River Alliance of CT > Priority Topics > Streamflow > Streamflow and Groundwater Overview in CT

  Know Your Flow!
   We update this webpage with the latest information weekly.

Paddling or fishing? Be careful out there!

Click here for DEEP's guidelines on spring boating safety.

Link to webpage
Rivers Alliance thanks Loureiro Engineering Associates, Inc. for their generous sponsorship of the Know Your Flow page.

CT Conditions At A Glance

(Click on any graphic below for more information)
Streamflow in CT Now
(click map to go to the data page)

 [color code for]  color code for  color code for color code for  color code for  color code for  color code for               [color code for]       
   High Flow     Low Flow        Not Ranked
CT Current Streamflow. Click to go to source page.
US Drought Monitor Map of CT
(click map for details)
Current US Drought Conditions. Click to open                  source page.
The U.S. Drought Monitor is jointly produced by the National Drought Mitigation Center at the University of Nebraska-Lincoln, the United States Department of Agriculture, and the National Oceanic and Atmospheric Administration. Map courtesy of NDMC-UNL.
Map of Below Normal 28-Day Average Streamflow

Recent Flow At a Typical CT Stream
(click graph for details)

Link to Burlington Brook flow data
Groundwater Level
(click graph to see more well data)
USGS monitoring well. Click to see more.
Check your forecast here
(click map to go to state weather page)
Link to NWS graphic page
Graphic courtesy NOAA showing CT area watches and warnings if any.

April 28, 2017 Update

Summary (details follow)

US Drought Monitor analysis released April 27 shows the area of Moderate Drought reduced to 28% of the state, down from 56% last week.  This change makes the Abnormally Dry area now 56% (it was 33% last week). The eastern 13% of Connecticut is in the No Drought category, up a little from 11% last week. The worst drought since the historic drought of the 1960s continues.

The NE River Forecast Center is not predicting any flooding in CT for the next five days.

Groundwater level showed a quick improvement after heavy rains but many monitoring wells are now dropping back close to or below normal.

The Interagency Drought Workgroup (IDW) met on March 8 but has not yet published any recommended changes to the state's drought alert level. The Drought Watch continues for six CT counties; Drought Advisory continues for the eastern two counties. The state is still requesting all residents to reduce their water use.

Streamflow Discussion

Only Up through the first week of this month, streamflow increased in almost all rivers and streams due to decent rainfall. In general it has fallen back since then to steady levels close to average for this time of year. Some stream gages are showing flow patterns quite different from the rest, most likely due to changes in releases from dams. Even the Saugatuck River, that was setting low-flow records since November, is now above normal. See below for a discussion of what a normal streamflow pattern looks like.

The 28-Day Average Streamflow map from the USGS now categorizes all of Connecticut as normal or higher.

In November, fishermen reported long stretches of rivers with no catch at all. How well river ecosystems recover from the drought will depend on flow from groundwater. Where groundwater is near normal, streamflow should remain healthy between storms unless rainfall patterns change as they did in the past four or five years. 

Click here for the USGS Connecticut stream gage web page with graphs of flow from every gage for the past 30 days compared to their daily average flows.

NOTICE (03/30/2017) -- Data collection at the following gages may be discontinued on June 30, 2017, due to funding reductions from partner agencies. Although historic data will remain accessible, no new data will be collected unless one or more new funding partners are found. Users who can contribute funding for the non-federal share of costs to continue operation of these stream gages should contact Jon Morrison at the USGS New England Water Science Center - Connecticut Office (860-291-6761) or email at jmorriso@usgs.gov.

North Branch Park River at Hartford-----_-52 years of records
Rooster River at Fairfield------------------39 years of records
Mill River near Fairfield--------------------44 years of records

Public Water Supply Information

Your water utility and town web pages are the best sources of information on water use restrictions that apply specifically to you.

The CT Department of Public Health (DPH) Monthly Reservoir Status Summary reports that at the end of March capacity was at 97% of normal. Note that "normal" means the average capacity reported at the end of each month over many years. The March average is 96% of total capacity. The 97% of normal is based on reports from 34 large water systems that use reservoirs. That percentage means those reservoirs averaged 93% of overall capacity. Nineteen of the 34 systems reported no restrictions on water use. "Voluntary Water Restrictions" is noted for three systems; "Water Supply Advisory" for four systems; one has a "Water Supply Watch"; "Water Supply or Drought Emergency" is listed for five systems; one has a "Water Supply Warning"; and one has declared a "PHASE 2 Drought Alert." The Monthly Reservoir Status Summary is one of several links on DPH's Reservoir Data monthly web page. 

Aquarion maintains on its website a weekly updated water usage report that includes customer usage trends, system transfers, system capacities, days usable storage remaining, and precipitation. Their graphs and charts indicate conditions are near or even above their trigger levels for water usage alerts.

