I embarked on creating a Minimum Working Example (MWE) script, and in the process, I stumbled upon the root of the issue. The COUNTYFP column had an object data type, causing matplotlib to generate discrete colors. Strangely, the orientation parameter only seems to function for colorbars. Once I converted COUNTYFP to numeric data, the problem was resolved.
Here's the code snippet that demonstrates the fix:
import geopandas as gpd
import matplotlib.pyplot as plt
# Read shapefile
data = gpd.read_file("path/to/shapefile.shp")
# Convert COUNTYFP column to numeric
data['COUNTYFP'] = data['COUNTYFP'].astype(int)
# Plot choropleth map
data.plot(column='COUNTYFP', legend=True, legend_kwds={'orientation': 'horizontal'})
plt.show()
In summary, the culprit was the object data type of the COUNTYFP column. By converting it to numeric, matplotlib was able to generate a continuous colorbar, and the orientation parameter functioned as expected.