FAQs

 

I am having trouble downloading the data. What could be wrong?

We recently instituted account registration for access to our data over ftp. You can create an account by visiting http://www.remss.com/register.

If you have downloaded data successfully but are having trouble reading it, it is likely that you need to change your ftp settings. One of the most common problems occurs when binary data files are downloaded as text (ascii mode), altering and corrupting data. Many FTP clients will do this by default. In command line FTP, type "binary" after connecting and before downloading. In FTP clients, check settings: file extensions are often used to determine binary v. ascii mode.

If you have downloaded data in binary mode, and it still seems corrupt, check file sizes. The number of bytes reported by the FTP server should exactly match the size reported by your local file system. If different, try downloading again.

Why aren't my ftp username and password providing access correctly?

Your default username and password are your full email address. If you are trying to access data through ftp via a url of the format [ftp://user:password@ftp.remss.com] with the "@" symbol included in your email, it is often unrecognized. For example:

[ftp://xxx@yyy.com:xxx@yyy.com@ftp.remss.com] will not work correctly. This can be remedied by encoding the "@" symbols in your email address to "%40". The example then becomes: [ftp://xxx%40yyy.com:xxx%40yyy.com@ftp.remss.com]

How can I find out when you update your data?

Many of the instruments we process data for are continually updating. Other datasets are updated once monthly (i.e. the wind and vapor products are usually updated on the 10th or 11th of each month, and CCMP is updated twice per year around January and July). For those that are updated infrequently, you can subscribe to our email updates for any or all of our products here: http://www.remss.com/about/user-registration. We typically only send 1-2 emails per month as necessary.

I noticed some data are missing. What should I do?

If you notice some missing or empty data files, please first check that they are not listed as missing on the webpage for the specific instrument/measurement in question. If they are not, please contact us via the support box.

Why are some files named .rt and others .v7?

Files in .rt format are “real time” which means that at least some portion of the input data are tentative. Remote Sensing Systems aims to make data available as soon as possible for operational use, but some improvements can be made post-hoc once full atmospheric analyses are available which include in situ measurements. When final processing is complete, the .rt files are removed from the server and replaced with .v7 files. The .rt files are not intended to be research quality. Some possible issues within them include erroneous brightness temperatures and geolocation errors.

Can we use your data on our web site or in our document?

Since our data are public, you are free to use the data for your research, in a document, and on your web site. We do ask that you properly cite us. For more information, see our data citation page. Our funding is dependent on our ability to show that the data are used by others. By stating that you used RSS data in your documentation and giving adequate credit, you are helping to make sure that these important data products continue to be freely available.

Are winds one minute or ten minute mean winds (or equivalents thereof)?

The satellite images are instantaneous snap-shots of a 25 km x 25 km area. This is probably more comparable to an 8 - 10 minute wind (if you consider the amount of space measured by a storm moving past a fixed point making observations for 8 minutes).

We validate satellite winds with buoy winds, most of which are temporal averages ranging from 6 minutes (TAO) to 8 minutes (NDBC), and generally obtain wind speed differences of less than 1 m/s.

I need a 5-day averaged product. Can I get that from you?

No. We do provide data from a few instruments at 3, 7, or 8 day averaging intervals due to instrument sampling patterns, but will typically not fulfill specific requests for averaged products. However since we offer all of our products in daily files, you can create your own 5-day running average using the data analysis software of your choice (IDL, Matlab, r, etc…)

How are the 3-day, weekly and monthly average data maps made?

The radiometer (SSM/I, TMI, and AMSR-E) and scatterometer (QuikScat, SeaWinds) data are all available as 3-day, weekly and monthly averaged data. These files are created from the earth gridded daily pass data. Each grid cell contains the average (mean) value of all valid geophysical data points in that cell for all daily passes within the averaging time period. For grid locations where no data exist within the averaging period (occasionally occurs in 3-day maps), a value of 254 (missing data) is assigned.

