EPIC version 0.14.2

What's Changed

• Optimise nearest algorithm tree resolve step by @matt-frey in #561
• Implement generalised damping, including surface damping by @sjboeing in #562
• Add optimised rejection method by @matt-frey in #563
• Update webpage by @matt-frey in #564
• Add labels to code by @sjboeing in #565
• Reset labels and dilution when writing to netCDF by @sjboeing in #566
• Support for 64-bit integers by @matt-frey in #560
• Set total number of parcels before parcel_communicate by @matt-frey in #567
• Fix input to acos by @matt-frey in #568
• Fix rejection method by @matt-frey in #570
• Explicitly set the attributes to write by @matt-frey in #571
• Bugfixes to labelling by @sjboeing in #569
• Move GitHub page to organisation by @matt-frey in #573
• Add index.html which is a symbolic link to html/intro.html by @matt-frey in #574
• Update version string from 0.14.1 to 0.14.2 by @matt-frey in #575

Full Changelog: 0.14.1...0.14.2

EPIC version 0.14.1

What's Changed

• Added/changed ifort style flags by @RuiApostolo in #494
• Fix .zenodo.json by @matt-frey in #542
• Generalise configure.ac by @matt-frey in #492
• Selective parcel attribute writes by @matt-frey in #543
• Need to calculate buoyancy extrema for total buoyancy by @matt-frey in #546
• Update writing statistics to improved interface by @matt-frey in #545
• Fix rk_timer by @matt-frey in #547
• Collect merger statistics by @matt-frey in #544
• Update conda environment by @matt-frey in #550
• Damping implementation by @sjboeing in #549
• Add missing stop_timer; fix resetting parcel split and merge diagnostics by @matt-frey in #552
• Fix loop for nearest parcel check by @sjboeing in #555
• Gridded strain by @sjboeing in #553
• Ensure link exists before calling netcdf routine by @matt-frey in #558
• New release 0.14.1 by @matt-frey in #559

New Contributors

• @RuiApostolo made their first contribution in #494

Full Changelog: 0.14.0...0.14.1

EPIC version 0.14.0

What's Changed

• Fix the write frequency by @matt-frey in #527
• Add boundary surface fluxes by @matt-frey in #451
• Add missing field statistics timer stop by @matt-frey in #537
• Calculate squared buoyancy frequency only if not provided by @matt-frey in #538
• Compilation flag based 1-point par2grid and grid2par by @sjboeing in #539
• Overwrite default Zenodo metadata by @matt-frey in #536
• Dev 0.14.0 by @matt-frey in #541

Full Changelog: 0.13.1...0.14.0

EPIC version 0.13.1

What's Changed

• Write specific fields only by @matt-frey in #520
• Remove -fPIC flag by @matt-frey in #519
• Do halo swap for nparg and nsparg by @matt-frey in #523
• Fix OpenMP pragmas by @matt-frey in #528
• Change default value from 1/40 to 1/20 by @matt-frey in #526
• Initialise buoyancy, vorticity and humidity by @matt-frey in #534
• Buoyancy perturbation mode by @matt-frey in #532
• Update Python packages by @matt-frey in #522
• Update Python tools by @matt-frey in #529
• Increase buffer in parcel communication by @matt-frey in #530
• Add data coarsening post-processing tool by @matt-frey in #521
• Update changelog for version 0.13.1 by @matt-frey in #535

Full Changelog: 0.13.0...0.13.1

EPIC version 0.13.0

MPI-parallel version of the 3D model.

