Tags: hepengz/PyDSD
Tags
Bugfix regresion spectrum calculation effective area arguments (josep… …hhardinee#99) * Added some fixes for passing in effective sampling area as a function not being handled appropriately. Also added regressions tests to make sure this doesn't become issue in future * Added test data for arm ld tests * Ran test_DropSizeDistribution.py through black for formatting
Vdisdrops processing (josephhardinee#98) This adds support for processing ARM's vdisdrops file which are 2DVD files that report individual drop sizes rather than bulk statistics. This adds ability to generate different binned dsds and time sampling, as well as updating filtering to work with this new type of data. SubCommit logs: * Added citations for terminal fall speed, and pressure adjustment * vdisdrops reader added to pydsd, needs refactoring and then filtering updated to be more general * Added two new fields, total_measured_drops and number_measured_drops to the vdisdrops reader * Added some options to expand object out to include entire day * Added spectral filtering updates for vdisquants from drop files. All filtering should now be disdrometer agnostic * Removed spectral filtering by size as something silly is happening in my tests * Rearranged how axes on spectra are handled slightly so that everything is consistent between parsivel and vdis now * Added option to maintain smallest drops/speeds when doing spectra filtering
Better drop spectrum support (josephhardinee#92) This adds better support for drop spectrums. Most work is focused on Parsivels to start with, but the support is fairly general. This adds a series of new features: readers that set the field drop_spectrum, and attribute spectrum_fall_velocity will have this tracked. For an example see l105-110 of the ARM_APU_reader.py This also adds support on the DSD object for converting spectra into N(D). Also added are several filtering/QC operations. Chief among these is the filter_spectrum_with_parsivel_matrix function in utility.filter. This filters the spectral matrix (for parsivels) with arbitrary fall speed assumptions. Similarly there is a filter_nd_on_dropsize added. Several other supporting features have been added. I need to make a notebook demonstrating these at some point. Individual updates: * Renamed notebook to reflect change to PyDSD * Initial support for reading in drop spectrum and storing them in fields * Added support for PCM 50% fall speed filtering * Added parsivel spectrum filtering, recovery of Nd from spectrum, and filtering on nd dropsize * Cleaned up parsivel spectrum filtering, added replace option, track history * Cleaned up several instances of inplace to be replace to make the purpose more clear * Run through black and removed default parsivel matrix * Removed Python 3.5 as we use f-strings heavily now * Moved py3.8 to allow failures for now. Issues with pyqt and sip that I'll troubleshoot later
ENH: VDIS reader works with ARM TWP vdis files (josephhardinee#69) * ENH: VDIS reader works with ARM TWP vdis files
PreviousNext