Any planet curves, but I will use the Earth since those calculations are public & easy . . . at sea level, a observer on a tower that is 30m off the ground can see out to a bit over 12 miles- and that is with a clear horizon. So IF a DS is landing 15 miles away from such a positioned radar station, it means the DS will not be observed landing and all the defenders have will be a direction. With a 5/8 mech, that means they can be at the target in 30 minutes or less if it runs- like closer to 20 minutes. Anyone want to guess what a prepared response time for defenders might be? Btw, for a person at sea level looking out to see? The 6 foot tall person can see a bit above 3 miles to the horizon.
Further, take that 15 miles away, it means the DS can divert from its initial trajectory while under the horizon from the radar site. The 12 miles would also not take into account ground scattering effect as the radar tries to work on the horizon. IF the DS landed closer the defenders would only have a bearing and a guess at range, to actually plot the location of ANY inbound contact it needs to be followed by at least 3 radar stations so its actually location be pinpointed by triangulation.
That's not wrong, but not the entire picture. A radar tower that's 30 meters high looking for a DS that's also 30 meters high will have a significantly wider area of coverage than if only one of them is substantially taller than ground level. Until a DS touches down, it's going to be some significant distance ABOVE the ground, and you don't dare fly as ungainly as a DS at Nape-of-Earth altitude for too long, unless you don't mind an extremely high risk factor for such a valuable asset. That 12 mile radius suddenly starts increasing to 30 miles or more, although hills and mountains could of course decrease that. The 20 minute race to target is now a half-hour or longer.
The point is that the defenders will be able to track the DS to low-altitude, and will know that the DS can't be TOO far from where it dropped beneath radar coverage, so overflights should have a decent chance of finding it, unless there are very significant terrain obstacles or it's WELL hidden somehow. It takes 3 radar stations to pinpoint its exact location, but two will give you a line along which it must be, with a decent reading of distance (time lapse from send to reception of a return "ping" gives a pretty accurate range), and even one station will at least give you a general direction and pretty good distance. That's sufficient for vectoring in air searches, either by conventional aircraft, VTOL, or ASF, and unless the LZ is extremely flat, the DS's own ASFs (if it's carrying any) will very likely have issues with takeoff and landing on unprepared ground in order to do anything about it.
Basically, the defenders will very likely know roughly where to look, if it's feasible to do so, and hiding something that large can only be accomplished in certain circumstances, because from the air it will stand out pretty clearly from a distance in any kind of open terrain. Whether or not they have the resources to conduct that search, or to do anything about it if they find the DS, will depend on the exact circumstances. Nearly all of the tricks mentioned in the various posts are merely ways of buying that little bit of extra time for the raiding party to complete its mission and get back before serious firepower can be called in, not ways of hiding a DS for an extended period of time in the face of an opponent who is aware that it's present.
The further from the target the DS lands, the harder it will be to find, and the longer it probably has until it's in danger, but the longer it will take the raiders to reach the objective. Basically, it all comes down to the exact situation at the time of the raid, so ground recon BEFORE the raid will be essential for choosing an optimal landing site and judging the response time of the defenders.
