In a Dec 2005 commentary on Feddema et al (2005) The Importance of Land-Cover Change in Simulating Future Climates. Roger Pielke Sr. writes on the role of land use change in shaping thunderstorms:

One example of how land use and land cover affects global climate is the changing spatial and temporal pattern of thunderstorms. Land use and land cover change and variability modify the surface fluxes of heat and water vapor. This alteration in the fluxes affects the atmospheric boundary layer, and hence the energy available for thunderstorms. As shown in the pioneering work of Riehl and Malkus and Riehl and Simpson, at any time there are 1500 to 5000 thunderstorms globally (referred to as “hot towers”) that transport heat, moisture, and wind energy to higher latitudes. Because thunderstorms occur over a relatively small percentage of Earth’s surface, a change in their spatial patterns would be expected to have global climate consequences. The changes in the spatial patterning of thunderstorms result in regional alterations in tropospheric heating that directly change atmospheric and ocean circulation patterns, including the movement and intensity of large-scale high- and low-pressure weather systems. Most thunderstorms (by a ratio of about 10 to 1) occur over land, and so land use and land cover have a greater impact on the climate system than is represented by the fraction of area that the land covers.

NASA has mapped global lighting strikes. The below image shows the global average annual occurrence of lightning at a resolution of ½° by ½°.

Compare this map against Gordon et al’s map of vapour flow changes, and it becomes apparent that some of the areas of strong vapour flow change are in areas of high thunderstorm activity.  It would be interesting to discover what effect the changes in land cover/land use are doing to thunderstorms and if this has any effect on regional/global climate.