Isabella Arzeno-Soltero Benjamin T. Saenz Kristen Davis
Updated: 2023-04-04
The full G-MACMODS code can be found at [ https://github.com/macmods/G-MACMODS.git].
G-MACMODS (Arzeno-Soltero et al. 2023) is a derivative of MACMODS, utilizing the same conceptual framework, and in many cases parameterizations, as the earlier model described in Frieder et al.,2022. Briefly, the main model differences include using 0D (single set of tracers) for the water column, seaweed type-specific parameterizations for up to 4 different seaweed types [2 temperate browns ('Saccharina','Macrocystis'), temperate red ('Pyropia/Porphyra'), tropical brown ('Sargassum'), and tropical red ('Eucheuma')], and a crowding parameterization derived from empirical growth data (Xiao et al., 2019) and type-specific tuning. See Arzeno-Soltero et al. 2023, including supplementary information, for a full model description.
G-MACMODS was translated into Python-3 for the purposes of generating and analyzing large numbers of simulations in a Monte Carlo manner, because at the time of model creation, seaweed biophysical parameters were very uncertain and a gross estimate of model error were needed. Python and various accelerated numerical and analysis packages provided an efficient scalability through parallelization.
The model code package 'magpy' contains model classes to generate a single 0D simulation (MAG0), and a G-MACMODS simulation, with its global input data dependencies(Arzeno-Soltero et al. 2023).Driver python scripts which created the simulations used in Arzeno-Soltero et al. 2023 are found in the outer directory that generate 'standard' simulations, simulations used for validation comparisons against a suite of other publications, and Monte Carlo simulations for estimating uncertainty.
-- The seaweed type parameterizations from Arzeno-Soltero et al. 2023 (see mag_species.py) were developed from a combination of values derived from literature, and where unavailable, from tuning, such that model output was within the range of wild and farmed seaweed biomass and growth rates that were observed in the oceans (not in a lab setting). Modification of these parameters will result in simulations not supported by research. Use of new seaweed type parameterizations with G-MACMODS should always be accompanied by supporting validation simulations and experimental data.
-- The seaweed genus named mentioned above were the primary sources for parameters for modeled seaweed types, however the types modeled here should not be seen as representative for any single species. Typically, G-MACMODS seaweed types have wider tolerances for environment that most single species, as our attempt was to model yield potential, and one of our assumptions is that strain selection will be used to select cultivars that are suited to the local aquaculture environment.
Surface nitrate concentrations and vertical nitrate fluxes: Long, M., B. Saenz. (2023). Nitrate flux and inventory from high-resolution CESM CORE-Normal-Year integration. Version 1.0. UCAR/NCAR - GDEX. https://doi.org/10.5065/hpae-3j62.
MODIS sea surface temperature (SST), MODIS surface photosynthetically active radiation (PAR), and net oceanic primary productivity (NPP) were downloaded from the Ocean Productivity website (https://sites.science.oregonstate.edu/ocean.productivity/index.php). Specifically, 8-day NPP can be found at http://orca.science.oregonstate.edu/1080.by.2160.8day.hdf.vgpm.m.chl.m.sst.php, whereas 8-day MODIS inputs can be found at https://sites.science.oregonstate.edu/ocean.productivity/1080.by.2160.8day.inputData.php.
Zonal and meridional surface current velocities were taken from the HYbrid-Coordinate Ocean Model (HYCOM99) Global Ocean Forecasting System (GOFS) 3.1, accessed from https://www.hycom.org/dataserver/gofs-3pt1/analysis.