The microbial community use the organic matter as their food and during consumption they release CO2 . This also causes break down of organic matter into different simple forms which also attracts other microbial community through succession. At this, most of the organic matter is converted into CO2 which emits from that environment, thus decreasing the concentration of organic carbon along with passage of time and biomass formation is however very less in amount , possibly due to P/R ( production/ Respiration) ratio less than one because of their activities in adverse situation. This could result in decrease in SOM.
Controls on microbial community structure are highly complex in soils. Soil architect at aggregate scale is recognized as one of the major drivers of microbial diversity in soils. There is a special issue in Frontiers in Env. Sci.
"Elucidating Microbial Processes in Soils and Sediments: Microscale Measurements and Modeling" worth looking into. Have a look at Iain Young 's and Claire Chenu 's work, you will find many more interesting examples.
SOM is one of the driving factors, but not the sole one. The increase of SOM usually accompanied with the change of other soil variables, which could increase or decrease microbial abundance and diversity. Thus, it is highly complicated. It's my viewpoint and hope it helps. Best wishes.
Let me know, would you like to know: "Why microbial community 'diversity' (instead 'structure') didn't increase with soil organic matter?"
To answer your question is necessary to think about two aspects: the organic matter quantity and quality. I mean, you need to know the nutrient pools of the OM and the number of different carbon sources present in your OM (i.e, the C sources diversity). This will be very important.
For example, you can increase SOM using a poor-nutrient material (high C/N relation, like sugarcane or eucalypt residues). Doing this, you not necessarily will affect positively the microbial community, because you added a high quantity of a specific C source in the soil. O the other hand, you can put a nutrient-rich material (low C/N relation, like soybean residues) and increase the microbial abundance exponentially, but not necessarily increase the microbial diversity. In the two cases, you will favor specific microbial groups only.
However, if you added different plant materials, or, naturally, different C sources derivated from the plant through root exudation, the possibility to increase both abundance and number of microbial groups is higher than above-mentioned practices.