The forecast river level graph (green line) is automatically updated using data from the SEPA guage at Alford and provides a 3 hour forecast from the latest data received.
To produce the predicted graph - a correlation of river levels between Alford and Kemnay (Milton Meadows) is being taken and a relationship table populated. Relationships are also being collated between the associated water levels and the downstream time to reach Kemnay from Alford. These tables are then used to calculate a predicted river level at Kemnay.
This is a fairly crude prediction method and does not fully account for localised rainfall events in the catchment area between Kemnay and Alford. Relationship data between Kemnay and Alford has been collected since Jan 2018, the accuracy will depend on the number of high river levels recorded. The relationship model is updated following high river level events.
This graphic is for information only and should not be used for decision making with regards to river levels.
This graph is unlikely to give meaningful until a range of river depths have been captured from the Milton Meadows river guage.
Use the Haughton prediction for potential river levels downstream from Milton Meadows.
The above graphic shows the current set of data used to drive the relationship model. Each occasion of a peak of river level is plotted against the relevant heights at Alford and Milton Meadows. The resulting scatter graph above shows the relationship between the two sites. The dots represent individual occasions of matched river levels. The trend lines are overlaid, where the blue dots are average heights where we have several data readings..
The heavy rains of 3/4th October 2020 resulted in a large rise of the river Don with 3.16m measured on the Milton Meadows gauge. This has allowed us to compare the new readings with the SEPA gauges at Bridge of Alford and Haughton.
The blue line above represents the measurements at Milton Meadows compared to Bridge of Alford and it closely flows the equation of a line. However, the blue line will significantly under predict the event on the 7 Jan 2016 (top right red dot) where we estimate the river rose to 5.5m. So, to reach this point our equation has to be modified when the river exceeds around 3.2m with the equation given for the red line - an exponential.
Our latest prediction model, updated 10/10/2020 now runs on two equations, the straight line to 3.2m and then the exponential. Using the data from the 4/5th October and estimated data from the 7th Jan 2016 the current model will give a good approximation for the next 3hrs.