Nope, petroleum does not include natural gas (though it does include byproducts such as petcoke); the data is straight out of the EIA historical generation data.
If we went to the monthly series it wouldn't surprise me if we've had crossovers already. I hadn't thought of celebrating about it, but if you do, feel free to hoist a beer for me.
found unusually low EROEIs for nuclear power I may have an answer for you.
Instead of the standard EROEI ratio Lenzen gives the ratio of kWh-thermal inputs required per kWh-electric generated.
It occurred to me that Lenzen may be counting the waste heat from generating electrical power inputs among his thermal inputs.
The nuclear fuel cycle is, of course, heavily dependent on electrical inputs, particularly for uranium enrichment.
If thermal power plants are about 33% efficient, then Lenzen could be increasing the size of these electrical inputs by about 3 times when converting them to thermal inputs.
At least one of the studies cited in Lenzen’s review is available online. http://www.world-nuclear.org/info/inf11.htm
For reference 9 at the end of table 13a Lenzen reports 0.171kWh-th/kWh-e for diffusion enrichment and 0.052 for centrifuge, both using ore from the Ranger mine.
Converting to a more standard EROEI by taking the reciprocal and then multiplying by 3 gives 17.54 for diffusion and 57.69 for centrifuge.
These EROEIs are almost exactly the same as the ones in the study referenced: 3020PJ/173PJ diffusion, 3020PJ/52PJ centrifuge.