Data for "Stable climate simulations using a realistic GCM with neural network parameteriz

Sampling data from a stable climate simulation using a realistic General Circulation Model (GCM) with neural network parameterizations. The data includes large-scale specific humidity and temperature, their advection rates, and heating and moistening rates from a cloud-resolving model, alongside surface and top-of-atmosphere radiative fluxes. The dataset was created by Xin Wang of Tsinghua University and represents a full year of simulation for 1998.

Use Cases

• Training neural network emulators for atmospheric moist physics based on variables like specific humidity (qv_nn_in) and temperature (T_nn_in).
• Validating hybrid climate model stability using the provided heating (stend_check) and moistening (qtend_check) rate outputs.
• Analyzing radiative energy budgets using surface (FSNS, FLNS) and top-of-atmosphere (FSNT, FLNT) net radiation fluxes.
• Benchmarking parameterization schemes against the cloud-resolving model's high-resolution tendencies.

Strengths

• The description provides specific physical units for all listed variables, aiding interpretation.
• The dataset is linked to a published research paper, suggesting a defined scientific context.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Xin Wang, Tsinghua University

Collection Method

Output from the SPCAM (Super-Parameterized Community Atmosphere Model) simulation.

Time Range

1998