air quantity

You need to explain more about what your exact need is. Is it for preliminary design consideration, is it for process control, is it for process evaluation?
You also need to explain what your exact information are: it is hard to believe that you would only have power and rpm. Don't you have technical data from the manufacturer, for example?

Stricktly speaking, if you only have power and rpm measured on an operating fan, you can't determine the air flow.

For a simple estimation,
assuming the temperature increase is small enough, you can just assume this elementary relation:

   P = h Q_v  Dp         (1)

where 

   power (P) is in W, 
    h is the efficiency, 
   Q_v is the volume gas flow in m³/s,
    Dp is the pressure boost delivered by the fan in Pa

The efficiency of fan might be typically between 20% and 30%, but will depend on the working point and the fan model.

I like to remind people that Pa = J/m³ , so that it is easy to check that the units in the formula (1) are consistent.

From equation (1), you can see what would be needed to evaluate the gas flow.

For a more detailled evaluation,
you should use the fan datasheets provided by the manufacturer. The minimal information usually available are the P(Q_v) and Dp(Q_v) curves for a given rpm and a given gas density  r . (the efficiency can also be evaluated from these data for any operating point). If your operating point is different from the datasheets, then you need to rescale the datasheets curves using the affinity laws as described on wikipedia or on pontyak.com. You can then read the gas flow from the properly scaled  P(Q_v)  curve. In this case, the rpm will indeed be needed, but also the density of the gas to be evaluated from the temperature.

A pending question is: should the amount of dust be taken into account when evaluating the density. I believe the answer is yes, but I am not sure. After all, the dust velocity might be unchanged after passing through the fan, leading to no effect on the energy balance.