Several typical loads are applied to muck specimens to simulate muck ground treatment conditions. Then tests of pore structural distribution of the soil are performed by the method of nuclear magnetic resonance (NMR), for seeking variation of their internal structure under the typical load level and rate, and thus to further determine the muck ground reinforcement mechanism and responses in micro-structure aspect. The results suggest that: (1) the max-pore decreases under a certain load, and the reduction increases with load level; nevertheless, when the impact load reaches a considerable level, both the max-pore and small-pore decrease; (2) the confining stiffness effect of constraint samples induces a decrease in the proportion of large-pore; (3) at a certain load level (680 kPa or less), the loading rate is the key factor determining the relative proportion of maximum void ratio: smaller lading rate will make the void ratio increase, larger lading rate will make the void ratio reduced, and its boundary value is between > 0.8 MPa/s and ≤ 1.6 MPa/s; (4) with a certain load level and rate, an the number of impact times increses, namely as the total energy is enhanced, the large-pore as well as the max-pore considerably decreases; however, this effect will be reduced when the interval time is short and as the number of impact times increases further. i.e. the large-pore effectively decreases. There exists a suitable value of the number of impact times for the muck specimens. These results reflect the rule of different loading effects in the micro of micro fine structue and provide a basis for the design and construction optimization for muck ground improvement.