Describe your issue in detail
Hello,
I’m trying to investigate the physicochemical processes (transport, emissions, deposition, chemistry, …) related to the formation and loss of NOx using the process analysis routine.
You would expect emissions to be the dominant contributor to NOx variability over the surface.
So I picked simulated surface NOx>20 ppbv in Oct 2023 to first double-check the closure:
Which looks okay. Now I’m showing the individual tendencies:
EMIS_NOx (EMIS_NO+EMIS_NO2) is literally close to zero! Transport is wrongly perceived as the driver of NOx variability. I decided to break down the transport into individual components (VDIF, HDIF, ZADV, HADV), and I do see VDIF mostly representing what emissions should look like (extremely high values at the surface and much lower values above the surface). Why is that?
One more question, shouldn’t the process analysis results be in ppmv/hr instead of ppmv?
Best,
Amir Souri
NASA 614/GESTARII
I guess, if my results are correct, they imply that surface emissions and vertical mixing are solved simultaneously to maintain numerical stability instead of simply injecting a volumetric addition (EMIS_NOX). Right?
Yes, you are correct. I put your question in the google search button and found this to be true for CMAQ for the following reasons:
- Avoiding Numerical Instability
- Physical Realism and Continuous Flux
- Impact on Non-Linear Chemistry
Link with AI answers for CMAQ:
For surface emissions (like (EMIS_NOX), CMAQ does not treat them as a volumetric source S in the first layer (layer₁). Instead, they are formulated as a flux lower boundary condition at the ground (z = 0):
This formulation prevents local concentration spikes in layer₁, stabilizes the operator splitting sequence, and prepares a smooth vertical profile for the gas-phase chemistry module (CHEM) that follows.
It’s important to note that one reason that transport is so important is both the simultaneous mixing solution and the depth of the first layer. CMAQ has a non-local PBL solution (ACM2) that allows for rapid mixing within the PBL. It also is typically applied with a 20m first layer. In a well-mixed 2km PBL, that means a rapid removal of surface emissions.
This post discusses using an analysis volume to better see the important role of each process.
