Back to event list
General notesThis site's purpose is to show event path maps (occultations and eclipses) overlaid directly on real-time weather forecast maps, to avoid having to flip between two different maps when checking the weather for mobile events. This is an evolving site: features change, and not everything is documented.Auto-refreshWeather imagery is automatically refreshed by default (unless changed in settings -- see below), so the page does not generally need to be manually reloaded to get up-to-date imagery.Auto time seekEvent time is the time set for this map's event to occur, e.g. the geocentric midtime of an occultation, or midtime of an eclipse. It is arbitrarily set for each map, and indicated by a star ( ) on buttons and widgets. On initial load, the map will automatically show the weather at event time if possible, i.e. if the default forecast is in range; or the star button can be pressed to select it. Thus one does not need to compute and select the correct forecast time each visit. Some weather pages may not have an event time defined.WidgetsMany widgets have a brief tooltip that pops up when they are hovered over. This may give help and/or status on the item. This includes some non-clickable items, like the house and star icons under the imagery slider. Buttons are white on blue when they are selected/on, black on white when they are unselected/off. Pop-up dialogs can be dismissed by clicking the X at upper right (if available), pressing Esc, or clicking on the page outside the popup (except for Help, which stays open).BookmarksMost GUI selections are stored in the URL anchor (the #... part). This means that a particular provider/region/time etc. can be set and the URL bookmarked or shared, and those selections will be restored on revisit. (However, this will lose the default provider/region; see that section below.)Choosing imagery type (clouds, wind, etc.) and regionProvider selects the overall weather service provider. This controls not only the type of imagery (forecast vs. satellite), but also what models, model runs, products, and regions are available, as different providers have different offerings. Most providers are focused on only one or a few continents; hence they are grouped that way. Product selects which product to show, e.g. cloud cover, temperature, wind etc. Total cloud cover or its equivalent is generally the default/first product. Region and the 8 arrows beneath it select which region to show. The arrows move roughly in the direction indicated. Note that different providers/products/etc. have different regions, and they are rarely due-north/south/east/west of each other, nor of consistent size. The arrows may thus not always be reciprocal; e.g. moving southeast and then northwest may not always return to the same region (or even be allowed). An arrow with no region reasonably adjacent in that direction from the current region will be disabled (greyed out). Note that some regions may only be accessible via the select box and not the arrows; e.g. moving to/from Contiguous United States and the various other regions in the NWS forecasts provider. Model selects which forecast model to show. A forecast model is a statistical tool, run by complex software, which is used to generate the forecast maps. Various weather agencies have developed different models, and their predictions often differ -- so check multiple models where possible. Some providers only offer one model, so it is unselectable for them. Satellite providers do not use any models, as they are showing observations not forecasts. Model run selects which model run to show. Each model is run several times a day as new weather information is added, to update the forecasts. The (latest) run is typically the most accurate -- as it incorporates the most recent data -- and thus is chosen by default. (If model runs are offered but none say (latest), then auto-refresh is off; see settings.) One way to see if a model getting "confident" about its forecast is to halt the display on a fixed forecast time of interest (e.g. event time), and choose different model run dates: if the weather starts to look consistent across the most recent several runs for the same forecast time, then the model is more sure of its predictions. (Be sure that the selected forecast time is in range of all model runs being checked: the image timestamp should stay the same.)Cascading optionsIf a used-car website were to have two select boxes for vehicle make and model, the available model options would likely change based on the make selected -- i.e. the make selection "cascades" down to the model options. Similarly on this site, Provider, Model, Product, and Region selections cascade downstream in that order. This means that selecting the provider will change the available models, products, and regions; selecting the model will change the available products and regions but not providers; etc. The interface will attempt to preserve downstream selections when their available options change due to an upstream selection, but this may not always be possible. In such cases, another upstream selection may need to be made to restore or change downstream available options.Default provider/regionSome maps have one or more default providers and regions defined, which are selected if none are set in the URL (no #... anchor info) on initial page load. These defaults (especially region) are defined to best fit the event map; e.g. an event that mostly occurs over the eastern US might have CMC and Eastern United States defined as its default. Since the preferred/"best" provider might not have the longest forecast range, alternate defaults may also be defined, which will be selected if the event time is out of range of the most-preferred provider. For example, many US event maps have Wetterzentrale forecasts / North America as a secondary default for a more-preferred default of CMC / Eastern United States: the former is lower-resolution but longer-range, the latter is higher-resolution but shorter-range. On such maps, the provider and region may thus change from one default to another on a later visit, when the event time becomes in range of the more-preferred provider.KML (event overlay) controls
Setting imagery date/time rangeThe Show imagery for widget set controls what date/time to try to show imagery for. Clicking on the date/time button will pop up a calendar/time widget, and the date and time can be selected. Clicking a day selects that day, but does not yet apply it; clicking a time selects that time and also applies it (closing the dialog). Closing the dialog (by pressing Esc or clicking a time) applies the selected date/time, which now becomes the (approximate) center of the desired imagery range. On the calendar, the current date is indicated with a small home icon, and the event date (if defined) with a small star.
Desired vs. displayed time rangeNote that while Show imagery for and Duration together select the desired date/time range, the actual displayed range may differ, depending on the current data available for the selected provider and model. If the selected range is partially or wholly beyond available data, the displayed range will be altered to fit. Looking at where the home/star time ticks are (see below) can quickly indicate whether the displayed range matches the desired range.Stepping/animating imageryThe horizontal slider controls stepping through imagery a frame at a time; [ and ] (square bracket) keys can also be used. (Curly braces -- { and } -- also step, but wrap around at the end of the range.) The slider handle is centered at the date/time being displayed, and its width indicates the image interval (time step to next image), relative to the time tick marks or overall range. E.g. for the same 5-hour range (duration), 1-hour interval images will show a narrower handle than 3-hour interval images. Thus a wider handle can indicate coarser/larger steps between forecast times. The time ticks are labelled with time, and/or day of month (with month or weekday -- changeable in settings), by default in UTC (also changeable in settings). Day-only labels imply a time of 00:00. Ticks are always aligned with image times, even if those are not integral (e.g. GOES satellite shows time ticks like 18:01). However not all images may have ticks (e.g. if there are many images for a large duration range). Note that sometimes the displayed range will have two different image intervals; late in a forecast's available range it may switch from e.g. 3-hour interval images to 6-hour intervals. The handle will stay the same size, while the time ticks will generally be denser for the longer-interval portion of such a range, indicating where the interval change occurs. Just below the imagery slider, two icon ticks may appear to identify specific points in time:
Progress/status barsWhile loading imagery, a grey progress bar appears beneath the slider, showing the fraction of images completed (whether successful or not). The bar disappears when loading is finished, so it may not be visible if loading is rapid. The inside of the slider span serves as a status bar: images that successfully load have their slider location colored green, failed images show red, and pending images white. So a mix of loaded and failed images can produce a green and red stripe. Images can fail to load for various reasons, besides just connectivity problems. A common reason is that images for the failed dates are simply not available yet. The GUI makes an educated guess as to when each model run or satellite image is available, but it could be wrong, e.g. there is a delay in the provider's processing. Failed images will automatically retry; there is no need to continually refresh.Animation buttons
Help and settings
|
|