What's Changed

• Rename comm to comm_world by @matt-frey in #364
• Extend NetCDF interface to MPI by @matt-frey in #363
• Fix MPI reduction by @matt-frey in #366
• Field decomposition by @matt-frey in #361
• Grid halo swap by @matt-frey in #369
• Parcel and field diagnostics in parallel by @matt-frey in #368
• Read and write fields in parallel by @matt-frey in #367
• Read and write parcels in parallel by @matt-frey in #365
• Parcel exchange by @matt-frey in #362
• Parallel parcel initialisation by @matt-frey in #373
• Parallel par2grid by @matt-frey in #374
• Parallel grid2par by @matt-frey in #375
• Parallel parcel split by @matt-frey in #376
• Parcel rejection method by @matt-frey in #377
• Parallel gradient correction by @matt-frey in #383
• Forward and backward FFT by @matt-frey in #438
• MPI parallel FFT by @matt-frey in #428
• MPI parcel Laplace correction by @matt-frey in #442
• MPI timer module by @matt-frey in #443
• Pull from main branch into 3d-mpi-model branch by @matt-frey in #444
• Improve MPI communication by @matt-frey in #445
• MPI passive communication in nearest algorithm by @matt-frey in #448
• Remove reference potential energy calculation for 3D by @matt-frey in #449
• Parallel nearest algorithm by @matt-frey in #384
• Improve the search for libraries by @matt-frey in #452
• Changes on halo swapping branch by @sjboeing in #456
• Fix MPI related issues by @matt-frey in #453
• Make RK4 variables part of the parcel container by @matt-frey in #457
• Fix MPI related issues by @matt-frey in #460
• Read parcels in chunks by @matt-frey in #459
• Explicit array bounds by @matt-frey in #458
• Recover lost changes from main to MPI branch by @matt-frey in #467
• Cleanup and additional debug check in parcel nearest by @matt-frey in #466
• Proper de-allocation of allocatable arrays by @matt-frey in #468
• Print on rank 0 only and use MPI_Abort instead of stop by @matt-frey in #469
• Communicate multiple fields in a single call by @matt-frey in #470
• Fix write_netcdf_axis_3d by @matt-frey in #473
• Single comm sum based by @sjboeing in #476
• Fix MPI unit tests by @matt-frey in #477
• Allocate RMA windows in MPI tests by @matt-frey in #480
• Change nearest comparison script by @matt-frey in #481
• Fix lower boundary of local box by @matt-frey in #482
• Fix the vorticity correction by @matt-frey in #486
• Use request-based RMA by @matt-frey in #472
• 3rd order RK stepper by @matt-frey in #487
• No allocatable strings by @matt-frey in #479
• Implement algorithm by Scherzinger and Dohrmann by @matt-frey in #488
• Ensure argument to dacos is within bounds by @matt-frey in #495
• Use amax instead vmax as parcel splitting criterion by @matt-frey in #489
• Resize parcel container by @matt-frey in #493
• Communicate veta and vtau by @matt-frey in #496
• Fix timer by @matt-frey in #499
• Use MPI_Win_allocate instead of MPI_Win_create by @matt-frey in #498
• Fixes from serial version by @matt-frey in #501
• Fixes issue with zero-indexing in inversion module by @sjboeing in #502
• Replace eta and tau as part of parcel_replace subroutine by @sjboeing in #503
• Parcel interpolation simplifications by @sjboeing in #505
• Fix closing bracket in the gradient correction by @matt-frey in #507
• Apply #507 to development branch by @matt-frey in #508
• Fix parcels on periodic edge by @matt-frey in #509
• Avoid array temporaries by @matt-frey in #504
• Remove max_vratio from configuration files by @matt-frey in #510
• Simplify parcel merge routine call by @matt-frey in #514
• Fix unit-tests -- register missing timer and fix function call by @matt-frey in #513
• Fix communicator issue by @matt-frey in #512
• Do not use MPI_COMM_WORLD as default in collectives by @matt-frey in #515
• MPI version of 3D model by @matt-frey in #511
• Fix OpenMP race condition in inversion_utils.f90 by @matt-frey in #516
• Update changelog version 0.13.0 by @matt-frey in #518

Full Changelog: 0.12.6...0.13.0

EPIC version 0.12.6

What's Changed

• Pull optimizations to serial version by @matt-frey in #500

Full Changelog: 0.12.5...0.12.6

Note: This is the last version of EPIC with a serial 3D model.

