Either dedicated repeaters or AP's will do it, but your variable either assumed or forgotten is obstruction and height.
Omnidirectional AP's are always best served the higher you can get them in this case since you don't seem to mention the option of elevating hundreds of feet off the ground and end users.
If this campground is thick foliage, you have to factor vegetation such as what types of line of sight "alleys" you'll faithfully have by seasons and wind.
I know this from my experiences pen testing WiFi from various surreptitious distances and my effective range worsens when seemingly smaller / thinner obstructions interrupt longer range probes.
Good site testing will help make the right decisions even if you can't take readings over the entirety of seasons, assuming you're uninitiated but comfortable with LINUX in monitor mode, a basic copy of Kismet will provide accurate readings.
Testing clients with AP's and WireShark on LINUX can output decibel levels by client (DOES Windows version of WireShark too, didn't see the option but didn't try to hard to determine if Windows version does)
Even Windows either NetStumbler or WiFiNetwork Monitor by Nirsoft gives detailed more than anecdotal signal readings than the five bars since your variance is literally 20% but worst, the Windows WiFi meter is delayed and you're reading the signal strength a few seconds ago than anything resembling real time.
Best testing in my opinion is high noon or bright sunny days when the sun is typically brightest, WiFi is microwave and bright sun can hamper signals, capacity planning for this time of day can go far ensuring the rest of the users days will have ample coverage.