The most expensive mistake a Hong Kong gardener can make does not involve plants. It involves ignoring what is underneath them. A garden is, before it is anything else — before it is a collection of specimens, a colour scheme, a view, a place to drink coffee on Sunday mornings — a soil system. Get the soil right, and the plants largely take care of themselves. Get it wrong, and you will spend the rest of your gardening life compensating for a problem you could have fixed in the first month. Every wilted leaf, every yellowed margin, every shrub that should be two metres tall and is instead a metre of disappointment — trace it back far enough, and you will find soil that was never given a chance.
I have killed enough plants to know this. Not through neglect; through the earnest, well-intentioned ignorance that mistakes activity for progress. I have planted Japanese maples in alkaline fill and watched them chlorose to death over three slow summers. I have buried expensive roses in compacted subsoil and wondered why they produced three blooms and then gave up. I have spread fertiliser like a dispensation, convinced that the problem was nutritional, when the problem was structural — the soil could not breathe, could not drain, could not host the microbial life that turns minerals into meals. The plants were starving in a larder full of locked food.
This article is about soil. Not soil in the abstract — not “the living skin of the Earth” or any of the other reverent phrases that sound lovely and tell you nothing — but the actual stuff under the actual feet of actual Hong Kong gardens. What it is. Why it is like that. What you can do about it. And, critically, what you cannot.
What Hong Kong Soil Actually Is
To understand Hong Kong soil, stand on any hillside in the New Territories and look down. What you are standing on is, most likely, decomposed granite — the crumbled remains of the volcanic and intrusive igneous rock that forms the territory’s geological skeleton. Geologists call it CDG: completely decomposed granite. Gardeners call it a problem.
Hong Kong’s soil is thin, acidic, and hungry. The average topsoil depth on undeveloped slopes is fifteen to thirty centimetres — compare this to the metre-deep loams of the English home counties or the rich mollisols of the American Midwest, and you begin to understand why a Hong Kong garden is, from the start, a project of creation rather than maintenance. The parent material — predominantly coarse-grained granite and fine-grained volcanic tuff, depending on which side of the territory you’re on — weathers into a sandy, free-draining substrate that holds water about as well as a sieve holds soup.
Then there is the pH. Hong Kong soils typically register between 4.5 and 5.5 — strongly acidic, thanks to millennia of leaching by monsoon rainfall. In practical terms, this means that phosphorus, calcium, and magnesium — three of the essential macronutrients plants require — are chemically locked into forms that roots cannot access. Your camellias, azaleas, and gardenias will be perfectly happy. Your vegetable beds will not.
Add to this the legacy of construction. If you garden in an urban or semi-urban plot — a Pok Fu Lam townhouse, a Clear Water Bay villa, a Sai Kung village house — the “soil” you are working with is almost certainly not soil at all. It is fill: a compacted mixture of subsoil excavated from somewhere else, construction debris, and whatever topsoil the contractor could source cheaply. I have dug into Hong Kong gardens and found, within the first thirty centimetres: broken tiles, rusted rebar, plastic packaging with expiry dates from the 1990s, and, on one memorable occasion, an intact Nokia 3310. The phone still had battery charge.
This fill layer — typically thirty to sixty centimetres deep, sitting on the original compacted substrate — is the real growing medium for most Hong Kong gardens. It is usually alkaline (pH 7.5–8.5, thanks to the lime in concrete and mortar), heavily compacted (bulk density often exceeding 1.6 g/cm³, at which point root penetration becomes physically difficult), and biologically dead. The earthworms, nematodes, springtails, mites, and the billion-strong bacterial and fungal communities that constitute living soil are simply absent. What you have is dirt. Dirt is what you track into the house. Soil is what grows things. Most Hong Kong gardens start with dirt.
