If plants become stressed later in the season, a time series of NDVI images could be a dramatic visualization of that event. Either agricultural or wild areas could be used to illustrate the event, although well irrigated areas might not show any effect. Interpreting the images will probably require comparing them to similar images from more normal years which you obviously don’t have yet. But monthly or weekly images of the same area could still raise awareness of the seasonal changes occurring there.
The predicted event of drought will be a regional phenomenon which could be best visualized with broader imagery. Data from the Modis sensor on the Terra satellite is used to produce NDVI results every day for your region. You can view or download 16 day averages of NDVI data (and also NWDI, which might be more relevant to drought) for the past dozen years for any area larger than 250 meters at http://pekko.geog.umd.edu/usda/test/. Making graphs or maps of these data available to your community could stimulate a conversation about the phenomenon. For example, the NDVI data for this spring in your region (I think that's where you are) has been a little below average, but not that much: http://bit.ly/KBektH. The NWDI has been more solidly below average: http://bit.ly/KBe6mm. You can play with that site and learn a lot about the recent history and geographic patterns of drought and plant growth in your area. Screenshots from the above links are below.
Combining that satellite perspective with more detailed kite or balloon views of NDVI at some important local area could be an effective way to get the community involved in the issue.