Energy2green Wind And Solar Power System

Wind Energy DIY Guide

Get Instant Access

May 2005

horizontal and vertical resolution (see Table 2). Furthermore, as mentioned, by the time the ERS-1 scatterometer wind data were ready for a preimplementation test in the early 1990's, the atmospheric analysis scheme also changed from an OI scheme into a spectral statistical interpolation scheme (Derber and Parrish, 1991; Parrish and Derber, 1992), which is essentially a three-dimensional variational analysis scheme (3D-VAR). It is important to note that the beginning of the operational use of satellite ocean surface winds from ERS-1/2, and SSM/I wind speed data, occurred in the 1990's, and the current operational NASA QuikSCAT wind data were implemented in early 2002. Thus, 3D-VAR has played a very critical role in the use of these satellite surface wind data in the NCEP operational GDAS. A similar 3D-VAR scheme (Courtier et al., 1998; Rabier et al., 1998) was designed for using the ERS-1/2 wind data in ECMWF's GDAS during the late 1990's and early 2000's. However, it should be noted that currently a new 4D-VAR scheme (Rabier et al., 1997) has been implemented for the use of QuikSCAT wind data in ECMWF's GDAS.

As stated earlier, satellite wind data from the scatterometer's measurements of ERS-1/2 and QuikSCAT contain both wind speeds and directions. However, since the wind vectors retrieved from the scatterometer measurements suffer the wind directional ambiguity problem, the quality of scatterometer wind data in the current operational application is to treat them as though they were of the same quality as ship wind data, and as such they can be readily applied to the 3D-VAR analysis of the NCEP GDAS. Fundamentally, scatterometer measurements are backscattered radiance from C-band or Ku-band radars. In principle, both the 3D-VAR and 4D-VAR variational assimilation systems at NCEP and ECMWF can make use of the brightness temperatures and backscattered radiance information in the atmospheric analysis. The current operational use of scatterometer wind data in 4D-VAR of the ECMWF GDAS is to use the backscattered radiance directly. This direct use of backscattered radiance from scatterometer measurements is similar to the use of TOVS radiance data, and in general requires greater computational resources. In a synoptic case study to be discussed in Sec. 5, Yu and Derber (1996) and Yu (1996) have demonstrated the use of backscattered measurements and the use of retrieved wind speed and wind direction information from ERS-1 in the NCEP 3D-VAR analysis. Results of these two studies show that the use of ERS-1 wind data did improve the central intensity of cyclonic pressure and location of the circulation center. However, a comparison of results between the assimilation of backscat-tered measurements and those from the assimilation of the retrieved winds shows that there is little difference in the analysis results for storm center pressure intensity and location in this case study. This is the reason that NCEP elected to use the retrieved wind vectors from scatterometer measurements from ERS-1/2 and QuikSCAT wind data in the current operational GDAS.

Satellite wind data from the radiometer measurements from SEASAT in 1978, and those from the follow on SSM/I of the DMSP satellite from the 1980's to the present time contain only wind speeds, and no wind directions. In an effort to make use of the SEASAT passive radiometer wind speed data in the OI analysis scheme, Yu (1987) has developed a technique which can deduce a unique wind direction for each of the radiometer-measured wind speed data, provided that a large scale sea level pressure analysis field is given. Such fields are typically available at an NWP operational center, such as NCEP and ECMWF, and they are in general of high quality, and the deduced wind directions are quite realistic and compare well with those from collocated buoy reports (Yu, 1987). Another way to use the radiometer-measured wind speed data is to assign wind direction to the speed data from the first-guess (background) wind field of a six-hour model forecast in the assimilation cycle. In fact, this is the method that has been in operational use for the SSM/I wind speed data at the current operational GDAS at NCEP (Yu et al., 1997). It should also be noted that the brightness temperature data from the SSM/I radiometer measurements can in principle be used directly in the 3D-VAR analysis system using a forward model associated with the varia-tional analysis methodology, without having to derive any ancillary wind direction information for the radiometer wind speed data.

3. Impact of Satellite Ocean Surface Wind Data

Observing system experiments (OSE's) have been a benchmark designed for investigating the impact of a satellite data set. Figure 4 is a schematic showing an example of an OSE designed to test QuikSCAT winds and infrared sea surface temperature data from the AVHRR and GOES satellites. An OSE typically invokes a parallel experiment in which a test run and a control run are conducted. The control run uses the conventional data, while the test run contains all conventional data and the additional satellite data set to be investigated in the data assimilation and forecast experiments. Before any new satellite data are implemented in the NCEP's NWP operations, two parallel experiments are conducted, one with the new satellite data, the other without, to assess the impact of the new data set. To be discussed in this

Figure 4. Schematic for the design of parallel global data assimilation experiments to test satellite data impact.

section are summaries of main results from scat-terometer data impact experiments that have been conducted to investigate ERS-1 in 1995 (Yu et al., 1996), and QuikSCAT winds in 2001 and 2003 (Yu, 2003), and those from SSM/I radiometer wind impact experiments in 1997 (Yu, 2001).

The forecast RMS vector wind errors shown in Table 3 are the model's 10 m wind forecasts when compared to midlatitude deep ocean buoys (between 25 North and 60 North) during two parallel data assimilation experiments conducted to investigate the impact of ERS-1 scat-terometer wind data on NCEP's operational global data assimilation system. As discussed in

Table 3. Forecast RMS vector wind errors of model 10 m winds when compared to midlatitude deep ocean buoys (25 North-60 North) during data assimilation experiments (from November 21 to December 31, 1995) conducted to investigate impact of ERS-1 scattero-meter wind data on NCEP's operational global data assimilation system.

Forecast Number of Without ERS-1 With ERS-1 Hours Buoys winds (m/s) winds (m/s)

Was this article helpful?

0 0
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