This is not a counsel of despair. It is a statement of fact, and facts are where good gardening begins. The gardeners I admire most — the ones whose gardens look effortless, whose roses bloom without black spot, whose hedges are dense and green without weekly intervention — all share one characteristic: they paid attention to what was under their feet before they paid attention to what was above it. They understood the soil first.
The Soil Test That Changes Everything
There is a particular kind of gardener who will spend three thousand dollars on a specimen tree, plant it in unimproved ground, and then wonder why it is dead within six months. This gardener owns a moisture meter but has never used it. This gardener has a pH test kit still in its shrink-wrap, purchased two years ago and buried under seed packets in the potting shed. I have been this gardener. The tree was a Magnolia grandiflora ‘Little Gem’. It cost $2,800. It lasted four months.
The soil test is the single most cost-effective intervention you can make in a garden, and it is also the one most consistently skipped. The reasons are understandable: it is not glamorous, the results are not immediately gratifying in the way that a newly planted border is, and there is a lingering suspicion that soil testing is something that farmers do, not gardeners. This is wrong. A basic soil test — pH, texture, and organic matter content — costs less than a bag of decent compost and takes thirty minutes. It will tell you more about why your plants are struggling than any amount of Googling symptoms.
Start with pH, because pH governs everything. The availability of every plant nutrient is regulated by soil acidity: in acidic soils (below pH 5.5), phosphorus binds with iron and aluminium into insoluble compounds; in alkaline soils (above pH 7.5), iron, manganese, zinc, and copper become unavailable. You can apply all the fertiliser you want, but if the pH is wrong, those nutrients might as well not be there. A plant growing in pH 8.0 soil can sit in a sea of iron and still show chlorosis — yellowing leaves with green veins, the classic sign of iron deficiency — because the iron is chemically inaccessible.
In Hong Kong, the pH question is complicated by the fill problem. Native soils are acidic; fill soils are alkaline. Your garden may contain both, sometimes in the same bed, sometimes in layers — a sheet of alkaline fill dumped over original acidic subsoil. A single pH test from the surface tells you almost nothing. You need to test at multiple depths: scrape away the top two centimetres (mulch, leaf litter, recent compost), then sample from 5–10 cm, 15–20 cm, and, if you are serious, 30–40 cm. What you will often find is a gradient: slightly acid at the surface (where years of organic matter have accumulated), strongly alkaline at 20 cm (the fill layer), and then back to acidic at 40 cm (the original substrate). Plant roots do not stay in one stratum. They traverse all three.
The texture test is equally revealing and costs nothing. Take a handful of moist soil, squeeze it, and try to roll it into a ribbon between your thumb and forefinger. If it won’t form a ribbon at all and feels gritty, you have sand — excellent drainage, terrible nutrient and water retention. If it forms a long, smooth ribbon that holds together, you have clay — excellent nutrient retention, terrible drainage, and a tendency to crack like a dried riverbed in summer. If it forms a short, crumbly ribbon, you have loam — the gardener’s lottery win. Most Hong Kong soils are at the sandy end of this spectrum, which is both a blessing (roots rarely drown) and a curse (you will water more than you think).
The organic matter test is the simplest of all: look at the colour. Dark brown to black soils contain 3–5% organic matter; pale brown to grey soils contain less than 1%. Most Hong Kong soils are on the pale end. This matters because organic matter is not just “compost” — it is the substrate on which the entire soil food web depends. Without it, you are not gardening. You are hydroponics with extra steps.
The pH Trap: Acidity, Alkalinity, and the Nutrients That Disappear
The relationship between pH and nutrient availability is not linear and it is not intuitive. It is a series of bell curves, each nutrient with its own optimum range, and these curves overlap in a sweet spot between pH 6.0 and 7.0 that soil scientists call “slightly acid to neutral.” This is where almost everything works. Outside this range, things start to break.