Greenwich, Stamford, Darien, and New Canaan are working with Aquarion on a water-demand study They are examining the overuse of water that, combined with years of below-average rain and snow, has caused a statewide drought. The study could result in permanent water conservation guidelines.

Groundwater Discussion

On Apr 28, one of the seven real-time monitoring wells run by the USGS with multi-year records was back below 75% of its records for the date, though it is not breaking low water records. Three of them are higher than 75% of their records, three are near normal.  The first week of April was the first time all of them were finally reporting near- or above-average levels, after a year of low groundwater measurements.  March 30 was the first time in months that none of the seven monitoring wells were breaking records for lowest daily average for the date. The water table's lowest levels occurred in September, with more stations breaking low-water-level records than those that were merely very low until November.

As of Apr 28, 7% of the most recent measurements at the 70 wells on the Groundwater Watch website were much below normal, though they were all last measured on Mar 30 and may show better numbers next week. This is a distinct improvement from April 13 when 21% were much below normal. None had measurements lower than any previous year's measurement at those wells going back 16 to 20 years. 

14% were rated below normal on Apr 28, down from 19% on April 13. 74% are now near normal, two of the wells are above normal, one is well above normal. On April 13, 70%  were near or above normal.

The water level in streams and rivers between storms depends on the flow of water out of the ground into the stream channel. The higher the water table, the more water is available to keep streams flowing. During a drought, smaller upland streams dry up faster than the larger rivers (details below below). An occasional brief, powerful rainstorm does not do much to help groundwater levels rise. The rain falls faster than the ground can absorb, and thus much of it becomes stormwater runoff. In that situation, public water supply systems that depend on reservoirs are affected less than those with wells because the runoff is collected by the reservoirs. How much groundwater is contributed by snow depends on several factors including slope, rate of snowmelt, and whether the ground beneath the snow is frozen or saturated.

graphThere has been an apparent pattern for the last five years in which normal groundwater levels in winter and spring have alternated with well-below-average in summer and fall. Click the graph to the right for a larger version in a new tab of a modified USGS graph. There are two trends of great concern. One concern is that the below-normal levels in summer and fall got worse in each of those years. The other concern is that the water table has not recovered yet this year as it did in previous years. Five years is not long enough to make any statistically valid conclusions about climate, however, and those station records only go back seven to 14 years. Click here to see all the USGS graphs for their real-time groundwater measurements that compare levels from the last four years to the median level for each day of the year (lots of data, so it takes a long time to load). USGS manually measures other wells once a month or so, and a quick look at those records verified the possibility that this pattern may apply to all our groundwater.

The USGS Groundwater Watch website is an excellent source of data for examining historic and current trends. The map of CT (example below) is color coded the same way the USGS Connecticut streamflow map is colored to show how the most recent measurements compare to historical records for each station. Graphs can be made to chart water levels for all 70 wells.

map of CT gw, link to website

Precipitation Patterns

As of April 28th the National Weather Service Advanced Hydrologic Prediction Service reports that the 30-day precipitation totals across Connecticut ranges from 25% below normal in western parts of the state to 25% above normal in eastern Connecticut. Western CT is now showing below-normal precipitation totals for 7, 14, 30, and 60 day time scales. Eastern CT is in better shape due to getting more rain recently than the western side.

Periods of unusually heavy precipitation do not compensate for periods of dry weather even if they average out to normal because heavy rain does not all soak in to recharge groundwater. Go to this link for detailed maps and data.

For most of 2016, storm tracks brought precipitation out to sea or north of us because a persistent high-pressure system called a blocking pattern was diverting weather fronts and storms. This was the same bad situation as in previous years (see below), caused not only by less total rainfall than normal over a long period of time but also by the intensity of the rain we did get. Not all drought conditions went away in winter 2015 and were worse this past winter. Generally, as our climate changes, heavy precipitation events are becoming more frequent, but heavy rain does not recharge groundwater aquifers as much as the same amount of rain over longer periods would. This means that even when long-term total rainfall returns to normal, streams and rivers could still be afflicted with drought conditions because between storms the flow will drop to low levels due to a lower than normal water table.

Some studies relate recent extreme weather events to an upper-atmospere situation during which the jet stream becomes confined into a stable pattern of alternating north and south flows resembling waves when viewed on a map. Normally the jet stream meanders in direction so that the storms and dry areas that are steered by the position of the jet stream travel west to east along more or less random paths. When the jet stream gets locked into position, however, the storms and dry areas follow the same paths over and over. If the position of the these persistent jet steam waves shifts, areas that were missed by most storms start getting one storm after another and vise versa. There is speculation that if this is a new atmospheric phenomenon, it may be caused by a shift in wind systems from the loss of Arctic sea ice that is causing unprecedented warming in the north. For more information, see these articles about the studies: Scientists Link California Droughts, Floods To Distinctive Atmospheric Waves (2017-04-06), Extreme Weather Events Linked to Climate Change Impact on the Jet Stream (2017-03-27).

