Missing records in hydrological databases represent a lack of information and have negative consequences for water management. An incomplete time series (for example, of measures of river flow)prevents the computation of hydrological statistics and indicators. Also, records with data gaps are not suitable as input or validation data for hydrological or hydrodynamic modeling.
gapIT is a JAVA-based software package that automatically calculates missing data using different data-infilling techniques, using hydrological discharge data series measures at gauging stations as input. Donor stations are automatically selected based on Dynamic Time Warping, geographical proximity and upstream/downstream relationships among stations. For each gap, the tool computes several flow estimates through various data-infilling techniques, including interpolation, multiple regression, regression trees and neural networks. The interactive visual application allows the user to select different donor station(s) than those automatically selected.
The results are validated by randomly creating artificial gaps of different lengths and positions along the entire records. Using the Root Mean Squared Error and the Nash-Sutcliffe coefficient as performance measures, the method is evaluated based on a comparison with the actual measured discharge values.
The interactive but automated approach of gapIT, coupled with a visual inspection system for user-defined refinement, allows for standardized objective infilling, where subjective decisions are allowed but are at the same time traceable.
The data-driven approach of the software can be reused in similar real-world situations (for any type of missing sensor data).
Senior Valorisation and Transfer OfficerEnvoyer un e-mail