RELEASE NOTES for CAMx v6.11, 12/15/14 The Comprehensive Air Quality Model with extensions (CAMx) is available at http://www.camx.com. Overview of Version 6.11 ------------------------ V6.11 includes a minor modification and several bug fixes from the previous release (v6.10). * Use the chemistry parameters files specifically labeled for v6.10. * There are no changes to I/O file formats from v6.10. * There are no changes to the CAMx control file namelist from v6.10. v6.11 Modifications ------------------- 1. Added a TREESUM algorithm for time-averaging concentrations. Implications: TREESUM improves the fidelity of single-precision end-of-hour sums used to time-average concentrations for output. This helps reduce output concentration differences between MPI and non-MPI runs. v6.11 Bug Fixes --------------- 1. Two improvements to PiG when using MPI parallelization: a) Fixed MPI pass of PiG puff location data at first hour of a model restart. b) Fixed MPI pass of gridded concentration data at first hour of restart when the LVISPIG option is invoked Implications: Both bugs were leading to small concentration differences between MPI and non-MPI runs at each model restart. 2. Fixed an MPI bug that was not passing ozone column maps to slave nodes when a simulation crosses through midnight. Implications: This bug was revealed only when CAMx was run across midnight, as opposed to running CAMx from midnight to midnight (as is usual). It led to concentration differences between MPI and non-MPI runs. 3. Fixed MPI bug that was passing gridded snow code only at start of simulation, not when updated hourly. Implications: No effect in cases where snow is absent or when snow cover is static during a single run. It led to minor concentration differences between MPI and non-MPI runs due to differences in surface albedo and possibly dry deposition. 4. Removed reading of dummy variables in UAM-formatted input files. Implications: The NSEG header variable is left over from UAM and has never been used for CAMx. The NSEG slot is now used for simulation time zone in meteorological input files, but ignored in other files (emissions, IC/BC). 5. Improved internal checks for optional plume rise/distribution parameters. Implications: A recent CAMx modification allows for point source plume depth override by setting a negative value for the time-dependent "flowrate" variable. SMOKE has historically set flowrate to -9 (missing value), which led to model crashes now that negative values are not ignored. This fix allows SMOKE files to continue to operate correctly while still allowing the optional plume depth override to be invoked. 6. Fixed a bug in the argument list for Zhang dry deposition routine when called from RTRAC. Implications: The Zhang dry deposition routine was recently updated to receive species-dependent parameters via an argument list, but this change was not carried over to RTRAC, leading to model crashes with a segmentation fault. 7. Added rounding of time stamp in met input files to handle small increments of numerical fuzz. Implications: Some users complained that CAMx was missing met input times because of imprecision in the file timestamps. This fix should alleviate the problem. 8. Removed trap to stop a simulation using RTRAC with Mechanism 2 (CB6r2). Implications: This left-over trap was inadvertently included in the last distribution. ================================================================================ RELEASE NOTES for CAMx v6.10, 04/02/14 The Comprehensive Air Quality Model with extensions (CAMx) is available at http://www.camx.com. Overview of Version 6.10 ------------------------ V6.10 includes new features, modifications, and bug fixes from the previous release (v6.00). * Use the chemistry parameters files specifically labeled for v6.10. * There are no changes to I/O file formats from v6.00. * The CAMx control file namelist has changed to support certain modifications. The CAMx makefile has been streamlined and updated to support a variety of MPI systems. It's functionality and execution syntax has changed; see the README file, User's Guide, or type "make help" or "make" for instructions. v6.10 Major Updates and New Capabilities ---------------------------------------- 1. Gas-phase chemistry mechanism 2 has been updated from CB6r1 to CB6r2 and includes the capability to run with emitted methane (ECH4). Implications: CB6r2 differentiates organic nitrates (ON) between simple alkyl nitrates that remain in the gas-phase (providing a reservoir of NO2) and multi-functional ONs that can partition into organic aerosols (OA). ON present in OA are then assumed to undergo hydrolysis to nitric acid. These changes tend to reduce regional concentrations of ozone and ON, and increase nitric acid. Regional CAMx simulations with CB6r2 show that accounting for ON hydrolysis in aerosols improved performance for ozone and the partitioning of NOy between ON and nitric acid. CB6r2 will run with CF or CMU aerosol chemistry and works with all Probing Tools except Process Analysis. ECH4 represents emitted methane over and above the global background for methane, which continues to be set internally at 1.75 ppm for all mechansisms. ECH4 decays by oxidation and has the same OH rate constant and products as global background methane. 