For other related climate news: Climate Study: More Intense and Frequent Severe Rainstorms Likely, No Drop Off Expected (UCONN, 2017-03-07), Climate Change Threat to Both the Natural World and to Human Civilization.

Rain? The National Weather Service's seven-day forecasts predict above normal rainfall, which should allow streams to show a healthy pattern of varying day to day from above to below average. If it does not rain, streamflow will obviously fall, but where the water table is near or above average it should stay at healthy levels. Streamflow could drop back to unhealthy levels where the water table is below normal. 

Long-range (16-day) computer models are also predicting above-normal rainfall due to storms tracking over us. A persistent shift in storm tracks to our north and south contributed to our drought conditions over the last few years (see discussion above), so monitoring storm tracks can help to determine if our drought will intensifying over the next few months as it did in previous years. As long as rainfall remains near normal, our streams can continue to recover from drought conditions, but not all parts of the state have the groundwater reserves necessary to maintain healthy between-storm flows should storm tracks again begin to consistently miss us. A persistent pattern that brings all storms over us could actually lead to flooding problems.

Each month, the NWS Climate Prediction Center issues analyses of global atmospheric conditions that could influence general temperature and precipitation patterns. Their April analysis for trends through August show equal chances for above-normal, for normal, and for below-normal precipitation. Above-normal temperatures are a little more likely than normal temperatures. This is pretty much the same three-month forecast we have seen over the past year, but actual precipitation was much lower than normal. Before the April analysis, for three months in a row, their prognostications for the summer gave a 40% chance for above-normal rainfall, a 33% chance for normal, and a 27% chance for below normal precipitation. That prediction has now shifted to the fall.

Lowest Baseflow Streams and Rivers in Connecticut

Very variable rainfall this spring has greatly affected our count of the number of low-flow rivers*.

Only 2 stream gages, the Mill River near Fairfield and the Housatonic at Stevenson were showing low flow* on the afternoon of April 28. The Housatonic River was at normal earlier in the day when the Stevenson Dam was generating power. Having single digit numbers of low flow rivers is expected to continue due to predicted normal or above normal rainfall.

Last week, on April 19, there were 12 USGS stream gages reporting low flows* in Connecticut (see list below) out of 62 (19%)  To see the status of any them right now, click the station's URL in the chart below. Streamflow had generally fallen steadlily since April 13 when only the Saugatuck River below the Saugatuck Reservoir, and Burlington (Bunnell) Brookk had low flows*. The Housatonic River at Stevenson Dam was briefly at its 17th percentile for part of that day. On Apr 7, just the Saugatuck had low flow.

On March 30, eight of 62 (13% of) Connecticut rivers were experiencing low flows*.  On March 23, 44 of the 72 (61%) stream gages listed by the CT office of the USGS at this link were reporting flows lower than 75 percent of all the daily averages for the date for those gages. This is a little better than on March 9, when we found that 47 of 62 (76%) Connecticut rivers and streams from a slightly different list were experiencing low flows*. There were 12 the previous week. For more information on that sudden jump in the low-flow list, please see the IDW freshet discussion from Jan. 20.

Before March 30, we last saw single-digit numbers of low-flow streams on February 24, when there were seven low-flow rivers on our list, the same number as the previous week. That was a significant improvement from February 14 when there were 23, and from February 6 when there were 34 low-flow streams. In January the number ranged from 27 to 35. During most of 2016, there usually were 50 to 60 low-flow streams; on December 23 there were 53.

All low-flow rivers raise a concern about the recovery of their ecosystems from drought conditions. This is especially true when the flow is lower than any record for the date, such as the Saugatuck River, the only that was one breaking records on April 13 and 19. There were two record-breakers March 23, an improvement from the seven on March 16 and the five on March 9. On March 1, only the Saugatuck River stream gage was reporting flow lower than any record for the date, as it has since November.

In most of February, there were only one or two record-breaking low-flow rivers, an improvement from Feb. 6 when there were seven record-breakers. In January the number of record-breakers ranged from five to nine. The number of record breakers were much higher for most of 2016; for example, there were 15 on Dec. 23.


