About Us - Geyser Observation and Study Association
GOSA Store
Geyser Observation and Study Association - Transactions


Geysers of the World   

Geysers of Yellowstone   



  Grotto Geyser
Feature Type: Geyser
Geyser/Spring Type: Cone geyser

Upper Geyser Basin
Grotto, Riverside, Chain Lakes

Grotto Geyser is the namesake of the group. It was named and first reported on by the Washburn expedition of 1870. While not tall, the maximum height, usually reached at the very start of the eruption, is only 30-40 feet, the eruption can last for a fairly long period, expelling large quantities of water. Grotto is a fountain-type geyser. The pool through which it erupts is hidden inside Grotto's unique sinter formation. The unique formation is the result of sinter coating old tree stumps that stand around the pool. The result is a shell like formation that is not seen anywhere else in the park. Grotto's sinter formation connects directly to the large raised rim of Rocket Geyser. Both geysers erupt simultaneously.

The duration of the eruption can be anywhere from one hour to 26+ hours. Eruptions are often classified as one of two types. Short Mode eruptions usually last for 1-2 hours but can last longer. These eruptions are usually but not always concluded by a major eruption of Rocket Geyser. About 3/4ths of Grotto's eruptions are short mode eruptions. Long Mode eruptions, sometimes called Marathons, generally last for 6-12 hours but occasionally can last considerably longer. These eruptions usually do not include a major eruption from Rocket Geyser but do include eruptive activity from Spa Geyser. Also, during Long mode eruptions, the water level in Marathon Pool often drops a few inches. When the Short and Long mode classifications first came into common use, intermediate length eruptions, 3-6 hours, were very rarely seen. Since then, these intermediate eruptions have become more common. No consensus has yet been reached on how to classify these intermediate length eruptions.

Grotto's interval depends on the duration of the previous eruption. The interval for short mode eruptions is usually around 7 hours. The interval for long mode eruptions can be over a day. The longer the previous eruption, the longer the next interval. Fortunately, even if you don't know the length of the pervious eruption, with a little practice, it is usually possible to determine if the start of an eruption is imminent.

What to look for:
In recent years, most eruptions of Grotto Geyser have been preceded by Grotto Fountain Geyser. Thus, look at the information for that geyser to understand what to look for. Even if Grotto is not preceded by Grotto Fountain, the things to look for are still basically the same.

Predicting Grotto Geyser:

Lew and Jan Johns have been studying Grotto Geyser for a number of years. They have determined that you can predict the next eruption of Grotto based upon the duration of the current eruption. Follow this link to read their report on using eruption durations to predict Grotto geyser. Their results work quite well. Their 1997 linear regression formula for predicting the Interval Between Eruptions (IBE) was:

IBE = 190 + 1.7D

Where: IBE is in minutes and D is the duration in minutes.

Electronic Monitor Files
Grotto eruptions for 2000.txtGrotto eruptions for 2001.txt
Grotto eruptions for 2002.txtGrotto eruptions for 2003.txt
Grotto eruptions for 2004.txtGrotto eruptions for 2005.txt
Grotto eruptions for 2006.txtGrotto eruptions for 2007.txt
Grotto eruptions for 2008.txtGrotto eruptions for 2009.txt
Grotto eruptions for 2010.txt 

Some of the temperature data used to derive the eruption times and durations used in this section were collected by Ralph Taylor under a National Park Service research permit, and the remainder was collected by personnel working for the Geology Department of the Yellowstone Center for Resources (including Ralph Taylor). The loggers are a combination of loggers owned by the NPS and Ralph Taylor. Analysis of the raw temperature data to extract the eruption data was performed by Ralph Taylor. The eruption time files on this website may be used provided that Yellowstone National Park is credited for the temperature data and Ralph Taylor is credited for the eruption times.

