For example: trees are constantly growing in a forest, and trees are also constantly dying, but the overall population size (number of trees in a forest) remains generally stable, in the absence of major changes in climate and animal predation of trees (such as clear-cutting forests).
A lot of uses of growth in economics appear to be related to growth of rates, not growth of something like mass. For example, carbon flows through grasslands at a greater rate than it does in forests; the carbon in a grassland was in the atmosphere at most a few years ago while a tree in a forest can be several hundred years old. The rate at which money flows through an economy (less misers, basically, mean higher transactions) is linked to economic growth.
However, growth in terms of growth of the number of buildings in a city and numbers of humans in a city, that's absolutely tied to the amount of arable land each human needs for food production, so obviously that's not an open-ended system, and in particular, if there's not enough water then the land isn't arable.
Again, if a city of fixed size improves the average quality of life for its inhabitants, then that's a kind of economic growth, isn't it? Even if resource use (in terms of conservation of mass and energy) is fixed, you can still have economic improvements within that steady-state system, can't you?