Sign inRegisterCheck connection

A Single Weather Image as a Citable Observation

Cite one weather photograph with capture time, camera identity, and caption. A clockless webcam JPEG is a picture, not an observation record.

Back to Personal weather station guides

A weather photograph is citable when a later reader can state what was seen, when, from which instrument, and in which direction. The display page for one file is that citation. It is not a lightbox, not a comment thread, and not a webcam frame that lacks a clock. Visitors who followed /gallery/displayimage.php were opening Coppermine’s single-image view. This article keeps that address. No historical JPEG is replayed. No live camera is attached.

The gallery index is the catalog: albums, required fields, and what may be listed. This page is one row from that catalog, fully written out. If the index is a station logbook, the display page is the signed entry.

Historical context

Coppermine’s displayimage.php was the permalink for a file ID. Incoming links, including query strings, pointed at one picture. The restoration does not keep those IDs as a media CDN. It keeps the scientific object the permalink implied: a single still with a caption and a time.

The original interactive gallery is not rehosted. Recovered neighboring pages show the same Coppermine 1.3.3 family as the album list and the password-reminder URL. Those surfaces were account and browsing chrome. The display URL was the citation surface, even when the software treated it as a social page with “next,” “previous,” and comments.

Citation, not a lightbox

A citable still is a small observation record. It needs the same kinds of labels a logger row needs, adapted to an image:

| Label | What it answers | |---|---| | Camera or file ID | Which instrument or which stable token | | Capture time | When the exposure was made, with timezone | | Time source | EXIF, overlay clock, filename, or “unknown” | | Viewing direction | Azimuth, or “not recorded” | | Caption | What the frame is allowed to show | | Weather note | Optional, and marked as an operator note | | Rights | Whether the file may be kept, copied, or only described |

Without capture time, the still can illustrate a sky type in a teaching note. It cannot join a time series. Without camera identity, two similar blue frames cannot be told apart. Without a caption that states the scene, a later reader will invent one.

Comments are not captions. A visitor remark is not part of the observation. If you preserve comments at all, keep them in a separate field labeled as commentary.

Time that traveled with the file

Three clocks commonly disagree on hobby stills.

EXIF DateTimeOriginal is the intended capture timestamp in the Exif standard (CIPA DC-008). It is only as good as the camera clock. Many network cameras write it late, in UTC, or not at all. DateTime in the same header is often a file-change time. Treating DateTime as the observation instant is a citation error.

An overlay clock burned into the JPEG is part of the picture. It is useful when the stamp was applied at capture from a clock you set and checked. It is misleading when the stamp is the web server’s time at upload, or when the overlay was added years later in a batch. If overlay and EXIF disagree, the display page must say so. It must not silently pick the prettier one.

Filename time is acceptable when the name encodes a sortable timestamp and a camera ID, and when that name was assigned at capture rather than at a later rename. cam01-20100312T1415MST.jpg can carry the observation. IMG_4588.jpg cannot.

FTP modification time is not an observation clock. A batch upload collapses a day’s stills onto one second.

A webcam JPEG served as current.jpg without a clock is a picture of a sky, not a citable observation. Interval and field-of-view policy for the camera itself are a different page. This article is about one file that already exists. If that file has no time, the display page should say “time unknown” and refuse to title the still as a dated event.

Do not use the browser’s “now” as the observation time. Page-load time is a website event.

Caption as scientific text

The caption is not marketing. It should be boring and checkable.

Good captions name the view in a way a stranger could verify from a site plan: “North sky above the rain gauge, 15-minute still.” They do not claim WMO cloud amount or cloud type unless a trained observer or a documented instrument produced that code. A consumer camera with automatic exposure is a qualitative diary. “Overcast-looking” is an honest note. “OVC” as if it were METAR is not.

Good captions refuse to smuggle logger numbers into the photograph. If you want temperature beside the still, print it as a separate observed value with its own timestamp, and only if you can defend simultaneity. A temperature from a 5-minute logger packet next to a still whose EXIF is 40 minutes off is two observations, not one.

The caption should state limitations that change use: night, dirty dome, rain on the lens, IR cut-filter change, or a pan that moved the azimuth. Those are instrument notes. They belong next to the image the way a station siting note belongs next to a thermometer.

Do not invent a storm narrative. If the file is a clear noon, the caption is a clear noon. If you do not know, you do not know.

What a single-image page is not

It is not the album index. Next and previous links are navigation. They are not a time-lapse. A sequence product has to prove order; that is a slideshow problem, not a citation problem.

It is not a live webcam. This URL does not proxy a camera, does not refresh a JPEG, and does not restore Mesa imagery.

It is not a password gate. If the file was private in Coppermine, the honest public page says the original object is not rehosted. Recovering operator access to a disk archive is covered on the archive-access page.

It is not an official warning. Lightning in a frame or flooding on a drive is not an NWS Watch, Warning, or Advisory. Hazard decisions go to weather.gov or the local authority.

Practical checklist

  • Print camera ID, capture time, timezone, and the source of that time.
  • If EXIF, overlay, and filename disagree, list the disagreement.
  • Caption states view and limitations; it does not invent cloud codes or storm stories.
  • Logger values, if shown, keep their own timestamps and are labeled observed or derived.
  • Comments stay out of the observation record.
  • No “live” label; no browser clock as capture time.
  • Missing time is stated; the still is not back-dated to make a neater citation.

Modern relevance

Share links and object-storage URLs made displayimage.php look obsolete. The citation problem did not go away. A CDN link without EXIF, without overlay, and without a caption is weaker than a 2005 Coppermine page that at least had a title field—if that title was used as science rather than as a joke.

Phone photos used as “the sky at the station” fail this page constantly: the phone’s timezone, the station’s timezone, and the logger’s archive interval are three systems. A still shot on a walk around the yard is a different viewing direction than the fixed webcam. Putting it in the same ID series is a silent instrument change.

TNET does not display a recovered file ID at this path. Anyone who followed /gallery/displayimage.php for a picture gets the citation rules that picture required.

TNET research bridge

A dated, identified still is an observed visual record. A caption is operator text. A slideshow assembled from many stills is a derived display. Those distinctions are the public vocabulary on how the service works. Provenance for the file itself—which camera, which time source, which archive—is the same class of question as data sources, quality controls, and methodology. The weather-station hub lists the neighboring instrument pages.

Sources