radarlidar_analysis

Scientific Background

Methods

This project aims to compine radar and lidar windspeed and direction data into one combined product. Therefor a new height and time grid is defined. The height grid consist of 26 vertical layer from 0m AGL to 13km AGL. The time grid for daily plots has time steps of 0.5h. The raw data from radar and lidar are projected onto this new grid with the NearestNeighbors-Algorith from sklearn.

Code Base

This project consists of three files:

RadarLidarWindSpeed.py

The file "RadarLidarWindSpeed.py" contains a class "RadarLidarWindSpeed". This class stores all data in a pandas dataframe. To access and change this data, a set of methods is provided:

• createDataframe() This methods creates a pandas dataframe to store all prcoessed data.
• mergeHeight() This method uses a neares neighbour algorith to map two height scales.
• mergeTime() This method uses a neares neighbour algorith to map two time scales.
• readSpeedFile() This method reads the file with radar and lidar wind speed data.
• readDirectionFile() This method reads the file with radar and lidar wind speed data.
• importDataset() This method summs up the import process and also calls readDirectionFile() and readDirectionFile().
• calculateDifferences() This method calculates the differences between radar and lidar data.
• getSpeedHeightProfile() This method calculates a coverage-height-profile for wind speed.
• getDirectionHeightProfile() This method calculates a coverage-height-profile for wind direction.
• getCoverageHeightTimeSeries() This method calculates a coverage-height-timeseries for data coverage.
• calculateSpeedFusion() This methods merges radar and lider speed data.
• calculateDirectionFusion() This methods merges radar and lidar direction data.
• calculateAvailability() This methods calculates a value [0,1,2, 3] for data availability.
• exportNCDF() This method exports all data from the datafram into a netcdf file.
• windspeedFullHeightCoveragePlot() This method creates a speed coverage plot for the full height.
• windspeedFullHeightOverviewPlot() This method creates a speed overview plot for the full height.
• windspeedBoundaryLayerCoveragePlot() This method creates a speed coverage plot for the boundary layer.
• windspeedBoundaryLayerOverviewPlot() This method creates a speed overview plot for the boundary layer.
• winddirectionFullHeightOverviewPlot() This method creates a direction overview plot for the full height.
• winddirectionBoundaryLayerOverviewPlot() This method creates a direction overview plot for the boundary layer.
• availabilityPlot() This method creats an availability plot.
• histogramDifferencePlot() This method creates a histogram of the differences.

In addition to this class two files are provided to run routines automaticly:

call_routine.py

This code creates Quicklook plots for current and last day. It can be started by

python3 call_routine.py

The file can be modified in the following way. With the line

days_range = (0, 2)

the number of days, for which quicklooks are created, is selected. In this case, quicklooks are created for the current and the previous day. With the line

storagePath = "/work/marcus_mueller/routine/"

the storage path can be changed. The software automaticly creates a folder structure based on the date under this path.

call_historyPlot.py create statistical plots on data availability

The fille call_historyPlot.py contains a routine to create availability plots for historic data. It can be started by:

python3 call_historyPlot.py

The following changes can made in the code:

storagePath = "/work/marcus_mueller/history/" Storage Path for Plots

start = datetime(2019, 4, 1) Start date for the plot

nMonths = 23 number of months to process from start date on