grid independence in WRF

Hi
I want to investigate "grid independence" in WRF (grid size: 2km to 12km for local simulation). I run my model. when I checked WRF outputs (velocity and temperature), there wasn't any independence in my outputs. Could you guide me about it, please?
I attached the files.
Thank you

Hi,
I first would like to apologize for the delay in response to this question. It seems this post was overlooked. As it has been so long since you posted, have you moved on from this, or would you still like assistance? If you'd still like our help, will you please be a bit more descriptive with your question? What are you expecting to see? What are the differences in the runs, etc.?

Hi kwerner,

Thank you for your reply. I want to use WRF to predict meteorology parameters in the urban domain. I want to use a smaller resolution. As you see in the pictures, in x=2000 m, the wind speed is close to observation, but the temperature is smaller than observation and for x=10000m we have the best performance for temperature.
At the tables: O=the means of observations, M= the mean of the model predictions, MB=the mean bias,r=the correlation coefficient, MAE=mean absolute gross error, IA=index of agreement, and RMSE=root mean square error.
How can I choose the best grid resolution for my simulation? Could you help me, please?
Thank you

Hi,
Unfortunately discrepancies in simulation outcome is just a part of modeling. As it's not the real environment, we can't always make it exact, and you may have to choose a set-up that best meets the criteria that you are interested in. You may have to play around a bit with the size of the domains, the resolution, and perhaps even the physics settings to find the best outcome that you can work with. There is no magical all-inclusive answer. It may be different for each case.