Year

Figure 1. Evolution of anomaly correlations for 500 hPa geopotential height forecasts at NCEP for the northern and southern hemispheres.

Figure 2. Evolution of anomaly correlations for 500 hPa geopotential height forecasts at ECMWF for the northern and southern hemispheres.

(Hollingsworth, 2004). Both figures demonstrate the current state of NWP forecasting skills, and convey the main conclusion that forecasting skills at Day 5 have steadily been improved for both the northern and southern hemispheres at these two major NWP operational centers over the last 40 years. Furthermore, they show that in the early days of NWP, owing to a lack of data in the southern hemisphere, the forecasting skill in the SH is far less than that skill in the NH. However, this gap in the forecasting skill has been considerably narrowed between the two hemispheres owing to an increase in satellite observations during the last few decades.

The advent of vast numbers of satellite data introduces a critical issue regarding the design of a viable atmospheric analysis and data assimilation scheme in order to most effectively use these satellite data for NWP. In the early days of NWP operations during the 1960's, the Cressman analysis scheme (Cressman, 1958) was used operationally at NMC (National Meteorological Center, the predecessor of NCEP). It was designed mainly for its efficiency. However, the main deficiency of the Cressman scheme is its arbitrariness of weights assigned to each observation as a function of distance between the observation and the grid point to be analyzed. With the rapid improvement in the computing speed and data storage during the last two decades, the current operational atmospheric analysis methods have evolved from the Cressman scheme during the 1960's and 1970's, and the Optimum Interpolation scheme (Gandin, 1963) during the 1980's and early 1990's, to the current variational analysis schemes (Sasaki, 1972; Derber and Parrish, 1993) from the late 1990's to the present. Most of these schemes are designed so that they can make use of various kinds of global observations and take into account unique characteristics for each type of satellite remotely sensed data.

Remote sensing satellite data can basically be vertical profile data, such as temperature and humidity soundings from TOVS (TIROS Operational Vertical Sounder); or surface level data, such as ocean surface winds, sea surface temperatures, albedo, and vegetation indices. Satellite data can also include many other types, such as those on vertically integrated total precipitable water, and cloud-tracked winds, etc. In general, profile data on temperatures and humidity are retrieved from the observed values of radiance from satellites. The radiance data, when used directly in the atmospheric variational analysis, are found to be most useful for NWP operations owing to their great coverage in three spatial dimensions. Single level surface wind data from various satellites during the last 20 years or so generally fall into two categories: one category contains ocean surface vector winds retrieved from active scatterometer measurements, and the other category contains wind speed information only observed from passive radiometer measurements. Owing to the wind directional ambiguity problems associated with the scat-terometer wind data, and the incomplete wind information (only wind speed information, but no wind directions) associated with the passive radiometer wind data, methodologies designed to use these single level satellite surface wind data in the atmospheric analysis and data assimilation systems are different from those for the use of radiance data. A brief discussion on the evolution of the atmospheric analysis schemes at operational NWP centers such as NCEP and ECMWF is given in Sec. 2, with a particular emphasis on data assimilation methodologies designed for the use of the single level surface wind data from satellites.

The use of satellite ocean surface winds in the operational global data assimilation systems (GDAS's) at NCEP and ECMWF has been found to have a positive impact on short range numerical weather forecasts. The first satellite to measure surface wind vectors over the oceans was SEASAT, which was launched in 1978, and had on board both an active scatterometer and a passive multichannel scanning radiometer, among other instruments. This was a proof-of-concept oceanographic satellite, which for the first time demonstrated that global ocean surface wind fields could be measured by a remote sensing technology. Following SEASAT, many other satellites have been launched, such as GEOSAT, ERS-1/2, SSM/I, NSCAT, and QuikSCAT, which were designed for observing ocean surface winds. Satellite ocean surface winds from ERS-1/2 were used operationally at NCEP and in ECMWF NWP operations during the late 1980's, while QuikSCAT winds are currently being used operationally at both centers. During the last two decades or so, there have been a considerable number of global data assimilation experiments conducted to test the impact of these satellite surface wind data before their implementation in the NWP operations. Some of the recent preimplementation impact experiments conducted at NCEP, NASA, and ECMWF on the use of ERS-1/2 and QuikSCAT wind data in various GDAS's will be presented in Sec. 3. However, only gross statistics over a large number of analysis and forecasts of these preimplementation tests will be presented. For details of impact investigations into the use of satellite ocean surface winds, see Baker et al. (1984), Yu and McPherson (1984), Duffy et al. (1984), Yu (1987), Hoffman et al. (1990), Stoffelen and Cats (1991), Ingley and Bromley (1991), Hoffman (1993), Thepaut et al. (1993), Yu et al. (1994), Yu and Derber (1995), Yu et al.

