这俩变量有什么不同 按官网的解释看起来是坐标系不同,但实际有什么区别还是不清楚
https://www.heywhale.com/mw/notebook/655ec29e998f42fde5818d35
以上解释得很清楚了,就不重复做了
在WRF中,温度剖面图(Sounding)通常使用eta层数据来绘制。eta层数据通常是从地面开始的一系列等间隔垂直高度层,这些高度层往往非常密集,尤其是在近地面层。然而,由于地形高度变化的存在,eta层在地形上可能会产生变化,导致某些高度层没有数据,进而出现图中近地面空白区的情况。 目前的解决办法是对nan值进行替换,具体可以参考wrf-python官方示例的处理
# To remove the slight gap between the dbz contours and terrain due to the
# contouring of gridded data, a new vertical grid spacing, and model grid
# staggering, fill in the lower grid cells with the first non-missing value
# for each column.
# Make a copy of the z cross data. Let's use regular numpy arrays for this.
dbz_cross_filled = np.ma.copy(to_np(dbz_cross))
# For each cross section column, find the first index with non-missing
# values and copy these to the missing elements below.
for i in range(dbz_cross_filled.shape[-1]):
column_vals = dbz_cross_filled[:,i]
# Let's find the lowest index that isn't filled. The nonzero function
# finds all unmasked values greater than 0. Since 0 is a valid value
# for dBZ, let's change that threshold to be -200 dBZ instead.
first_idx = int(np.transpose((column_vals > -200).nonzero())[0])
dbz_cross_filled[0:first_idx, i] = dbz_cross_filled[first_idx, i]
效果如上
在WRF中,parent_grid_ratio参数是用于定义父域和子域之间的水平分辨率比例。一般情况下,建议将parent_grid_ratio设置为奇数,例如3或5,而不是偶数。
这是因为在WRF模型中,父域和子域之间的网格点位置是通过插值计算得到的。当parent_grid_ratio为奇数时,插值计算相对容易,因为可以使用更多的父域网格点来计算子域网格点位置,不会出现网格点位置落在父域网格点之间的情况。而当parent_grid_ratio为偶数时,插值计算会更加复杂,因为子域网格点需要考虑如何与父域网格点进行插值,可能会导致数据插值误差的增加。
此外,将parent_grid_ratio设置为奇数还有助于避免在边界处产生奇异点(singular point)的问题。奇异点是指在网格交错区域出现的数值不稳定现象,可能导致数值计算错误。
总之,使用奇数的parent_grid_ratio可以减小插值误差,并提高模拟结果的准确性和稳定性。因此,建议在WRF模型中将parent_grid_ratio设置为奇数,如3或5
from chatgpt
xesmf库或者ncl里面的rcm2rgrid_Wrap函数可以实现
xesmf库使用可以参考我往期推文关于WRF插值站点的二三事
LWC计算: 说法不一,有人说是QCLOUD ,有人说用大气物理学书上的公式
气象家园截图
能见度计算:方法一:
https://www.heywhale.com/mw/project/6210d9b368364e0017a4659f
方法二:
Hi, You can always use a formula to convert liquid water content (LWC) (QCLOUD) at the lowest level into visibility. See for example the formulas given in the next paper: Kunkel BA. 1984. Parameterization of droplet terminal velocity and extinction coefficient in fog models. J. Clim. Appl. Meteorol. 23: 34–41. Best regards. Carlos 2016-02-16 17:09 GMT+01:00 Tiago Luna <tiagoluna at ua.pt>: Some years ago I came across with this work: https://www.researchgate.net/profile/Ji_Woo_Lee/publication/277004271_Predictability_Experiments_of_Fog_and_Visibility_in_Local_Airports_over_Korea_using_the_WRF_Model/links/555d00b608ae8c0cab2a6aef.pdf Please take a moment to read it and search for some of it's references. I don't know if there is any recent work about this. Regards, Luna On 15-02-2016 05:27, Sundar J wrote: As the subject suggest, is there any formula exist for calculating surface visibility? I have seen some of the weather sites which shows the visibility plots (mostly from GFS as input data). Is there any way that to plot the surface visiblity parameter using parameters from wrfout file? I have read some of the some of the scientific paper which indicates none of the models has visibility as one of the output parameter. Why does it so? Regards
The formula for calculating Liquid Water Content (LWC) in the WRF model is not explicitly provided in the standard output. However, you can calculate LWC using the cloud water mixing ratio (QCLOUD) from the model output. The LWC can be calculated by multiplying QCLOUD by the density of air (ρ) at the corresponding level: LWC=QCLOUD×ρ Where: ( LWC ) is the Liquid Water Content in ( g/m^3 ) ( QCLOUD ) is the cloud water mixing ratio in ( g/kg ) ( \rho ) is the density of air in ( kg/m^3 ) Please note that QCLOUD is a standard variable in the WRF model output, and the air density can be calculated from the model’s pressure and temperature fields using the ideal gas law1. For visibility calculations or other specific applications, you may need to refer to specialized formulas or research papers that provide parameterizations for converting LWC to the desired units or for use in specific models1. For example, Kunkel (1984) provides parameterizations for droplet terminal velocity and extinction coefficient in fog models, which may be relevant depending on your application1. Remember to check the specific details and assumptions of the WRF model configuration you are using, as these can affect the calculation of LWC. by newbing
在WRF模型的输出文件中,"north-south-stag"和"north-south"是与垂直方向(南北向)网格点相关的变量,它们之间的区别在于:
在处理这两种类型的变量时,需要根据具体的需求和使用情境进行处理。一般来说,如果需要进行插值或者计算某些物理量在实际数据点上的值,通常会使用"north-south"类型的变量。而如果需要处理网格节点之间的物理量取值,可能需要考虑使用"north-south-stag"类型的变量。
处理这些变量的具体方法可能涉及到插值、网格点对应关系的处理等操作,具体的处理方法可能取决于你所使用的数据处理工具或编程语言以及具体的需求。
from chatgpt
已有公开的数据集与程序,网址:https://www.scidb.cn/en/detail?dataSetId=791587189614968832 程序为Fortran,需要的自行探索
https://wrf-python.readthedocs.io/en/latest/plot.html
以上为官网示例
meteva的showdoc有格式转换的介绍,用bing搜meteva第一个网址就是
订正WRF模式输出的风场数据是一个很常见的问题,尤其对于风电场的运营和管理非常重要。以下是一些可能有用的方法:
无论采用哪种方法,订正模式输出的风速都需要有足够的实测数据来验证修正的效果,并且需要进行大量的试验和调整以获得最佳结果。建议在使用任何方法之前,先对模式输出数据和实测数据进行详细的分析,以了解偏差的来源和性质,并根据具体情况选择合适的订正方法。
from chatgpt
可以修改,可以用restart文件,restart相关讨论
https://bbs.06climate.com/forum.php?mod=viewthread&tid=36404