ABSTRACT:

This is raw environmental time series data stored in a sqlite database with a data schema loosely based off of ODM1.1. This scheme is shown in the data model figure included in the resource. The geographical location of these data is in the Hampton Roads region in South East Virginia. The variables of the time series are rainfall, tide, wind, and water table elevations. These data were processed and used as input for data-driven modeling for street flood severity prediction. The processing and modeling are described in this Journal of Hydrology Paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Python 2 Jupyter notebook that aggregates sub-daily time series observations up to a daily time scale. The code was originally written to aggregate data stored in the sqlite database stored in this resource: https://www.hydroshare.org/resource/9e1b23607ac240588ba50d6b5b9a49b5/

ABSTRACT:

sqlite database containing data related to street flooding in Norfolk, Virginia USA. The tables include pre- and post-processed data for machine learning models. The raw data that are preprocessed are found in this resource: https://www.hydroshare.org/resource/9e1b23607ac240588ba50d6b5b9a49b5/. The pre-processing is done via a Python 2 Jupyter notebook stored in this resource: https://www.hydroshare.org/resource/e46c995f38194c41934930a10079042b/. The preprocessed data are used in the model stored in this HS resource: https://www.hydroshare.org/resource/ae53ae6bd4374dd1a292b3555b9fa5f7/.

ABSTRACT:

A Python 2 Jupyter notebook that models flood counts using the Random Forest model. The input and output data are from Norfolk, Virginia USA.

ABSTRACT:

Helper functions for accessing data stored in an sqlite database. There are generic functions, such as one to read a database table into a Pandas dataframe . There are also more specific functions designed to interface specifically with the data schema implemented in the database in this HydroShare resource: https://www.hydroshare.org/resource/9e1b23607ac240588ba50d6b5b9a49b5/.

ABSTRACT:

This resource aggregates several resources related to street flood severity modeling in Norfolk, Virginia USA. The resources include raw and pre-processed data, scripts used to perform the pre-processing, scripts used to train data-driven algorithms, and results from the models. The models used crowd-sourced street flood reports as target values and environmental data as input values. The resources in this aggregate resource are used to generate the results for this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

A diagram showing how these resources relate is shown in the "Resource workflow diagram for street flood severity modeling in Norfolk, VA 2010-2016" resource.

ABSTRACT:

Figures resulting from RF modeling of flood counts

ABSTRACT:

This is tabular output data from two data-driven models used to predict flood severity, Poisson regression and Random Forest regression. Both outputs from the training and testing phases of the modeling are included in the resource. Additionally, results indicating the relative importance of each predictor variable in the Random Forest model are provided in the "rf_impo_out.csv" file. This work is described in the following paper published in the Journal of Hydrology: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

This is tabular input data originally used in two data-driven models (Poisson regression and Random Forest) for predicting flood severity. The inputs to the model (or predictor variables) are environmental conditions such as cumulative rainfall, high and low tides, etc. The outputs (or target variable) of the model is the number of flood reports per storm event. This data was used in work that is described in the following paper published in the Journal of Hydrology: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

This is a script written in the R programming language. The script is used to train and apply two data-driven models, Random Forest and Poisson regression. The target variable is the number of flood reports per storm event in Norfolk, VA USA. The input variables for the models are environmental conditions on an event time scale (or daily if no flood reports were made for an event). This script was used to produce results published in a paper in the Journal of Hydrology: https://doi.org/10.1016/j.jhydrol.2018.01.044.
---
Original run configurations:
R version = 3.3.3
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)
Packages used:
'randomForest' (version 4.6-12)
'caret' (version 6.0-73)

ABSTRACT:

