I looked a little more closely at the average annual inflows and outflows of the three major reservoirs above Lake Powell: Flaming Gorge, Navajo, and Blue Mesa. The water database only shows data back to 1969 for Blue Mesa and Flaming Gorge, and back to 1964 for Navajo. So I focused on data since 1969, which we have for all three reservoirs.
This is the hypothesis, put in the form of a question: "Are the reservoirs above Lake Powell impeding Lake Powell's ability to refill?" To test that hypothesis, we simply need to compare annual inflows and outflows from each reservoir. If the inflows exceed the outflows, those reservoirs are holding something back that might otherwise reach Powell. If those numbers are the same, there is no effect on Lake Powell. I'd expect average annual inflows to be slightly greater than outflows, or else those reservoirs would quickly be drained.
Here's the data, for different periods since 1969:
Net Average Annual Inflow by Reservoir (acre-feet)
Flaming Gorge
1969-80 - 159,000 af
1981-99 - 106,000 af
2000-23 - 27,000 af
1969-2023 - 83,000 af
Blue Mesa
1969-80 - 23,000 af
1981-99 - 12,000 af
2000-23 - 11,000 af
1969-2023 - 14,000 af
Navajo
1969-80 - 101,000 af
1981-99 - 179,000 af
2000-23 - 222,000 af
1969-2023 - 181,000 af
All 3 Reservoirs Combined
1969-80 - 283,000 af
1981-99 - 297,000 af
2000-23 - 259,000 af
1969-2023 - 277,000 af
Well, the first thing you can say from this is that by looking at the "combined" data, the amount held back above Powell has been very consistent since 1969--an average of about 277,000 af. Sounds like a lot, but in the context of releasing about 8.2 maf from Powell through GCD each year, the input only amounts to about 3% what is released from GCD. Put another way, 277,000 af amounts to about 5 feet in Lake Powell, assuming the lake is roughly half full. That volume means less when the lake is closer to full, and means more when it is down. But about 5 feet. Of course, that's a meaningless statistic, since the USBR can easily modify that number simply by holding back 277,000 af from being released down through GCD in any given year. Bottom line is that the reservoirs above Powell do next to nothing to impede Powell being filled when viewed in the big picture.
That said, there are interesting trends that come out of the data. For example, the amount "held back" in the upper reservoirs during the period when Lake Powell was approaching full (1969-80) was almost 10% more than the amount held back during the sustained drought of 2000-23. This suggests that if anything, USBR has been trying to help Powell since 2000, not hurt it.
It's also worth noting the relative amount held back in each reservoir. Of the three, by far Navajo holds the most back each year, and that's been especially true since 2000. But at the same time, substantially more has been released (relative to inflow) from Flaming Gorge since 2000 than any period before that.
USBR has its management methods, but one thing you can't say is that their operation of the upper reservoirs in the past two decades has been hurting Lake Powell relative to how management was done in the past.