Analysis of 3-day cumulative data
for Chelmsford hydrometric stations
by Mike Parker
updated 2025.09.08
An article presenting analysis of the hydrometric data for the three hydrometric stations in Chelmsford. The analysis examines the cumulative inflow (in millions of cubic metres) over three day periods, identifying an alarming increase over five decades between 1970 and 2020, with the greatest increase being observed on the River Can, a smaller increase on the Chelmer, and no increase on the Wid.
If you like this work, please consider supporting the Chelmsflood project by making a donation,
and if you would like to know more get in touch.
Previously I published my analysis for Rushes Lock hydrometric station, downstream of Chelmsford towards Ulting. The benefit of this station is that this part of the Chelmer carries all of the water flowing through Chelmsford. Therefore it gives a good overview of the situation with a single set of graphs. However, this station was disbanded in 2010.
There are three active hydrometric stations monitoring waterflow in Chelmsford. One at Springfield on the Chelmer. One at Beach's Mill on the Can. One at Writtle on the Wid. Repeating the analysis for all three of these stations, gives us the complete picture for Chelmsford, whilst also helping identify which areas are responsible for the increases we found previously.
We can calculate the volume of water flowing through each station by multiplying the GDF by the number of seconds in a day. I call this the Volume of Flow (VF) in millions of cubic metres (M m3). The cumulative VF over 3 days (VF 3D) is then the sum of this value over 3 consecutive days. In the graphs below we count the number of days, in each decade, where VF 3D exceeds some value.
The latest average daily GDF figures for each station were obtained from the Environment Agency.
At the Springfield station there is a notable rising trend observable in the number of days when VF 3D exceeds 1 million cubic metres.
0.2 M m3 --- rising by +2.0% per decade
0.5 M m3 --- rising by +1.5% per decade
1.0 M m3 --- rising by +6.2% per decade
1.5 M m3 --- rising by +11% per decade
2.0 M m3 --- rising by +25% per decade
The sharp increase in the 2000s coincides with the beginning of development at Beaulieu Park. Since most of this development is located on, or to the west of the ridge connecting Springfield to the Walthams, run-off from this part of the development will end up in the upper Chelmer. It is plausible that some of this rise was caused by the Beaulieu Park development, and by the same logic, that the Channels development will lead to a further rise in the current decade.
At Beach's Mill the rise is dramatic and worrying.
0.5 M m3 --- -1.4% per decade
1.0 M m3 --- +4.0% per decade
2.0 M m3 --- +24% per decade
3.0 M m3 --- +65% per decade
4.0 M m3 --- +288% per decade
In the 1970s there were only three days where 4 million cubic metres flowed through Beach's Mill over 3 days. By the 2010s (2010-2019), this had increased to 24 days. Some 8 times what it was.
Unlike at Springfield, this rise has been incremental, decade on decade, since the 1970s. The exception being the relatively dry 1990s (Rushes Lock analysis). If development is the cause here, we are looking at the consequence of incremental development at Writtle and elsewhere in West Chelmsford and upstream on the Can. A lot of the houses in this part of Chelmsford were built with front gardens that have gradually been paved or covered with tarmac, increasing run-off.
Interestingly there is no trend observable in the data for the Wid at Writtle.
0.2 M m3 --- -1.1% per decade
0.5 M m3 --- -1.3% per decade
1.0 M m3 --- +0.6% per decade
1.5 M m3 --- -0.2% per decade
2.0 M m3 --- +2.6% per decade
Whereas the North of Chelmsford district falls entirely outside of the Green Belt, and the West of Chelmsford falls partially outside of the Green Belt, the South of Chelmsford, drained by the Wid, falls entirely within the Green Belt and consequently has been spared from major development. Potentially this could explain why we have not seen the same increase at this station.
Also, since the Wid is not experiencing the same increase in extreme waterflows, it probably means that Margaretting was not the correct location for a Flood Alleviation scheme.
In my analysis for Rushes Lock, I found no increase in rainfall over the catchment since the 1970s, and therefore concluded that the observed increase in the number of days with extreme 'flood level' cumulative inflows was likely attributable to over-development. I concluded that analysis saying "if similar trends are observed at other hydrometric stations we should think very carefully about aspirations for further urbanisation". Well, this analysis shows that similar trends are indeed observable at Springfield and Beach's Mill hydrometric stations.
At the moment the link to over-development is purely a rational one, based on reasoned observation. The link needs to be confirmed by the indentification of drainage pipes/channels/ditches leading into the rivers and the monitoring of waterflow. A system of distributed water storage could then be implemented close to each of the sources of run-off to reduce the flood risk to Chelmsford.
If you like this work, please consider supporting the Chelmsflood project by making a donation,
and if you would like to know more get in touch.