Mid-winter water supply update

As we are now about half-way through our snow-accumulation season and half-way through the critical winter period for survival of juvenile trout, it is a good time to assess our current water supply and what it may mean for the upcoming spring and summer.

Reservoir contents

As a result of heavy late-summer rain, which increased water supply and decreased irrigation demand, the upper Snake River reservoir system began the 2015 water year at around 44% of capacity, a little above average. Island Park Reservoir content started the water at 70,000 acre-feet, a little over 50% of capacity and 16% above its average value on October 1. Rate of fill across the reservoir system as a whole has exceeded average so far this winter. The system currently contains 2.7 million acre-feet, 67% of capacity and about 20% more than average for this time of year. Despite this rapid fill rate—around 10,000 acre-feet per day—roughly 1000 acre-feet of water per day has been available downstream of the reservoir system for recharge of the Eastern Snake Plain Aquifer. Filling the reservoir system and recharging a large amount of water to the aquifer in the same winter is a double water-supply bonus and will contribute to increased water supply for all resources, including fisheries. Closer to home, Island Park Reservoir is currently at 77% of capacity and increasing at a rate sufficient to fill it in early April.


Flow in the creeks and rivers of the upper Snake River basin is currently approaching its lowest value of the year, what we refer to as “baseflow” in hydrology. Stream baseflow is maintained by groundwater inputs and shallow subsurface flow and is thus a good indicator of long-term trends in water supply. Right now, it is actually kind of difficult to assess streamflows, because most stream gages in the upper Snake River basin are affected by ice and are therefore not recording accurate measurements. However, the high rate of system-wide reservoir fill indicates that streamflow is good overall, reflecting above-average snowpack across much of the basin last spring and the late-summer rains.

Unfortunately, streamflows in the upper Henry’s Fork are not as high as streamflows elsewhere in the Snake River basin. Inflow to Henry’s Lake so far this winter has averaged 37 cfs, 14% lower than the 35-year mean of 43 cfs. Unregulated flow in the Henry’s Fork at Island Park Dam (what flow would be in in absence of Henry’s Lake and Island Park Reservoir) has averaged 443 cfs, 16% below the 35-year mean of 528 cfs. Although these streamflows are higher than they were at this time last year and a little higher than I had predicted at the beginning of the water year, they indicate that the upper Henry’s Fork watershed has still not recovered from the effects of low snowpack in 2012 and 2013, despite a 2014 snowpack that was near average.

Since December 1, outflow from Island Park Reservoir has averaged 249 cfs, compared with a long-term mean of 347 cfs. This is primarily a function of inflow; if inflow were at its long-term mean, outflow would be around 334 cfs, given the same contents and management targets for Island Park Reservoir. Thus, although the reservoir started the water year with higher-than-average contents, below-average inflows have limited the amount of water available for release. The current plan is to release about 260 cfs for the remainder of the winter, which will yield a December-February average outflow of 255 cfs, 30% lower than the average of 365 cfs but 28% higher than the 200 cfs outflow we received during the 2013-2014 winter. As a result, we expect more trout to enter the fishable population in the spring of 2016 than will enter the population this spring, even though we predict that recruitment (entry of new trout into the fishable population) will be below average in both years.  

Precipitation and Snowpack

Total precipitation since October 1 is currently 103% of average across the entire upper Snake basin and 98% of average in the Henry’s Fork watershed. However, water-year-to-date precipitation is currently lower than this at the three Snotel sites in the Island Park vicinity: 94% of average at Black Bear (on the Continental Divide, east of Big Springs), 88% at Island Park (near Ponds Lodge), and 82% at White Elephant (on the east side of Mt. Sawtelle). Snow water equivalent (SWE, the amount of water contained in the snowpack) follows similar patterns; SWE is currently 127% of median across the upper Snake basin and 109% of median in the Henry’s Fork watershed. However, SWE is currently only 93% of median at Black Bear, 99% at Island Park, and 92% at White Elephant. Worse, SWE is currently only 78% of median at the Crab Creek Snotel site, which is located in the Centennial Range, west of Island Park. Some analysis I did a few months ago showed that SWE at the Crab Creek site is a statistically significant predictor of spring runoff into Island Park Reservoir, so low SWE values at Crab Creek are indicative of low inflow to the reservoir next spring.

Note that the precipitation values are reported relative to the 30-year (1981-2010) average (the mean), whereas the SWE values are reported relative to the 30-year median. Remember from your high school or college math classes that the “average” or “mean” of a set of values is the arithmetic average (add all the values and divide by the number of values), whereas the median is the value in the middle of this set, when it is ordered from smallest to largest. This is an important distinction, because for data sets pertaining to hydrologic variables, the median is almost always less than the mean. For example, the median value of total water-year precipitation at the White Elephant site is 3.1% less than the median. I’ll skip the mathematics lesson for now but may revisit it in a future blog. What is relevant is that if SWE is, say, 92% of median, it is only about 88% of the mean. In other words, our current SWE values range from about 75% of average at the Crab Creek site to 96% of average at the Island Park site.  

Summary and Outlook

Taken together, the various water-supply indicators and their trends show that the upper Snake basin as a whole is recovering from the drought of 2012 and 2013. Some of that recovery is due to heavy rainfall during August and September of 2014, which greatly reduced late-season irrigation demand, albeit at the expense of quality and quantity of grain and hay crops. The recovery has been a little slower in the upper Henry’s Fork watershed, which did not receive as much precipitation in 2014 as other areas to the east and south. In composite, water supply in the Henry’s Fork watershed is currently about 15% below average, despite above-average reservoir storage. In fact, increased carryover storage in Henry’s Lake and Island Park Reservoir is due as much to increased supply and decreased demand elsewhere upper Snake basin as it is to supply and demand trends within the Henry’s Fork watershed. This provides one of many examples of why and how water management in the Henry’s Fork watershed is tied to water management across the entire upper Snake River basin.

At this point in the winter, SWE accumulation has reached about half of its annual maximum, which occurs around April 1 at most sites. Thus, at sites where SWE is currently around average for this time of year, average precipitation for the remainder of the winter will result in average SWE on April 1. At sites in the upper Henry’s Fork watershed, where SWE is currently around 90% of average, we will need precipitation to be around 110% of average for the remainder of the winter to achieve an average SWE on April 1. Current forecasts call for equal changes of above-average and below-average precipitation between now and then. Assuming average precipitation through March, SWE will end up at about 95% of average at the end of the winter in the Henry’s Fork watershed and at about 110% of average across the upper Snake basin as a whole. Given current streamflows, this amount of snow will be more than sufficient to fill the reservoir system in the spring or early summer. However, the amount of storage be required to meet irrigation demand will depend greatly on how much precipitation we receive in April, May and June, which is beyond the range of current forecasts.

Current water-supply indicators and weather forecasts suggest that slow but steady improvement in water supply will continue in the upper Henry’s Fork watershed through the spring, and this improvement is predicted to increase trout recruitment in 2016 over what we anticipate will be relatively poor year-class in in 2015. At this point, very high runoff in the upper Henry’s Fork and Fall River drainages does not appear likely this spring, but our certainty in that forecast will greatly improve in about 7 weeks, when I will next update the outlook.