Use of astronomy is a good litmus test (because it leaves clear archaeological evidence) of technology levels in early New World civilizations. Calendars appear to be key to agriculture in any place that has seasonal climate variation.
Chaco does indeed have the earliest datable evidence for astronomical alignments in the northern Southwest, and possibly in the Southwest as a whole. (There is some possible evidence for earlier alignments among the Hohokam of southern Arizona, but it’s somewhat questionable.) This provides some tentative support for the theory.
I don’t think it’s likely that the Chacoan leaders developed their astronomy on their own, though. There is plenty of evidence for contact and communication between them and Mesoamerica, though it isn’t always clear how direct this may have been (as opposed to indirect and mediated through groups in between such as the Hohokam). The much more complex societies of Mesoamerica also had much more elaborate astronomical and calendrical systems than anyone in the Southwest, so they are an obvious source for this as well.
They also presumably developed their knowledge earlier, so as I was thinking about my Chaco theory it occurred to me that it would be good to look into when exactly astronomical alignments and other evidence of this knowledge appear in Mesoamerica and how they spread and changed over time. Basically, the question is whether what is known about the origin and spread of astronomical knowledge in Mesoamerica is consistent with what appears to be true of the origin and spread of similar knowledge further north. Also, it would be helpful to know just how similar the alignments and other phenomena known from Mesoamerica are to those in the Southwest, again to judge the plausibility of a connection.
Luckily for me, an article published last year addressed this exact issue. Written by the Slovenian scholar Ivan Šprajc, it was published in the Journal of Archaeological Research and discusses the temporal and spatial distribution of different building alignments in Mesoamerica. . . . You might expect alignments to the summer and winter solstice sunrises and sunsets to be common, and they were to some extent, but they were by no means the most common. (Alignments to cardinal directions were also present but were even less common.) Much more common, especially in the Maya region, were alignments to certain points on the horizon that do appear to reflect particular sunrises and sunsets, but on different days than the solstices. The specific days cluster in February and October for sunrises and April and August for sunsets. Based on comparisons to ethnohistoric and modern ethnographic accounts of agricultural cycles, Šprajc proposes that these dates marked significant points in the cycle of planting and harvesting cycle, especially for maize, and that marking them would have been part of a very practical system of timekeeping that would also presumably have had ritual importance.
Furthermore, the numbers of days separating many of these dates that pattern together at particular sites tend to reflect multiples of 13 and 20, which are key numbers in the Mesoamerican calendar system, particularly in the 260-day ritual calendar. (Note that 260 is 13 times 20.) Based on the practices of some modern Maya communities that still measure their agricultural cycles this way, it appears that the alignments to mark the key dates would have allowed people to count from those points to figure out the rest of the cycle using these intervals. Since the same dates recur at these intervals in the ritual calendar, which is not calibrated to the solar year, people could have easily used them to keep track of the times for specific activities without worrying about a general calibration.
As a simplified example, if the alignment of a building in a community marked the beginning of the planting season based on the position of the sun, and the community knew that the harvest would come 260 days later, they could take note of the ritual calendar date (number and day-sign) of the beginning day marked by the alignment, the correspondence of which to the solar calendar would vary from year to year, and know that when that date came up again it would be time for harvest. This seems to me like a clever way to deal with the eternal problem of calibrating a solar calendar to seas.
From Gambler's House which explores the details of the Meso-American developments in more depth.