
-
Strongly support reanalyses, especially the need for new reanalyses that are tasked to address changes in the observing system and the impacts on the analyses.
-
Consider fleets of constant level balloons that can be tracked for winds, especially in the Southern Hemisphere, as a relatively low cost option.
Variable: Clouds
Main climate application
Cloud feedback is the single most uncertain aspect of future projections and is responsible for much of the range of sensitivity of atmospheric and climate models. The key issue is how will clouds change as the climate changes and it is necessary to deal with this in terms of amount, height, and radiative properties. The issue is extremely complicated because clouds will change due to changes in aerosols as well as the changing climate. In addition, it is difficult to distinguish clouds from snow and ice fields from space.
Contributing baseline GCOS observations
There is no GCOS baseline system for observing cloud at present, and an effective strategy needs to be developed for monitoring cloud, as well as the related variables of water vapour and aerosols.
Share with your friends: |