Redan idag har många kommuner i Sverige drabbats av skyfall och ett förändrande klimat indikerar att skyfall kommer bli vanligare. Översvämningar till följd av skyfall kan ge konsekvenser som skador på hus, vägar och även utgöra fara för människors hälsa. Skyfallskartering är ett verktyg som kan användas för att identifiera utsatta områden vid inträffandet av ett extremt regnförlopp. DHI har utvecklat MIKE 21 och MIKE Flood som kan användas vid skyfallsmodellering. MIKE Flood är en kopplad 1D-2D modell där ledningsnätet integreras. MIKE 21 är en 2D modell där regnets transportväg på ytan simuleras, ledningsnätet representeras av ett schablonvärde som dras bort från regnet som belastar modellen. Denna studie avser att jämföra dessa modeller för att undersöka behovet av att integrera ledningsnätet, samt utreda hur olika belastningar på ledningsnätet påverkar resultatet. Denna komparativa studie bygger på att MIKE 21 och MIKE Flood sätts upp för samma område lokaliserat i Nacka. Modelleringen resulterar i en karta som visar vattendjup och översvämningens utbredning. Resultatet visar att den största differensen mellan vattendjupen uppstår för djup mellan 0.1 till 0.3 meter. Vidare ger MIKE Flood generellt en större översvämningsutbredning än MIKE 21. Anledningen till denna skillnad är interaktionen mellan 1D och 2D modellen i MIKE Flood. Valet av schablonvärde är viktigt för MIKE 21 modellen då det har en stor påverkan på översvämningsutbredningen. Ledningsnätet betydelse minskar desto mer extrem nederbörden är. / Several municipalities in Sweden have been affected by cloudbursts and a changing climate indicates a future where these events gets more common. Cloudburst are events where extreme precipitation takes place under a short period of time. The consequences is flooding where damages to property and impede accessibility for emergency vehicles are some possible effects. Flood modelling is a way to identify vulnerable areas so that mitigations can be implemented. For example municipalities can use it when planning for new household areas and infrastructure. The Danish Hydrological Institute (DHI) has developed MIKE 21 and MIKE Flood. MIKE Flood is a coupled 1D-2D model while MIKE 21 is a 2D model. For the coupled model the storm water collection system can be integrated and represented by the model MIKE Urban. In the MIKE 21 model the collection system is represented by withdrawing a standard value from the applied rainfall. This comes with the assumption that the collection system fills up completely with the assumed capacity represented by the standard value. Since, MIKE Flood is a coupled 1D-2D model it requires more time and data to set up than the 2D model. Therefore, this study aims to compare these models in order to evaluate the impact of implementing the storm water collection system. Further, to evaluate how the division of the rain loads to each model affect the modelling results. This thesis is performed by setting up MIKE 21 and MIKE Flood for an area in Nacka. The models will be run for six scenarios, which results in a total of 12 simulations. For the scenarios two different recurrences are tested, a 100 and 50 year cloudburst event. Further, the collection system is represented with three different loads corresponding to a 5, 10 or 20 year rainfall event. The modelling results in maps containing water depths and flood extent. In order to compare the models the generated water depths and extent was compared. If the difference between the models was lower than 0.1 meter they was assumed to generate approximately the same result. In order to examine how well MIKE 21 and MIKE Flood matched in the different scenarios an index, measure of fit, was calculated. Moreover, the interaction between the 1D and 2D model was investigated for MIKE Flood. The result showed that there was a difference in flood depth and extent between MIKE 21 and MIKE Flood. Generally there were more flooded cells in MIKE Flood than in MIKE 21 for each scenario besides one. The larges difference between generated water depths was for depths between 0.1 to 0.3 meters. MIKE flood generates a bigger flood extent downstream along the collection system due to the lack of capacity in those parts of the collection system, along Vikdalsvägen the collection system seem to have the capacity to handle a 5 year load. The probable reason for the difference is the interaction between the collection system and 2D model. The collection system, MIKE Urban, enters new paths for the rain to travel that is not available in a 2D model. In areas where the capacity is reached there will be a flux from the 1D model to the 2D model and in areas whit capacity the flux can be in the opposite direction. The assumption that the collection system fills up entirely is not correct for Nacka. Further, the difference between the results from MIKE 21 and MIKE Flood increased with a higher network load. Which indicates that the importance of the storm water collection system reduces with more extreme precipitation events. That is the reason the index, measure of fit, is higher for the scenarios with a cloudburst of a 100 year return period.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-283608 |
| Date | January 2020 |
| Creators | Andersson, Elin |
| Publisher | KTH, Hållbar utveckling, miljövetenskap och teknik |
| Source Sets | DiVA Archive at Upsalla University |
| Language | Swedish |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | TRITA-ABE-MBT ; 20655 |
Page generated in 0.0022 seconds