Possible reasons are,
Loose fitting of casing pipe. The seepage water from the top loose formation while trickling down will carry the colloidal clay particle and enter into the bore well through the gap between seating and pipe end. The trickling would be a flow if you have met with good water yielding zone in the loose weathered part. This over the nonpumping period will fill the bore along with the source water from the other zone. If the static water level stands above the casing bottom then the flow will be less.
You might have got a sheared and fractured zone as your yielding zone for the bore well at any depth. The crushed and powdered sheared zone will dissolve the loose clay like soil when the ground water enters into bore well and this powdered particle will come out as mud while you pump out the water.
If you use a pump with higher capacity than the yielding rate or close to the yielding rate of the bore well, then the pump will suck more water from the storage zone. Due to high upward sucking rate, the mud settled at the bottom of the bore well will come out at the end of each cycle when the water level comes close to your pump. Normally some mud flow occurs during early minutes immediately after the start of the pump and seizes, when the contribution from the aquifer starts.
If you have obtained a soft weathered zone after the hard rock zone and if the water yield zone falls in the soft weathered zone at the bottom then also there is possibility of the inherent clay colloidal entering into the bore well. If it is in such a zone then, the solution is constructing a tube well with tiny pebbles and slots.
Contributor: A. Raja Mohamed (https://www.indiawaterportal.org/users/rajamohamed-ambalam)