CVEN9612 – Catchment Modelling

CVEN9612 – Catchment Modelling

Assignment 1 Part 1– Rainfall-Runoff Modeling and Routing

his part of the assignment is worth 15% of the total grade for CVEN9612. The assignment

answers are to be submitted online in Moodle as a short report.

Assignment 1 Part 1 is due 10pm 30th September (Week 4)

Total marks available for each part are listed below.

Question 1 – Runoff routing (15 marks)

There was a large flood on the Richmond and Wilsons Rivers in February 2024, which led toextensive damage in Lismore and surrounding areas. The flood hydrograph on the RichmondRiver was measured at Wiangaree (EL. 91m AHD) and we would like to be able to model theydrograph at the town of Kyogle, 21km further downstream (EL 71m AHD). The observedtreamflow hydrographs at Wiangaree and Kyogle are available on Moodle.Using Manning’s Equation calculate a representative wave velocity for flow down the channelassuming that the hydraulic radius of the channel is R=5.2 and it has a Manning roughness n of0.05. Assume the wave celerity is 1.7 times the Mannings velocity.Employ the Muskingum method to determine the best estimate of the flood hydrograph at Kyogle,assuming that an approximate value of K can be calculated by dividing the reach length by thewave speed. Calibrate your model to estimate the best value for x by minimizing the sum of

squared error of the prediction.

Provide a short report (less than 2 pages), with appropriate figures to answer the followingquestions. Show your working where appropriate and justify any assumptions that are required.

1. a) What time step did you use for your calculations (in hours) and how do you know if it isppropriate? (1 mark)

1. b) What is the peak of the calculated flow hydrograph at Kyogle? (5 marks)
2. c) What value of x did you use and how did you know this was the optimum value?Demonstrate with the use of an appropriate figure. (3 marks)

1. d) What is the RMSE of the calculated outflow hydrograph compared to the observed data atKyogle? (2 marks)

1. e) What is the impact of changing the value of x to 0? What is the practical meaning of x =0? (2 marks)

1. f) If there was interest in investigating Nature Based Solutions in the Richmond catchmentand riparian zone was revegetated and as a result Mannings n was 0.1, what is the newpeak flow at Kyogle? Explain how revegetation affects the routing of flows. (3 marks)12

Question 2 – Model Calibration (15 marks)

Model aim In this part of the assignment, you need to model the runoff and Lake Werri Berri in the Thirlmere

Lakes National Park wetland system using GR4J. The model is to be set up to understand

1. how often the lake is suitable for water sports and
2. the risk of bushfires affecting the sensitive peat ecosystem of the lake.Water sports can only occur when the lake is more than 2 m deep and bushfire risk is too high if

water levels in lakes are less than 0.1 m for more than 30 代 写CVEN9612 – Catchment Modelling consecutive days. The management planfor the lakes requires decisions to be supported with 8 years of data hence why a model is requiredas flow data is only available since 2011. The management plan requires the frequency of thesetwo events (i.e. water sports and bushfire risk) to be documented.

The aim of the modelling exercise is twofold:

(i)to understand the implications of choices in modelling such as objective function anddata transformation on lake water level predictions and what the most appropriatehoices are given the aims of the model.

(ii)To assess an alternative GR4J model for the same catchment which uses satelliteretrieved surrogates using the SRM methodology of Yoon et al., 2023 (week 3 lecture).

Assignment details

Use the AirGR and SRM packages in R to model the catchment. Instructions for AirGR and ashort video on setting up a model in AirGR are provided on Moodle. Please be aware that in theSurrogate River discharge Model (SRM), only one objective function is available, which isfocused on minimizing the Surrogate River discharge Model Error (SRME).Data, code, and the SRM Package can be downloaded via this link: https://deciduous-camp-

995.notion.site/Sharing-SRM-for-CVEN9612-e7a48b40de9b4f38a16915d1c6a07a4dRainfall, evaporation and streamflow data is provided for the catchment on Moodle. Data in all

les is provided in mm/day. The catchment area is 84 ha.You can assume that the wetlands have a plan area of 10.5 ha and that water levels can rangebetween 0 m and 5 m. Assume that the catchment average rainfall and evaporation also applydirectly to the wetland water balance. Assume that there are no groundwater interactionsand thatif water levels are higher than 5 m that all flow is lost instantaneously on that day.You will have to select an appropriate calibration period given the data available for the catchment.It is up to you to decide on how best to use the data that is available in setting up the model. Youwill also have to decide what objective function and data transformation is the most appropriategiven the model aims.In your short report (less than 3 pages), you need to assess at least two objective functions and twodata transformations and comment on the sensitivity of your conclusions to the modelling choices

you made when calibrating using observed flow data. You also need to compare the SRMcalibration with that for the model version where observed flows are used for calibration, notingkey similarities and differences. Provide suitable figures and tables of results to support yourconclusions.3