Most people know that deserts are classified for lack of watery precipitation because water fluctuations replenish groundwater, and that feeds vegetation, springs, streams, and so on. Stagnant water becomes crudded; dry earth becomes parched and loses the multitude of pores and other soil structures which convey water through the topsoil layers.
I've always preferred to be more ontologically concise and simple: an ‘aridland’ is one which has less water in some form or another than certain standard definitions; a ‘desert’ is one which is limited in habitability to below certain standard definitions. I.e.
This permits you to classify the following regions as deserts:
- smog–covered wastelands, where only intrepid mutants roam in few and small groups.
- seasonal inland salt lake which receives loads of rainwater but supports nothing except nanoscopic halophiles.
- an alien planet which is pristine and quite viable, but somehow devoid of life.
- similarly, there could be much and many fertile nutrients, but radiation results in sparsity of life and lifespans.
Similarly, these would probably not be considered deserts:
- smog–covered wastelands which are, never–the–less, densely populated.
- somewhere artificially irrigated in sufficient quantities and with competent ecological management that enables it to support a jungle, or hydroponics and vertical farming, or whatever.
- the aquatic frogfolk cities under the Europan ice ceiling.
Measuring the hospitality of a zone — the ease with which a certain group or class of lifeforms are able to thrive — should be an entirely different metric.
Ah, yes: you wanted a desert with “high humidity” and “moderate temperature”.
So, Arrakis but with much dissolved water in the air?
My suggestion relies on the composition of the terrain and earth.
So long as there is some water, there are species of vegetation which will take root with even very small amounts of vital minerals. They are variously called ‘pioneer plants’ or ‘weeds’. Eventually, you will get a parade — figure of speech — of animal critters and less–hardy vegetation which build upon the gradually increasing concentration of organic detritus and minerals brought in by the decomposition of the aforementioned parade.
Very simplified, but you get the idea.
Even very coarse earth suspended on a mesh grate over a vast subterranean river will eventually accumulate compacted crud on which mosses will be able to survive — and, so on.
If, however, the terrain is of some chemical composition which is either poisonous or simply incompatible with simple things such as lichens, then you'll never get anything to take root. Such a chemical would need to contaminate whatever soil is deposited — ergo, simple copper–plated pebbles probably wouldn't work for very long.
Of course, if you don't mind having a desert which becomes less of a desert due to recurrent traffic, then the copper pebbles could be adequate. Why copper? Most microbes, lichens, and mosses can't grow on cuprous surfaces.
Indeed, any unlimited accretion of foreign material which supports life would eventually dilute whatever chemical compels the desert to be so deserted.
Probably the best way to go, if a worldbuilder chose to use my suggestion, would be to have coppery pebbles and thorough drainage in a region. Use the quantities of water to flood and wash the area clean.