We present a new procedure to measure the large-scale carbon monoxide ( CO ) emissions across cosmic history .
As a tracer of large-scale structure ( LSS ) , the CO gas content as a function of redshift can be quantified by its three-dimensional fluctuation power spectra .
Furthermore , cross-correlating CO emission with other LSS tracers offers a way to measure the emission as a function of scale and redshift .
Here we introduce the model relevant for such a cross-correlation measurement between CO and other LSS tracers , and between different CO rotational lines .
We propose a novel use of Cosmic Microwave Background ( CMB ) data and attempt to extract redshifted CO emissions embedded in the Wilkinson Microwave Anisotropy Probe ( WMAP ) dataset .
We cross-correlate the all-sky WMAP7 data with LSS data sets , namely , the photometric quasar sample and the luminous red galaxy sample from the Sloan Digital Sky Survey Data Release 6 and 7 respectively .
We are unable to detect a cross-correlation signal with either CO ( 1-0 ) nor CO ( 2-1 ) lines , mainly due to the instrumental noise in the WMAP data .
However , we are able to rule out models more than three times greater than our more optimistic model .
We discuss the cross-correlation signal from the thermal Sunyaev-Zeldovich effect and dust as potential contaminants , and quantify their impact for our CO measurements .
We discuss forecasts for current CMB experiments and a hypothetical future CO focused experiment , and propose to cross-correlate CO temperature data with the Hobby-Eberly Telescope Dark Energy Experiment Ly \alpha -emitter sample , for which a signal-to-noise ratio of 58 is possible .