National Weather Service United States Department of Commerce

Logging Daily Observations


Most observers will use National Weather Service form B-91 (WS B-91) for reporting their daily observations.  Quality control of these forms has indicated that some confusion exists on how to properly log the reports, as well as the station information entries.

Station Header block:
STATION (Climatological) -- This is the name assigned to your station. Include the distance and direction from the main post office, if this has been assigned, or any other identification (used if there is more than one observing site in the same town). For example, "Avon 5NE", "Springfield 2", and "Lincoln" are valid.

RIVER STATION -- For those of you who read river data at a location different than your normal observing site, you would put the assigned name in this block. Otherwise, leave it blank.

MONTH -- This is the 3 letter identifier for the month (JAN, FEB, MAR, etc.)

YEAR -- Fill in the last two digits of the year.

STATE -- Use the two-letter postal identifier for the state ("IL").

COUNTY -- Name of the county where the station is located. In some cases, this may be a different county than the "parent" town listed as the station name. For example, Pana is in Christian County, but Pana 3E is in Shelby County.

RIVER -- Only if you report river stages, write the river name in this block. Otherwise, leave it blank.

TIME OF OBSERVATION RIVER, TEMP, PRECIPITATION -- All 3 of these use the same basic conventions. For example, if you take your observations at 6:00 AM, then you would write "6a" in the box, 6:00 PM would be "6p", etc. Midnight is coded as "12m", and noon is coded as "12n".

STANDARD TIME IN USE -- For Illinois observers, there are only two valid entries: C (Central) is to be used from November to March, and DC (Daylight Central) is to be used from April to October. (Observations are taken at the same clock time, regardless of CST or CDT being used.) We have seen many incorrect entries, e.g. "CST", "Central", "DCST", "Yes" (assuming this was a yes/no question), and leaving the box blank.

TYPE OF RIVER GAGE, ELEVATION OF GAGE ZERO, FLOOD STAGE, NORMAL POOL STAGE -- If you take river readings, the NWS will tell you what to write in these boxes. If you do not take river readings, please leave these boxes blank.

At the bottom right of the observation form, there are 3 blocks that also need to be filled out:

  • OBSERVER -- Please sign or write your name in this block. If your station is a business, where several people take the observations, you may put "STAFF".   We are not allowed to write in your name if the box is blank.

  • SUPERVISING OFFICE -- If this box is not already stamped when the B-91 booklet is sent to you, please write "ILX" in this box.

  • STATION INDEX NUMBER -- If your form has not already been stamped with your index number, please write it in this box.  It is in the form 11-xxxx-d, where "11" is the code for Illinois, "xxxx" is the station number, and "d" is the climate district.

Common Errors

  • These are 24 hour observations, ending at the observing time you indicate. For example, say on your 6 am observation on the 5th, you register a high temperature of 61, a low of 25, and a current temperature of 26. These are marked down in the row labeled "5". Then on the 6th, your high is 70, the low is 26, and the current temperature is 40. Even though the temperature on the morning of the 6th may have only dropped to 40, for the 24-hour period ending at 6 am, the coldest temperature was 26 (the "at-observation" temperature from the previous day). Therefore, you log the low temperature on the 6th as 26 degrees. If you wish, you may put a comment such as "AM low 40" in the remarks line for the 6th.

  • The "at-observation temperature" must be included when determining the high and low. For example, let's say you reported a high of 68, a low of 54, and an at-observation temperature of 50. Your low temperature in this case would actually be 50. (This relates to the example above.)

  • Precipitation totals should be logged for the day that the observation was taken, not the day that it actually occurred. On your 7 am observation on the 15th, you read 0.56 inches of rain in the gage, but you know that all of it fell the evening before. You would still log the 0.56 inches in the box for the 15th, because it is a 24 hour total for the period from 7 am on the 14th, to 7 am on the 15th. On the B-91 forms, there is also a space to mark the times the precipitation fell (if known). This is logged on the day it actually fell. For example, if it rained from 7 to 10 pm on the 5th, it would be marked on the 5th, even though the actual total will be written on the 6th.

  • Report precipitation using the correct number of digits. Liquid precipitation totals use two digits to the right of the decimal point; 24-hour snowfall uses 1 digit to the right of the decimal point, and current snow depth is rounded to the nearest inch (i.e., no decimal point). If any of these values is considered a trace, enter "T".  If none occurred, please put in a "0", instead of leaving the box blank. (This clarifies that there was actually no precipitation, instead of saying that no observation was taken.) For example, the precipitation total may be reported as "T", "0.02", "1.10", etc. Snowfall is in the format "T", "0.2", "1.1", etc. A snow depth of 1.1 inches would be logged as "1", 0.5 inches as "1", and 0.3 inches as "T".

  • For precipitation less than 0.01 inch, snowfall less than 0.1 inch, and snow depth (rounded) less than an inch, put a "T" in the box. Do not use "Trace", "Tr", "0.001", etc. You may include remarks such as "snow flurries", "few sprinkles", etc., but these actually do count toward the precipitation totals, and thus should be reflected in the 24-hour precipitation/snowfall totals.

  • Do not change your observation time without consulting with us! Minor changes, such as routinely taking the observation an hour earlier, may be permissible (after discussing the situation with us), but going from evening observations to morning observations (and vice versa) are not permitted. (If you are simply taking the observation slightly earlier or later than usual, put the actual observation time in the column labeled "Time of observation if different from above".)

Transmitting Observations
Several techniques are available for you to send your daily observations.  (For those of you who currently do not do this, we would greatly appreciate it if you would begin.   Daily reports are used for river forecasting, and for composing summaries for external distribution.)

  • IV-ROCS is an automated telephone system that allows you to enter your observed data. The system codes the observations, and they are transmitted to the NWS for relay around the network.

  • WXCODER uses a computer with Internet access for entering data onto a special home page.  This system also has quality control to avoid observation errors (e.g. low temperature warmer than high temperature). 

  • If observers do not have access to any of the above, they may phone their observations to us using our 800 number.

Regardless of how you send your observations, there are some things to keep in mind:

  • Do not send in your observation more than 1/2 hour before your coded observation time. If you do, then your observation will not be recognized by our computer.

  • Please use the correct value for "trace". For IV-ROCS, press the * (star) key at the prompt. For WXCODER, enter a "T" at the appropriate prompt.

  • Please use the correct value for below zero temperatures. For IV-ROCS users, this is coded by pressing the star key before the number (e.g. *18 for a temperature of -18 degrees).  In WXCODER, enter the value using a negative sign.

  • If no precipitation is reported, do not include a precipitation type code.  However, go ahead and send the precip amount of "0.00".

  • Report the precipitation type code. (Applies to IV-ROCS.)  If more than two precipitation types are observed, you will be asked one at a time. Any code values of 5 through 9 will prompt the system to ask you to enter a snowfall value.

Cooperative Observer Program Topics:
COOP Home, Latest ObservationsObservation Techniques
Equipment, National COOP Page