Alternate: Taking WaPUG to a New Practical Dimension: Introducing an Innovative H/H Model Validation Tool

H/H Models are the main tool for evaluating and improving the performance of collection systems. Flow prediction quality to variety of storm levels is important to ensure optimized improvements. Maintenance and design activities are expensive and cost-effective improvements is very proportional to good model prediction quality. WaPUG model validation criteria and variety of its form have been widely used to qualify models. Model performance is assessed through a few metrics that determine if the model is closer to observation at acceptable level, including peak flow, volume of flow and the flow depth. WaPUG also states that the observed and model flow hydrographs should closely follow each other in shape. An envelope of acceptable range for the significant peak flow events is the most commonly acceptable criteria. But the open questions remain: what is the acceptable envelope range, how many events within the envelope is required, and how close the hydrograph shape is considered good? In many cases, the goodness-of-fit based on peak and total volume does not guarantee good estimate of overflow volumes or confirm the model prediction is applicable to all seasons. This study presents an easy to use H/H Model Validation tool that overcomes these WaPUG limitations. The H/H Validation Tool reads the observed and model hydrographs and segregate each of them into base flow (dry weather flow with seasonal variation) and storm flow. It then uses user-defined parameters to split the data into individual storms and perform automatic metrics to qualify the model prediction quality at a deeper level. The few automatic metrics include: 1. Visual representation of model quality is replicating dry weather flow conditions throughout the data period by finding maximum and minimum four consecutive base flow days in each month. 2. A procedure to calculate the maximum flow generated within various sliding time spans of each event. The time spans represent the collection system reaction to three distinct flow period: the flow conveyance period to confirm the model quality to be used for pipe capacity studies, the flow storage period targeting the model quality to be used for offline storage studies, and the treatment period targeting the treatment volume at the treatment plants. 3. Present the maximum volume regression comparisons for the conveyance, storage, and treatment periods for all storm. These regression plots give an easy visual indication on the model prediction quality of each event's shape. 4. A heat map comparison for each event. This heat map gives a quick visual understanding. of the events level and their quality of fit. 5. A calculation for the Pearson Correlation Coefficient (the leaner association to the best fit line) and the Concordance Correlation Coefficient (how far from the perfect agreement line) to assess the goodness-of-fit to each metric. The H/H Model Validation Tool provides an easy to use model quality review platform. It eliminates the need for lengthy model review process and does not require the use of any model platform. Comparison results are obtained in few minutes and presented in visual format that is easy to present to variety of stakeholders.