You can always reduce the volume in software before sending it to whatever API plays your sounds. This is especially easy if you're synthesizing the sound with the desired output volume known.
It should be noted that this operation (digital gain staging) reduces headroom/resolution, although in practice this doesn't really matter. Especially not with ambient rain sounds.
