As temperatures rise in the spring, snow and ice that have accumulated throughout Long Island Sound’s watershed, which extends far north into Vermont and New Hampshire, begins to melt. This leads to high levels of runoff into small streams and rivers which, in turn, drain into the Connecticut River, which provides about 70% of the fresh water input into Long Island Sound, as well as other smaller rivers. This process is called the spring ‘freshet’.
Changes in the timing of the freshet may have implications for some aquatic species and human activities along the coast. Flooded fields and marshes along the river during the freshet provide critical feeding areas for migratory waterfowl. So if the freshet comes earlier, waterfowl could be impacted if they do not adjust the timing of their migration. Changes in the timing of flooding may also provide a competitive advantage to invasive plants (such as purple loosestrife and Phragmites) in the marshes since some of these species emerge earlier than the natives. In the past, these invasives were flooded in early spring and often rotted due to submergence for prolonged periods. So, if the flooding occurs earlier, the invasives (still emerging before the natives) will no longer rot in early spring and may gain a competitive advantage over natives.
By looking at 80 years of river data, scientists at the US Geological Survey and UConn have determined that the spring freshet is occurring earlier in the spring. This indicator is derived from measurements of river flow at a gauge at Thompsonville, CT, near the Massachusetts border). The indicator is the date (we use Julian days, or # of days into the year, to account for leap years) that the total volume of water that has passed by the gauge exceeds half of the total for the year. The critical date is called the “winter-spring center of volume” or WSCV. While spring weather in New England is quite variable, the WSCV usually occurs in late March or early April.
The blue line in the graph uses a smoothing function called a LOWESS regression to take out some of the yearly variability by computing a weighted average WSCV for each year that considers several years of data on each side of that year, but weights the closest years most heavily. This gives us a better picture of short- to medium-term (5–10 years or greater) patterns in the WSCV.
Although this analysis is relatively modern, the USGS Thompsonville gage dates back to 1929, and we can calculate WSCV for this entire time period, which means that this is one of our longest running climate indicators. Particularly when looking at weather and climate, which have a great deal of annual and decade scale variability, long datasets (75–100+ years) are essential to understanding long-term patterns, which makes this one of our most valuable indicators of changing climate.
Despite large oscillations, the freshet is getting to Long Island Sound on average about 10 days earlier than it did a hundred years ago. While the exact magnitude and timing of the freshet in any given year is highly dependent on local and regional weather patterns during the late winter/early spring period, the long term shift towards an earlier center of volume is indicative of a general warming trend throughout the region.