What resolution of static data should be used?

[g_vijayan]

Hi,
I am using three two-way nested domains of resolutions 9km, 3km and 1km.
Static data is available upto 30s resolution. How do I decide what resolution data should be used with each of my domains? Is it okay to use static data with resolution much finer than the domain resolution? What is the resolution of static data used when geod_data_res is set to 'default'? I could not find anything regarding that in the userguide.
What happens if our static data resolution is much coarser or much finer in comparision to our domain resolution ?

Kindly help me out.
Thankyou.

 

[Ming Chen]

Hi,
Please stay with the default option. When you look at GEOGRID.TBL, you will find what dataset the 'default' option point to.
It is totally fine if the static data resolution is higher than the model resolution, because WPS takes care of this issue and gives representative value for each grid box. If the static data resolution is coarser than the model resolution, WPS will interpolate the static data to model grids. In this case, due to the low resolution of static data, surface information may not be accurate, and you will lose some details.

 

[mmah1986]

Hi,

Ming: You mentioned that "It is totally fine if the static data resolution is higher than the model resolution, because WPS takes care of this issue and gives representative value for each grid box.". However, generally, if domain data is interpolated from a source that has much higher resolution, this could cause the resulting output to be too noisy. Does WPS geogrid include a mechanism (other than interpolating with interpolation schemes defined in the geogrid table) that alleviates high resolution contrasts that I just mentioned?

Thanks,
Mani

 

[mgduda]

Besides the selection of appropriate interpolation options in the GEOGRID.TBL file, there isn't a secondary mechanism to treat large differences in the resolution of source data and WRF grids. Note, though, that the "average_gcell" interpolation option (described in the Available Interpolation Options in Geogrid and Metgrid section of Chapter 3 of the ARW User's Guide) does provide a way to average the source data to the scale of the WRF grid. And for categorical datasets, the "nearest_neighbor" interpolation option actually considers all pixels of the categorical field that lie within each WRF grid cell when computing the dominant category and fractional values of categories (you can see this code in the accum_categorical routine in the geogrid program).

It's also worth noting that several interpolation options can be specified with the interp_option keyword in the GEOGRID.TBL file for a field, so that a single GEOGRID.TBL entry should be suitable for WRF grids that are much coarser than the source data as well WRF nested grids that are higher in resolution than the source data. For example, setting

interp_option = average_gcell(4.0)+four_pt

would instruct the geogrid program to use grid-cell averaging when the source dataset is more than four times higher in resolution than the WRF grid, but to use a four-point bilinear interpolation option otherwise.