2. The Plume-in-Grid (PiG) submodel has been extensively updated to include a condensed gas-phase chemical mechanism for GREASD PiG, reduced nocturnal puff growth rates, and other minor updates to improve performance and diagnostic output. Implications: GREASD PiG chemistry now runs much faster. Reduced nocturnal growth rates lead to puff sizes in better agreement with in situ aircraft measurements, and prolonged puff life for both GREASD and IRON PiG. These updates can result in a wide range of ozone, precursor, and PM impacts on the grid, especially for coarser grids that allow for extended puff lifetimes. 3. PM-DDM has been added to the model. This change includes an update of ISORROPIA from v1.6 to v1.7. The "short" names of DDM sensitivity tracers have been changed to accomodate 3-digit chemical species pointers. Implications: DDM sensitivity analysis is now available for PM. Only first order sensitivities are calculated with PM-DDM, but they can be calculated for all PM species. PM-DDM works only with the CF aerosol option. See the User's Guide for more information. The short names of DDM output sensitivities written to output files support up to 999 species, compared to 99 species in previous versions. 4. The RTRAC probing tool has been expanded to allow tracers to directly use photolysis and deposition rates calculated for any core model species. Implications: This update allows RTRAC to be used to emulate any core model species. For example, RTRAC could track the transport and decay/removal of ozone boundary conditions for any spatial stratification designed by the user. See the User's Guide to see how to use RTRAC to point to specific core species. 5. A surface model has been added that performs chemical reactions on deposited material and re-emits those species and chemical products back to the atmosphere. Implications: Heterogeneous reactions can be simulated at the surface and chemical products can be released to the air. The algorithm includes irreversible loss to soil leaching and vegetation penetration, and removal by precipitation. The surface model currently works with the Wesely deposition option. v6.10 Modifications ------------------- 1. The treatment of aqueous chemistry (RADM-AQ) has been modified: a) Bi-section solver was reverted from H+ back to pH; b) Check on high SO2 > 1 ppm (low pH) added to skip aqueous chemistry c) Minimum cloud water content was reverted from 0.01 back to 0.05 g/m3. Implications: Changes to aquous chemistry implemented in v5.41 and v6.00 are reverted back to pre-v5.41 code, with the added safety of checking for high SO2 leading to low pH. Many user's complained of model crashes in RADM-AQ due to excessive looping in the pH solver. Expect minor changes to sulfate and nitrate concentrations and wet deposition. 2. A single output species labelled "ALL" can be set in the namelist, which will trigger the output of all gas and PM species for the chosen chemical mechanism. Implications: Saves time typing long lists of species names, ensures every species is included, and alleviates typographical errors. Use this in cases when all state species (radicals excluded) are needed for output, such as generating boundary conditions in successive 1-way nesting applications. 3. The HDF5 output option has been disabled. Implications: Only the default Fortran binary output files are generated. If conversion to a common self-describing file format is needed, use the tool "camx2ioapi" to convert fortran output files to netCDF/IO-API. v6.10 Bug Fixes --------------- 1. An MPI bug in passing the variable "depsp" was fixed. Implications: Corrects occasional MPI+OMP segmentation faults using MVAPICH. 2. Point source HRVOC tracking was reinstated in DDM. Implications: HRVOC emissions can now be tracked by DDM for sensitivity analsis. Previously added to developmental versions of CAMx v4.5, it was erroneosly dropped from the code. 3. The passing of layer 1 height to MICROMET.F was corrected from deltaz to deltaz/2 in calls from ACM2 and PiG. Implications: This fixes an inconsistency that can lead to minor differences in surface layer stability parameters. It negligibly affects only ACM2 diffusion and PiG growth rates. 4. An RTRAC indexing bug loading boundary conditions for NOx was fixed. Implications: This correction properly maps RTRAC boundary conditions for NOx species. 5. Mapping indices were added for new Hg species for DDM and IEH chemistry. Implications: These pointer mappings were erroneously left out when Hg chemistry was updated in v6.00. 6. Point source mapping bug in nested grid boundary cells was fixed. Implications: This error led to the improper indexing of point sources located in nested grid boundary (buffer) cells, such that these sources were dropped from the simulation.