This dataset includes: (1) matching and sorting data for constructing an artificial neural network (ANN) ensemble model to simulate the concentration of dimethyl sulfide (DMS) on the sea surface; (2) The ANN model simulates the global daily sea surface concentrations of dimethyl sulfide from 1998 to 2017, as well as the calculated total transport velocity (Kt) and air sea flux. The input variables of this ANN ensemble model include chlorophyll a, sea surface temperature (SST), mixed layer depth (MLD), nitrate, phosphate, silicate, dissolved oxygen (DO), downward shortwave radiation (DSWF), and sea surface salinity (SSS). The spatial resolution of the simulated dataset is 1 °× 1 °. The units of DMS concentration, Kt, and flux are nmol L-1, ms-1, and μ mol S m-2d-1, respectively.
The in-situ DMS measurement data used to train machine learning models comes from the Global Surface Seawater DMS (GSSD) database. The GSSD database contains 266 cruise and fixed location observation activities from March 11, 1972 to August 27, 2017（ https://saga.pmel.noaa.gov/dms/ A total of 87801 DMS measurement data were obtained.
An artificial neural network ensemble model was developed based on 9 environmental factors, which demonstrated high accuracy and generalization in predicting DMS concentration. Subsequently, a global sea surface DMS concentration and flux dataset (1 °× 1 °) was established from 1998 to 2017, with a daily resolution of 1 °.
The data quality is good.
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:11738.11.NCDC.ZENODO.DB6451.2024
10.5281/zenodo.7898187