Street flooding reports made by mostly City of Norfolk staff from 2010-2016. The coordinate system used for the X and Y coordinates is "Virginia state plane, south zone, feet (NAD83)." These data were processed and used as target values for street data-driven flood prediction severity modeling. This modeling is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Script and accompanying notebook written in Python 2.7 for processing street flood reports made by City of Norfolk staff. The output data from this script were used as target values for street data-driven flood prediction severity modeling. This modeling is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Processed street flooding data from street flood reports made by City of Norfolk, VA staff 2010-2016. These data were used as target values for street data-driven flood prediction severity modeling. This modeling is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Script and accompanying notebook written in Python 2.7 for combining flood report data (output) and environmental data (input) into a format suitable for a data-driven model. These data used as target values for street data-driven flood prediction severity modeling for Norfolk, VA 2010-2016. This modeling is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Daily observations data for rainfall, wind, tide, and water table levels. These variables are more fully defined in the raw source data. These data are used as input for data-driven prediction of street flood severity in Norfolk, VA 2010-2016. This modeling is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Script and accompanying ipython notebook written in Python 2.7 for aggregating sub-daily environmental data (rainfall, tide, wind, groundwater) to a daily timescale. The input data are from Norfolk, Virginia. Several different methods of aggregation are used including averages and maximums. The processed/aggregated data are combined with street flood report data to be used in data-driven, predictive modeling. The script in this resource was used in the analysis described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

Diagram depicting the relationship between 10 different HydroShare resources used to produce results for data-driven street flood severity modeling done for Norfolk, VA for 2010-2016. The analysis is described in this Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044.

ABSTRACT:

This is a Python script used to plot results from a street flood severity model. The script plots predicted flood reports against true flood reports and was originally used for making a plot for a Journal of Hydrology paper: https://doi.org/10.1016/j.jhydrol.2018.01.044. The data files used to produce the plot for the paper are found in another HydroShare resource: https://www.hydroshare.org/resource/54df00b15c02458685fa3b622f2ecc7b/. For the script to work as is, the script has to be in the same directory as the data files and the files have to be named as follows: "poisson_[suffix]_train", "poisson_[suffix]_test", "rf_[suffix]_train", "rf_[suffix]_test". The "suffix" value should be the same as the suffix specified when using the R code that produces the data files. This code is also part of a HydroShare resource: https://www.hydroshare.org/resource/712cd2ce8f604c8f824d6836ee3fcb53/. The script is used as follows "python plot_count_model_results.py [suffix]".

Python version 2.7
Required matplotlib, pandas, and numpy

ABSTRACT:

This is a SWMM5 model that used for demonstrating swmm_mpc, a Python package for simulating model predictive control using SWMM5 as the process model. swmm_mpc is on github: https://github.com/UVAdMIST/swmm_mpc. To run these, you will need to install the swmm_mpc python package or use the Docker image according to the instructions on the github repo readme. This model was used as a demonstration in the following manuscript:

Sadler, J. M., Goodall, J. L., Behl, M., Morsy, M. M., Culver, T. B., & Bowes, B. D. (2019). Leveraging open source software and parallel computing for model predictive control of urban drainage systems using EPA-SWMM5. Environmental Modelling & Software, 120, 104484. https://doi.org/10.1016/j.envsoft.2019.07.009

ABSTRACT:

This resource contains data and models that were used to produce results for a paper published in the Journal of Hydrology. The models are for a neighborhood in Norfolk, Virginia USA that suffers from frequent coastal flooding. The paper describes the use of active stormwater controls to mitigate this problem which will worsen with sea level rise. The particular type of control approach explored was model predictive control (MPC) and the Stormwater Management Model (SWMM) was used to simulate the stormwater system. The swmm_mpc Python package (https://github.com/UVAdMIST/swmm_mpc) was used to simulate MPC in the SWMM model. MPC was simulated for a number of sea level rise scenarios and the amount of flooding was compared to the system with no controls. The Python script that ran swmm_mpc for the sea level rise scenarios is "models/runs/hgv11.py." The results were compiled and plotted with scripts in the "models/results/" directory.

The citation to the Journal of Hydrology paper is
Jeffrey M. Sadler, Jonathan L. Goodall, Madhur Behl, Benjamin D. Bowes, Mohamed M. Morsy, Exploring real-time control of stormwater systems for mitigating flood risk due to sea level rise, Journal of Hydrology, Volume 583, 2020, 124571, ISSN 0022-1694, https://doi.org/10.1016/j.jhydrol.2020.124571.

ABSTRACT:

Output of PRMS/SNTemp. The spatial domain of this model instance is the Delaware River Basin.