The city’s population is growing, but by just how much seems to be a subject of some disagreement.

Population estimates done annually by the U.S. Census Bureau were released this week. The population estimate for Suffolk as of July 1, 2016, was 89,273.

But in January, the Weldon Cooper Center for Public Service at the University of Virginia estimated the city’s population as of July 1 to be 91,722.

That’s a difference of 2,449 Suffolkians on the same day. So what gives?

The missing residents come from the methodology used to determine the estimates.

The Weldon Cooper Center’s population estimates are produced using housing stock, school enrollment, births, deaths and licensed drivers.

The U.S. Census Bureau, however, uses only births and deaths, as well as migration, to figure the population change from the last census.

“We estimate the resident population for each year since the most recent decennial census by using measures of population change,” the Census website says.

The census estimates start with the last decennial census — in this case, the one conducted in 2010 — add births, subtract deaths and then add net migration.

About 7,036 births and 4,583 deaths were estimated for Suffolk since 2010, as well as a total of 2,141 migrations — 730 international and the rest from other localities in America.

The 2010 census counted 84,585 Suffolk residents.

While the census estimates still don’t have Suffolk above 90,000 residents, the Weldon Cooper Center first estimated the city topping that number last year.

Both the census and the Weldon Cooper Center numbers are used in various official capacities. The census estimates are used in many federal funding formulations. But the Weldon Cooper numbers are considered the official ones for the state and are also used in funding formulations based on per capita allocations, as well as in planning, budgeting and applying for grants, among other tasks.