April 28, 2017 Low Flow Rivers
Station Name
Station Website
HOUSATONIC RIVER AY STEVENSON https://waterdata.usgs.gov/ct/nwis/uv?site_no=01205500
MILL RIVER NEAR FAIRFIELD, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01208925


Last Week:

April 19, 2017 Low Flow Rivers
Station Name
Station Website
BUNNELL BROOK NEAR BURLINGTON, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01188000
HOCKANUM RIVER NEAR EAST HARTFORD, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01192500
HOUSATONIC RIVER AT FALLS VILLAGE, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01199000
HUBBARD RIVER NEAR WEST HARTLAND, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01187300
LITTLE RIVER NEAR HANOVER, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01123000
MILL RIVER NEAR FAIRFIELD, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01208925
NATCHAUG RIVER AT WILLIMANTIC, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01122000
SAUGATUCK R BELOW SAUGATUCK RES NR LYONS PLAIN, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01209005
SAUGATUCK RIVER NEAR WESTPORT, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01209500
SHETUCKET RIVER NEAR WILLIMANTIC, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01122500
STILL RIVER AT ROBERTSVILLE, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01186500
YANTIC RIVER AT YANTIC, CT https://waterdata.usgs.gov/ct/nwis/uv?site_no=01127500

*We define low flow as below the 25th percentile for that stream, OR below 25% of the mean OR below 25% of the median flow for that stream for the date from the USGS data file at: https://waterwatch.usgs.gov/webservices/realtime?region=ct&format=xml.

Click here for a list of all CT stream gages comparing their flow right now to their mean and median flows for today's date.

Streamflow Graphs and Storms

Click to enlarge. Streamflow graphics courtesy USGSThe bottom graph to the right illustrates what can be considered an average pattern of streamflow, in that the stream shows a flow that peaks above its average for the time of year but then drops to below average. The streams that maintain healthy flows between storms are those draining the parts of the state where there has been decent rainfall or those controlled by managed dams.

How quickly the flow drops after each rainstorm, and how low it drops, is unique for each stream, because it depends on how much water ran off the surface compared to how much soaked in to increase groundwater levels. If the watershed of the stream has not received average long-term rainfall, the low water table will allow the flow between storms to drop to unusually low levels. In watersheds with lots of impervious roads, roofs, parking lots, patios, and compacted soil in lawns, a lot of the rain typically flows over the surface or through storm drains directly into the streams, leading to flooding problems far more often than in watersheds with mostly natural surfaces. Their streamflow then drops quickly after the storm ends, and the flow levels off far lower than in streams whose watersheds have more natural surfaces.

The U.S. Drought Monitor publishes a weekly analysis of drought conditions across the entire nation based on a variety of types of data that include streamflow. The report comes out every Thursday based on data from that Tuesday. The U.S. Drought Monitor is produced by the National Drought Mitigation Center, which is also on Facebook.

Their analysis released April 27 shows the area of Moderate Drought in Connecticut reduced to 28% of the state, down from 56% last week.  This change makes the Abnormally Dry area now 56% (it was 33% last week). The eastern 13% of Connecticut is in the No Drought category, up a little from 11% last week.

Their narrative says: "Widespread showers continued the (northeast) regionís slow recovery from long-term drought ... Although short-term dryness is currently not a concern, slower-to-recover drought impacts remained. In particular, groundwater levels remained unfavorably low (10 to 25th percentile, locally below the 10th percentile) in the regionís lingering Abnormal Dryness and Moderate Drought areas, and additional soaking rainfall will be needed to recharge the regionís aquifers as water demands increase in response to seasonally warmer weather."

The "additional soaking rainfall" they refer to would actually have to all soak in to help.

From the end of January to the beginning of April, Connecticut had the dubious distinction of being the most drought-afflicted state of all 50 based on percentage of area. Click here for details from an archived Know Your Flow! page.

AsAs the graph below shows, Connecticut has been labeled Abnormally Dry, Moderate Drought, or worse pretty often over the last few years. This graph was made before any of the state was rated as Extreme.

How does the current drought compare to historic droughts in Connecticut? This graph is from NOAA's National Environmental Satellite, Data, and Information Service (NESDIS), displaying the Palmer Hydrologic Drought Index (PHDI) for our state for the past century (click on it for a larger view in a new tab). Since groundwater and streamflow did not fully recover from the annual return of drought conditions of the past few years, we can consider our current drought as starting in 2010 or 2012. The overall length of time we have had these intermittent drought conditions is comparable to the continuous drought of the 1960s. At times in 2016, the severity of the current drought approached the maximum severity in the 1960s. It is therefore fair to say the current drought is one of the worst in 100 years. Area weather forcasters have been reporting that the current drought is the worst since the 1960s.  


Are small and large streams affected by dry conditions equally?

This graph plots the rate of flow on September 4, 2015 as a percentage of the median for each stream gage. For example, a dot at the 40% line means that stream that day was flowing at 40% of what could be considered normal for that date. The few rivers and streams flowing at or above 100% of their median rate are mostly those with flows managed by dam releases.

Click here for a search of news articles about the drought


NASA's climate monitoring programs to be axed?