Activity Recorded by Data Logger - by Ralph Taylor  

Grotto Geyser has been monitored electronically since 2000, and was monitored at times before that, including the 1997 summer season. My data for 1997 covers the span 22 July to 14 October, with several gaps. Thermal volunteers Lew and Jan Johns also studied Grotto Geyser in 1995-7; their data can be found at http://www.shol.com/johns/web1-2.htm. The 1997 data described below and begins after the Johns' data for that year.

The form of the data presentation for Grotto differs from the presentation that I used for many of the other geysers because Grotto has some unique characteristics. Grotto has two types of eruptions, short eruptions and marathon eruptions. Short eruptions last about 1h30m and are followed by intervals of about six hours. Marathon eruptions have a much longer duration, some in excess of 24 hours, and are followed by a quiet period of a full day or more.

Since Grotto's eruptions fall into two groupings, much of the analysis that follows treats the non-marathon and marathon activity separately. It could be argued that the marathon eruptions are no different from the non-marathon eruptions other than lasting longer; on the other hand, there is a often a pattern of eruptions consisting of several short eruptions followed by a marathon, and marathon eruptions do deplete the energy not only from the Grotto group but also from the Giant group. At any rate, I did separate the eruptions by duration and provide statistics for the short and marathon eruptions separately.

In the analysis that follows I have chosen 4h50m duration as a dividing point between marathon and non-marathon eruptions. I initially used 3h30m based on an apparent gap in durations and the shape of the duration frequency distribution. However, analysis by Tara Cross presented in her paper on Giant Geyser in GOSA Transactions Vol X used 4h50m (electronic) to separate marathon and non-marathon eruptions based on observation of Grotto and related geysers.

It has been suggested by Paperiello that activity by Spa Geyser be used to distinguish marathon eruptions, but in this analysis I have only the electronic data for Grotto, so cannot use that criterion for this analysis.

Marathon eruptions, at least those that continue long enough, also cause visible changes in the Giant Geyser group, but again, that information is not available in the data set that I have for this analysis.

Activity in 2010  
The interval and duration graphs shown here simply plot all of the intervals (time from one eruption start to the next eruption start) and duration (time from the start to the end of the eruption) along a timeline.

The first graph shows the intervals for all of 2010 to date and will be updated as new data are obtained. The second graph shows the activity for the most recent three month period.

Click for a larger image

The graph at the right shows Grotto's activity for the last three months.
Click for a larger image

The interval histogram at right shows the distribution of Grotto's intervals. The intervals below 4h50m represent the non-marathon eruptions. Note that the marathon intervals are fairly evenly distributed with little obvious clustering. Note that in this and the other histograms displayed here the labels shown on the X-axis represent the upper boundary of the class, not the midpoint. Geyser times are traditionally truncated.

The graph at the right has class widths of one hour. The bar appearing above the label "18:00," for example, contains intervals from 17h01m through 18h00m .

Click for a larger image

The next graph shows Grotto's durations for 2010.
Click for a larger image

The next graph shows the durations for the most recent three months. The pattern of non-marathon and marathon durations is more clearly seen on this graph.
Click for a larger image

Finally, the duration histogram shows the distribution of the intervals for 2010.
Click for a larger image

Activity Since 2000 (and in 1997)  
The interval and duration graphs shown here simply plot all of the intervals (time from one eruption start to the next eruption start) and duration (time from the start to the end of the eruption) along a timeline.

The interval graph shows the range of intervals, and reveals some changes in Grotto's behavior. First, note the gap at intervals around the 12 hour line. This represents the break between non-marathon and marathon eruptions. The actual distinction is made based on duration, but the break is also visible here. Second, note that until early 2004 there were no intervals below three hours, but since then, and especially in 2005, there were several intervals well below three hours.

Lastly, note the marked increase in long marathon intervals (those in excess of two days), beginning in late 2005 and continuing in 2006, 2007, and 2008.

Click for a larger image

The duration graph shows similar trends (not surprising, since Grotto's intervals are known to correlate with the preceding duration). One notable trend is the decrease in the duration of non-marathon eruptions, some of which have had duration of only a few minutes.

