Using self-gravitational hydrodynamical numerical simulations , we investigated the evolution of high-density turbulent molecular clouds swept by a colliding flow .
The interaction of shock waves due to turbulence produces networks of thin filamentary clouds with a sub-parsec width .
The colliding flow accumulates the filamentary clouds into a sheet cloud and promotes active star formation for initially high-density clouds .
Clouds with a colliding flow exhibit a finer filamentary network than clouds without a colliding flow .
The probability distribution functions ( PDFs ) for the density and column density can be fitted by lognormal functions for clouds without colliding flow .
When the initial turbulence is weak , the column density PDF has a power-law wing at high column densities .
The colliding flow considerably deforms the PDF , such that the PDF exhibits a double peak .
The stellar mass distributions reproduced here are consistent with the classical initial mass function with a power-law index of -1.35 when the initial clouds have a high density .
The distribution of stellar velocities agrees with the gas velocity distribution , which can be fitted by Gaussian functions for clouds without colliding flow .
For clouds with colliding flow , the velocity dispersion of gas tends to be larger than the stellar velocity dispersion .
The signatures of colliding flows and turbulence appear in channel maps reconstructed from the simulation data .
Clouds without colliding flow exhibit a cloud-scale velocity shear due to the turbulence .
In contrast , clouds with colliding flow show a prominent anti-correlated distribution of thin filaments between the different velocity channels , suggesting collisions between the filamentary clouds .