如何从二维数组中极值绘制等值线？

Question

如何在极坐标图中高亮显示一个特定的轮廓值？

azi是极坐标方位角的度数（有360个值）。dev是水平方向的偏差度数（有90个值）。W是一个二维数组，形状为(90,360)，数值范围为0到110。

我需要高亮显示等于90的W值，并对其进行轮廓处理。以下是我试过的代码。

---

a=np.radians(np.linspace(0,360,360).round()) # borehole azimuth - degrees converted to radians
d=np.radians(np.linspace(0,90,90).round())     # borehole deviation - degrees converted to radians
azi,dev=np.meshgrid(a,d)  # create numpy array 90x360 for borehole deviation
W=np.random.randint(80,100, size=(90,360))

ax2 = plt.subplot(111, projection= 'polar')  
ax2.set_theta_direction(-1)  
ax2.set_theta_zero_location("N")  
data=ax2.contourf(azi,dev, W)  
plt.contour(azi, dev, np.where(W == 90))  
plt.colorbar(data)  
plt.show()

---

但是，我得到了一个错误的结果：

TypeError: Input z must be at least a 2x2 array.  

我不知道如何对等于90的2D - W数组值进行轮廓和索引。

另一种解决方案可能是创建一个新的W==90值的2D数组，并对这些值进行轮廓处理。

Answer 1

import numpy as np
import matplotlib.pyplot as plt

a=np.radians(np.linspace(0,360,360).round()) # borehole azimuth - degrees converted to radians
    d=np.radians(np.linspace(0,90,90).round())     # borehole deviation - degrees converted to radians
    azi,dip=np.meshgrid(a,d)  # create numpy array 90x360 for borehole deviation
    W=np.random.randint(80,100, size=(90,360))

# first we need to create a new 2D Numpy array (Z) from W where W=90
Z = np.zeros_like(W)
mask = np.isclose(W, 90)
Z[mask] = W[mask]

ax2 = plt.subplot(111, projection= 'polar')  
ax2.set_theta_direction(-1)  
ax2.set_theta_zero_location("N")  
data=ax2.contourf(azi,dip, W)  
plt.contour(azi, dip, Z)  # then plot contour using the new Z array
plt.colorbar(data)  
plt.show()