Opened 5 months ago

Last modified 5 months ago

#3548 new help

SDGVM: small values of vegetation carbon & soil carbon

Reported by: wcai Owned by: jules_support
Component: Land Surface Modelling Keywords: SDGVM, units, soil carbon, vegetation carbon
Cc: Platform:
UM Version:

Description (last modified by pmcguire)

Dear Dr. McGuire,

I'm a PhD student in Imperial College London supervised by Prof. Iain Colin Prentice. My recent focus is on simulation of CO2 seasonal cycle amplitude using 'P model', and with vegetation and soil carbon outputs from SDGVM.

I found out that at some grid cells the value of vegetaion and soil carbon is extremely small, so I was wondering if the unit of vegetation and soil carbon 'kg/m2' refers to per square meter grid cell, or per square meter land cover? In other words, are the extremely small values coming from the fact that the grid cell is mainly occupied by water or ice etc.? I'd really appreciated it if you could solve my puzzle. Thank you very much!

Best regards,
Wenjia Cai

Change History (6)

comment:1 Changed 5 months ago by pmcguire

  • Description modified (diff)

comment:2 Changed 5 months ago by pmcguire

Hi Wenjia
I can try to help.

What vegetation parameter are you looking at? Are you looking at the
NPP or the GPP?
Are you looking at the NPP or GPP for different plant functional types?
I think the m2 in kg/m2 refers to the grid cell area.

Patrick McGuire

comment:3 Changed 5 months ago by pmcguire

Hi Patrick,

Thank you so much for quick reply! I'm looking at vegetation and soil carbon (Cveg and Csoil) from TRENDY v8 outputs. And I'm using overall vegetation and soil carbon because I need to quantify the vegetation turnover rate using overall vegetation biomass carbon and NPP simulated by P model, and quantify soil respiration using soil carbon and soil respiration rate.

When I had extremely small values of vegetation carbon, for example, 0.1g/m2 vegetation carbon with 55g/m2 NPP (NPP simulated from P model) I got very large and unrealistic vegetation turnover rate. This is why I was wondering if there's some other reason, or just due to the disagreement between land masks/cover map we used. Fortunately this is seldom seen, but it would be good to get to the bottom of it.

Many thanks,

comment:4 Changed 5 months ago by pmcguire

Hi Wenjia
Are the units both g/m2 or kg/m2? You have 0.1g/m2. Should it be 0.1 kg/m2? What locations do you see these low values for?


comment:5 Changed 5 months ago by pmcguire

Hi Patrick,

The original unit for Cveg is kg/m2, I multiplied by 1000 to reach the same g/m2 as NPP I simulated. The majority of these low values are at the edge of continent and desert so this is why initially I though it's due to differences between land cover/mask. But my supervisor asked the question about the unit of Cveg and whether the m2 refers to grid cell or land cover. So I thought it would be nice to ask and confirm that. If the m2 refers to grid cell the same as other SDGVM outputs (as I've also used GPP outputs before), then the issue is solved.

Many thanks,

comment:6 Changed 5 months ago by pmcguire

Hi Wenjia
The units of NPP are kg/m2/s, not kg/m2, as you say.
Have you looked at the NPP from SDGVM? You said you looked at Cveg. But I also delivered the NPP output from SDGVM to the TRENDYv8 project.
The best that I can tell from looking at the source code for SDGVM, is that m2 refers to gridcell for the "biot" variable (bio total) that is delivered to TRENDY as Cveg.

The TRENDY SDGVM NETCDF files for npppft and gpppft have layers for each PFT, and for those, the units say (if you do an ncdump on the files) that it is "kg m-2 s-1, per unit land area occupied by the PFT".

Note: See TracTickets for help on using tickets.