regCoef返回的有个属性就是t检验的值,直接通过查表看一下是否通过检验即可,不需要再次进行val = onedtond(rcc@tval , dimsizes(rcc)) ;t-statistic
df = onedtond(rcc@nptxy, dimsizes(rcc)) - 2 ;自由度
b = tval ; b must be same size as tval (and df)
b = 0.5
prob = betainc(df/(df+tval^2),df/2.0,b)
的计算
stres = True ; set up a second resource list
stres@gsnDraw = False ; do not draw the plot
stres@gsnFrame = False ; do not advance the frame
stres@cnLineLabelsOn = False ; do not use line label
stres@cnFillOn =False ; color fill
stres@cnLinesOn = False ; do not draw contour lines
stres@cnInfoLabelOn = False
stres@lbLabelBarOn = False
plots_11=gsn_csm_contour(wks,tval,stres)
plots_11=ShadeLtGtContour(plots_11,-tt,17,tt,17); fillpattern=17
overlay(plot,plots_11)
我懂你的意思了,应该是这样的,这次是有点,只是打点的位置明显有问题,传不上图片。