Ice (a value of 252) is assigned to the grid cell if ice was present more often than valid data within the averaging period.

Do you have any data over land?

No, the algorithms used by RSS to calculate wind, cloud water, atmospheric water vapor content, and rain-rate are tuned for use only over the ocean. CCMP winds are the only dataset produced at RSS that has data over land, because it uses the ERA-interim model wind field as a base to be adjusted by the satellite and in-situ wind measurements. However, the land data remain unaffected by RSS satellite input, so that data is simply the ERA-interim background.

What is the resolution of the data?

The single-satellite data products available at RSS (GMI, TMI, SSMI/SSMIS, AMSR2, SMAP, WindSat) are all provided on 0.25 by 0.25 degree grids. The Cross-Calibrated Multi-Platform (CCMP) gridded surface vector winds are also on a quarter-degree grid. The other multi-satellite products (wind and vapor) are at 1-degree resolution.

Why are there limitations on F15 data use? What is wrong with it?

Data from the F15 satellite experienced biases stemming from interference between the radiometer and a radar calibration beacon that was activated in August 2006. Although corrections to the data were made, data quality continued to degrade, especially in 2008 and early 2009. Interference is affected seasonally, and worsens during late winter and early fall when F15 enters the earth’s shadow. More detailed information can be found in the following blog post from 2009 here.

Why do you have negative cloud values in your data? How is this possible? Is this an error?

We use a small negative offset to account for random noise in the data. In clear sky conditions, all cloud values should be zero. However, there is some noise inherent in the data. So in practice, clear sky values average zero. Some are a little above and some are a little below zero. We keep the clear sky values that land a little below zero so that they average out with the clear sky values that land a little above zero. If we set negative cloud retrievals to zero, then it would tend to push the average a little too high.

What is the difference between an 11 GHz wind(lo) and a 37 GHz(med) wind?

We produce two standard rain-free radiometer wind products: WSPD_LF (low-frequency) and WSPD_MF (medium frequency). The first, WSPD_LF is created using the frequency channels at 10.65 GHz and above (see tables above) and is most similar to the only wind offered in the version-5 AMSR-E data files or the first wind of the TMI files. The second, WSPD_MF, uses frequency channels at 18.7 GHz and above and is most similar to the SSM/I and SSMIS winds.

Each wind product has distinct advantages. The WSPD_LF is less affected by the atmosphere and rain, but is affected by 10.65 GHz RFI and sun glitter effects. The WSPD_MF has a higher effective spatial resolution, is less affected by ice and land contamination, and is only slightly affected by sun glitter effects and RFI. The WSPD_MF are a little noisier than the WSPD_LF.

Why do I see data on daily maps with tomorrow's date, when tomorrow has not yet occurred?

Each daily map is for ascending or descending passes of a satellite on that date at Universal Coordinated Time (UTC), also known as Greenwich Mean Time (GMT) and Zulu (Z) time. UTC is the mean solar time at the prime meridian, and is the primary standard time used to regulate clocks around the world. UTC does not observe daylight savings time.

For example, data may be collected in the New York area at approximately 8:30 pm local time. But 8:30 pm in New York City is 1:30 am in Greenwich, England (8:30 + 5 hour time difference). If it is August 17th, local time in New York, it must be August 18th in Greenwich, England, a location with GMT (the date of the map). Thus, some data collected on Aug 17th, local time, will appear on the August 18th map.

Similar time "issues" occur for each of the satellites. The table below demonstrates another example, this time for data collected by the F13 satellite. F13 crosses the equator at approximately 5:30 am.

Location

Lat/Lon

F13 local Time/Date

F13 GMT Time/Date

Map Date

New Guinea 0 / 133 5:30 AM 20-Sep 8:30 PM 19-Sep 19-Sep
Jarvis Island 0 / 203 5:30 AM 19-Sep 4:30 PM 19-Sep 19-Sep
Ecuador 0 / 280 5:30 AM 19-Sep 10:30 AM 19-Sep

19-Sep

Here, the F13 morning pass time of 5:30 (local time) is used to determine the GMT of the map date. The corresponding GMT time and date are then used to determine the date at the given location. When the September 19th F13 morning pass was collected, it was actually September 20th in New Guinea, a location that is 8 hours ahead of Greenwich.

