此範例處理降雨估計資料-QPESUMS過去1小時定量降雨估計格點資料 O-B0045-001¶
Data Source: CWB opendata
已下載xml或json檔案
from cwbplot import opdata
xmlobj = opdata.read_xml("O-B0045-001.xml",)
#會回傳tuple物件，分別為(降雨估計值, 經度格點, 緯度格點, 觀測時間)
jsonobj = opdata.read_json("O-B0045-001.json")
#會回傳tuple物件，分別為(降雨估計值, 經度格點, 緯度格點, 觀測時間)
若想要全自動化，直接使用api
from cwbplot import opdata
jsonobs = opdata.json_api("O-B0045-001","/home/staffid/cwbopdata/readme.api")
#會回傳tuple物件，分別為(降雨估計值, 經度格點, 緯度格點, 觀測時間)
畫個圖
import matplotlib.pyplot as plt
from cwbplot import cwb_colorbar, opdata, projinfo
qpe, lon, lat, obstime = opdata.json_api("O-B0045-001","readme.api")
fig, axs = plt.subplots(1,2)
raincm = cwb_colorbar.rain()
for cnt, ax in enumerate(axs.flatten()):
proj = projinfo.wrfd(lon=lon, lat=lat, res = "f", domain="d02", cut = [lon[0,0],lon[-1,-1],lat[0,0],lat[-1,-1]], ax=ax)
if cnt == 0:
graph = proj.imshow(qpe, cmap=raincm["cmap"], norm=raincm["norm"], origin="lower")
plttype = "imshow"
else:
graph = proj.contourf(lon, lat, qpe, cmap=raincm["cmap"], norm=raincm["norm"], levels=raincm["levels"], latlon=True)
plttype = "contourf"
proj.drawcoastlines()
proj.drawmeridians([120,122], color='dimgray', labels=[0, 0, 0, 1], yoffset = 0 )
proj.drawparallels([22,24,26], color='dimgray', labels=[1, 0, 0, 0], xoffset = 0 )
cbar = plt.colorbar(graph,ax=ax)
cbar.set_label("mm")
ax.set_title(f"1 hour QPE {obstime}\n plot type {plttype}")
plt.savefig("O-B0045-001.png", format='png', bbox_inches='tight', dpi=250)
圖如下
