2003ApJ...589..770S

We examine recent developments in the cluster cooling flow scenario following recent observations by Chandra and XMM-Newton. We show that the distribution of gas emissivity versus temperature determined by XMM-Newton gratings observations demonstrates that the central gas, when the cooling time is less than the age of the cluster, in cooling flow clusters cannot be in simple steady state; i.e., M{dot} is not a constant at all temperatures. On the basis of the measured gas emissivity, the gas can be in steady state only if there exists a steady heating mechanism that scales as H(T)∝T^α, where α=1-2. A heating mechanism that preferentially targets the hottest and highest entropy gas seems very unlikely. Combining this result with the lack of spectroscopic evidence for gas below one-third of the ambient cluster temperature is strong evidence that the gas is heated intermittently. While the old steady state isobaric cooling flow model is incompatible with recent observations, a moderate cooling flow model in which the gas undergoes intermittent heating that effectively reduces the age of a cooling flow is consistent with observations. Most of the gas within cooling flows resides in the hottest gas, which is prevented from cooling continuously and attaining a steady state configuration. This results in a mass cooling rate that decreases with decreasing temperature, with a much lower mass cooling rate at the lowest temperatures. Such a temperature-dependent M{dot} is required by the XMM-Newton RGS data and will produce an increasing amount of intermediate-temperature gas that will then be reheated during the next heating cycle. We show the compatibility of this model with the cooling flow cluster A2052. This paper strengthens the moderate cooling flow model, which can accommodate the unique activities observed in cooling flow clusters.