EPIC version 0.12.5

What's Changed

• Update requirements.txt by @matt-frey in #450
• Fix the calculation of the vorticity correction by @matt-frey in #484

Full Changelog: 0.12.4...0.12.5

EPIC version 0.12.4

What's Changed

• Fix field reading in 2D genspec and add 3D genspec by @matt-frey in #346
• Divergence form and nearest parcel fix by @matt-frey in #399
• Rename RK4 constants by @matt-frey in #401
• Use named indices for n-rank arrays by @matt-frey in #400
• Clean up flux form code by @matt-frey in #402
• Unreadable timings by @matt-frey in #405
• Apply nearest fix in 2d by @matt-frey in #403
• Generalise NetCDF dimension and axis names by @matt-frey in #406
• Monitor number of splits and merges by @matt-frey in #407
• Clean up par2grid by @matt-frey in #409
• Zeta zero on boundary by @matt-frey in #408
• Fix compilation with config.f90 by @matt-frey in #410
• Update the manual by @matt-frey in #411
• Fix python scripts due to NetCDF interface by @matt-frey in #412
• Fix writing z axis to NetCDF by @matt-frey in #413
• Changes to moist setup: shorter run, smooth edge bubble by @sjboeing in #414
• Humidity diagnostics by @sjboeing in #415
• Sgs diagnostics by @sjboeing in #416
• Restarts with zeta flags by @sjboeing in #417
• Update requirements.txt by @matt-frey in #418
• Update the input section of the manual by @matt-frey in #419
• Add more statistics by @matt-frey in #420
• Fix parcel interpolation by @matt-frey in #423
• Use central differencing in solenoidal correction and tendency calculation by @matt-frey in #425
• Peref in NetCDF output by @matt-frey in #422
• Linear extrapolation central diffz by @matt-frey in #426
• Fix buoyancy extrema calculation by @matt-frey in #427
• Check if file exists by @matt-frey in #430
• Domain averaged quantities in 2D and 3D and protected parameters in 2D by @matt-frey in #431
• Update python scripts by @matt-frey in #432
• Setting 3D default n per cell to 8 (a cube) by @sjboeing in #434
• Add missing factor 1/2 to peref calculation by @matt-frey in #437
• Fix verbose mode and update initialisation in parcel correction unit tests by @matt-frey in #435
• APE calculation by @matt-frey in #440
• Initialise parcels like grid2par by @matt-frey in #441
• Add 3d setups by @matt-frey in #433

Full Changelog: 0.12.3...0.12.4

EPIC version 0.12.3

What's Changed

• Make variables private for safety by @matt-frey in #359
• Calculate KE on the grid by @matt-frey in #358
• Change vmin and vmax and lambda_max by @matt-frey in #345
• Use relative error by @matt-frey in #360
• Change to criterion used in paper by @matt-frey in #344
• Fix gridded KE calculation by @matt-frey in #381
• Add split merge unit test by @matt-frey in #380
• Flux form including new inversion and net vorticity correction by @matt-frey in #379
• Fix B matrix advection by @matt-frey in #388
• Write time in nc_fields.py by @matt-frey in #389
• Write vorticity tendency in diagnose mode by @matt-frey in #390
• Add parcel enstrophy diagnostic by @matt-frey in #391
• Fix energy unit by @matt-frey in #392
• Write nparg and nsparg to NetCDF by @matt-frey in #393
• Add missing symbol in strings by @matt-frey in #394
• Additional FFT unit test by @sjboeing in #395
• Update requirements.txt by @matt-frey in #396
• Porting new inversion to EPIC by @matt-frey in #397
• Lagrangian vorticity tendency calculation by @matt-frey in #398

Full Changelog: 0.12.2...0.12.3

EPIC version 0.10.8 with static TG

This version of EPIC can only simulate the static Taylor-Green vortex test case.