The best way to determine the time of observations for scientific comparisons is to use the time data in the daily binary data files. Each pixel location of an ocean parameter map has a specific GMT minute of day listed in the time array. Use of time in this manner will keep your data processing and intercomparisons in the correct order.

How can I tell when a satellite collected data over a particular spot on a daily data map?

The time parameter can be extracted from each daily binary data file using read routines available on our FTP site (see the Data Description for each instrument). A visual method for quick approximations is discussed in: Support / Crossing Times / Swath Time Labels.

Who names tropical cyclones and how do they generate the names?

Tropical cyclones are named by a World Meteorological Organization committee. Here is a great page that describes the naming of tropical cyclones and provides current lists of names:
Worldwide Tropical Cyclone Names.

Do you have an older version of a data product?

When we update to a newer version of a data product we usually make significant improvements to the data.  We therefore discourage using older versions.  That said, if you need access to an older version for confirmation of a previous analysis, please contact us.

I need brightness temperature data.  Where can I get it from?

Brightness temperatures are available for many of the microwave radiometers.  We do not currently provide the files at RSS.  They are available from some of the NASA DAACs and NOAA data centers.  See the links in the Brightness Temperature Data Access Table.

What is oceanographic vs meteorological wind direction?

Wind directions are described as either "coming from" a direction (used by meteorologists) or "blowing to" a direction (used by oceanographers).  The figure shows how we report the wind directions in our data products.  This oceanographic convention means a 45 degree wind blows towards the northeast. Oceanographic Wind Direction Key

What is LTAN?

LTAN stands for Local Time of the Ascending Node and represents the time when the satellite crosses the equator when traveling from the south pole to the north pole (ascending).  This time is the time of day at that Earth location when the satellite is overhead.  For sun-synchronous orbiting satellites, this time is the same for each time zone (that is, a satellite with an LTAN of 6 am will be overhead at 6 am for every location on the equator each day.

What is a sun-synchronous orbit?

Sun-synchronous orbits combine the satellite’s altitude and inclination to allow measurement of the earth’s surface at the same local time in every time zone. A satellite in sun-synchronous orbit is typically located about 600-800 km above the earth’s surface. Most earth observation satellites used for climate purposes are in this low-earth orbit, in stark contrast with geostationary satellites typically used for short-term meteorological purposes, which orbit at nearly 36,000 km and have a much wider view of the earth’s surface. For more detailed information on orbit types, including images and a link to an animation, please see here.

What are the differences between the low-frequency and medium-frequency winds?

We produce two standard rain-free radiometer wind products:  WSPD_LF (low-frequency) and  WSPD_MF (medium frequency).  The first, WSPD_LF is created using the frequency channels at 10.65 GHz and above (see tables above) and is most similar to the only wind offered in the version-5 AMSR-E data files or the first wind of the TMI files. The second, WSPD_MF, uses frequency channels at 18.7 GHz and above and is most similar to the SSM/I and SSMIS winds.
Each wind product has distinct advantages.  The WSPD_LF is less affected by the atmosphere and rain, but is affected by 10.65 GHz RFI and sun glitter effects. The WSPD_MF has a higher effective spatial resolution, is less affected by ice and land contamination, and is only slightly affected by sun glitter effects and RFI. The WSPD_MF are a little noisier than the WSPD_LF.

“I saw this plot on a denialist web site. Is this really your data?”

We get this question quite a bit since many people use and look at our data to help them understand climate change. We encourage you to read more in the blog post containing a full explanation of Carl Mears' point of view on the RSS AMSU data and the differences with climate models used in the IPCC AR5 report.


If you did not find an answer to your question, please contact us.  We aim to respond to support submissions within 2 - 3 business days.