For this analysis I originally chose seven hours of duration as the dividing point between non-marathon and marathon eruptions. I used the seven hour criterion for the 2000 and 2001 data. Beginning in 2002, I changed to a six-hour break since the duration distributions since 2002 have had a region around six hours was the bracket with fewest members. Finally, in 2006, after discussions with other observers and after looking at the data further, I moved the dividing point to 3h30m duration, since that seemed to be a more appropriate choice based on behavior of the Grotto-Giant system.

The choice of duration to mark the separation between non-marathon and marathon is somewhat arbitrary, but is made on the basis of the annual duration distributions, which always show a break at some duration.

Click for a larger image

The next four charts show the change in monthly statistics for both non-marathon and marathon eruptions.

For this analysis, as noted before, all eruptions with durations over 3h30m are in the marathon eruption data set. The marathon interval chart gives the intervals from the start of one marathon eruption to the start of the next marathon eruption, not the next non-marathon eruption.

The graph of monthly minimum, mean, median, and maximum non-marathon intervals reveals that the average (both mean and median) interval between non-marathon eruptions has remained steady at right around six hours. The range between minimum and maximum has begun to increase beginning in early 2005. There were no intervals below three hours detected before June of 2005, but in the last half of 2005 numerous very short intervals were recorded.

Click for a larger image

The graph of the monthly statistics for non-marathon eruption durations shows similar characteristics. Again, the average has remained around the 90 minute range, but since early 2004 has gradually declined. The maximum is, of course, limited by the choice of 3h30m as the upper bound of non-marathon eruptions. Perhaps the most notable change is the decline in the minimum intervals, to nearly zero starting in September 2005.
Click for a larger image

The next graph shows the statistics for marathon eruptions. For this analysis, the marathon interval is the time between marathon eruptions, not the time from the start of the marathon to the start of the next eruption (which is usually a non-marathon eruption). The reason for this choice is that marathon eruptions are of particular interest because of the connection between the activity of the nearby Giant Geyser group and Grotto Geyser.

During Grotto marathons, the activity on Giant's platform slowly ebbs until a few hours into the Grotto marathon the Giant platform is quiet.
Interesting Giant Geyser activity, including hot periods and eruptions, tends to be related to Grotto marathon activity. It is this relationship that suggested that the marathon interval for Grotto be tracked.

On the average, marathon eruptions occurred about every two days since 2000.
The marathon frequency dropped (that is, the marathon interval increased) during 2003 and 2004, but in the last months of 2005 marathons became much more frequent (and Giant Geyser became more active, which may or may not be related).

Click for a larger image

The final graph of this group is the marathon duration graph at right. Since July of 2003 Grotto's marathon durations have increased slowly on average, reaching over 20 hours for much of 2005 but then declining. Note also that the maximum marathon intervals have increased.
Click for a larger image

The graph at the right shows the percentage of marathon eruptions for each month for which I have electronic data. I added this graph after noticing the increase in the number of marathon eruptions in early 2006.
Click for a larger image

Activity in 2009
Activity in 2008
Activity in 2007
Activity in 2006
Activity in 2005

Click for a larger image

Click for a larger image

This vintage Haynes picture of Grotto Geyser shows the start of a Grotto eruption. The water is not nearly as tall later in the eruption.

Click for a larger image

Click for a larger image

Please note - this site is currently under constuction. Please visit for more information.  Last update 01-29-2017

Home | Contact GOSA | GOSA Store | About GOSA | Transactions | Geysers of the World | Geysers of Yellowstone

Facebook Page Google+ Page YouTube Channel
Designed and Hosted by: Computer Point

2006 COPYRIGHT© 1997-2017 GOSA
All images © 1999-2017. All rights reserved. No form of reproduction, including copying or saving of digital image files, or the alteration or manipulation of said image files, is authorized without written permission