(1996), Stoffelen and Anderson (1997), Yu et al.

(1997), Andrews and Bell (1998), Yu (2000), Atlas et al. (2001a), Atlas et al. (2001b), Yu (2003), and many others.

The main purpose of this article is to review recent progress in the use of satellite ocean surface wind data in atmospheric analysis and data assimilation, to discuss the impact and current status of operational applications of these wind data in NWP, and to demonstrate that in some synoptic cases where applications of satellite ocean surface winds can lead to very significant improvements in the ocean surface wind and sea level pressure analyses over the southwestern Pacific region. The outline of this article is as follows. In Sec. 2, the instrument error characteristics and data coverage of various satellites of the last two decades designed for measuring ocean surface winds are described. Then, a brief account of the evolution of the atmospheric analysis schemes at NCEP is given, with special emphasis on the use of satellite ocean surface winds and their quality control procedures in the global data assimilation experiments. Section 3 discusses results of impact investigation related to various satellite ocean surface wind data, such as ERS-1/2, SSM/I, and QuikSCAT wind data, followed by the current status on the use of satellite ocean surface winds in NCEP and other operational centers' NWP operations. In Sec. 4 a synoptic case is discussed where use of satellite surface winds shows the most significant impact.

2. Assimilation of Satellite Ocean Surface Winds in Atmospheric Analyses

2.1. Characteristics of satellite ocean surface winds

As stated earlier, the two microwave instruments that can measure ocean surface winds are the scatterometer and the radiometer. The main principle of these two instruments lies in their ability to detect very small scales of motions of gravity and capillary waves over the ocean surface, from which surface winds can be deduced. Scatterometer measurements of winds contain both wind speed and direction, with wind directions consisting of as many as four values, of which only one is correct. This wind direction ambiguity problem is inherent in the scatterometer measurements for all the follow-up satellites, such as ERS-1/2 from the European Space Agency during the late 1980's, and most recently NASA's QuikSCAT during the late 1990's. On the other hand, radiometer measurements from SEASAT and various follow-up satellites such as GEOSAT during the 1980's, and SSM/I (Special Sensor Microwave Imager) of the Defense Meteorological Satellites Program (DMSP) during the 1990's and up to the present time, contain only wind speed, without wind direction information.

Satellite ocean surface wind specifications for various platforms such as radiometer

Table 1.

Tsann-Wang Yu Satellite ocean surface wind specifications for various platforms.

Table 1.

Tsann-Wang Yu Satellite ocean surface wind specifications for various platforms.

Satellite

DMSP f11, f13, f14

ERS-1/2

QuikSCAT

Sensor

Passive microwave

Active microwave

Active microwave

Number of antennas

7 channels

3 antennas

4 antennas

Observed field of view

Scanning

1-sided

2-sided

Frequency/Bands

19 (H,V), 22V,

C-band radar

Ku-band radar

Renewable Energy 101

Renewable Energy 101

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable. The usage of renewable energy sources is very important when considering the sustainability of the existing energy usage of the world. While there is currently an abundance of non-renewable energy sources, such as nuclear fuels, these energy sources are depleting. In addition to being a non-renewable supply, the non-renewable energy sources release emissions into the air, which has an adverse effect on the environment.

Get My Free Ebook


Post a comment