At pH 4.5 — common in Hong Kong’s undisturbed hillside soils — phosphorus is the first casualty. It binds with iron and aluminium to form insoluble phosphates, and no amount of bone meal or superphosphate will change this until the pH rises above 5.5. Molybdenum, a micronutrient essential for nitrogen fixation in legumes, also becomes unavailable at low pH. Calcium and magnesium, both critical for cell wall structure and chlorophyll production, leach away in acid conditions faster than roots can absorb them. Meanwhile, aluminium and manganese — which are toxic to most plants at high concentrations — become soluble and available. The plant is simultaneously starved of what it needs and poisoned by what it doesn’t.
At pH 8.0 — typical of Hong Kong’s construction fill — the problem inverts. Iron, manganese, zinc, copper, and boron all become unavailable, locked into insoluble hydroxides and carbonates. This is why plants in alkaline Hong Kong gardens so often show interveinal chlorosis: the classic yellow-leaf-with-green-veins pattern of iron deficiency. The irony — and I use the word deliberately — is that the soil may contain abundant iron. It is simply in a form the plant cannot use. Applying chelated iron as a foliar spray will green up the leaves temporarily, but it treats the symptom, not the cause. The cause is pH.
For Hong Kong gardeners, the practical question is what to do about it. The answer, in most cases, is: adjust slowly, and only if you must. Attempting to raise the pH of acidic soil by dumping agricultural lime is a common impulse and a common mistake. Lime works, but it works slowly — months, not days — and over-liming creates a new set of problems. The target should be pH 6.0–6.8 for most ornamental gardens, achieved gradually through the addition of organic matter (which buffers pH toward neutral) and, if necessary, modest amounts of dolomitic limestone (which adds both calcium and magnesium). For alkaline fill soils, the standard advice is to add elemental sulphur, which soil bacteria oxidise into sulphuric acid, gradually lowering pH. In practice, on the scale of a domestic garden, this is difficult to do effectively. The better strategy is to work with the alkalinity rather than against it: choose plants that tolerate high pH (rosemary, lavender, clematis, dianthus, many Mediterranean herbs) and amend planting holes generously with ericaceous compost for the acid-lovers (camellias, rhododendrons, blueberries) that you cannot live without.
The gardener who understands pH does not fight it. She reads it, respects it, and plants accordingly.
Organic Matter: The One Amendment That Always Works
If you do exactly one thing to improve your soil — one thing, in a lifetime of gardening — make it this: add organic matter. Not fertiliser. Not soil conditioners in plastic bottles. Not the mysterious blue granules that old men at the flower market swear by. Organic matter. Compost, well-rotted manure, leaf mould, green manure crops dug into the soil. The stuff that was once alive and is now decaying, because decay is the engine of fertility.
The case for organic matter is so well-established that stating it feels almost redundant. It improves soil structure: in sandy soils, it acts as a sponge, holding water and nutrients that would otherwise drain away; in clay soils, it opens up pore spaces, improving aeration and drainage. It feeds the soil food web — the bacteria, fungi, protozoa, nematodes, arthropods, and earthworms that mineralise nutrients, suppress pathogens, and create the crumb structure that good tilth depends on. It buffers pH, slowly releasing organic acids as it decomposes. It contains, in modest but balanced amounts, almost every nutrient a plant needs. It is, in short, the closest thing to a universal soil improver that exists.
In Hong Kong, the organic matter deficit is severe. The combination of rapid decomposition (heat and humidity accelerate microbial activity), heavy rainfall (which leaches soluble organic compounds), and a gardening culture that has historically favoured chemical fertilisers over compost means that most soils contain far less organic matter than they should. A target of 5% organic matter by volume is ambitious but achievable; most Hong Kong garden soils I have tested are below 2%, and many are below 1%. At 1%, you are not gardening in soil. You are gardening in mineral dust.
The practical question is how to add it. The standard recommendation — a 5–8 cm layer of compost applied as mulch twice a year, in spring and autumn — works, but it works slowly. In Hong Kong’s climate, a 5 cm mulch layer can decompose to near-invisibility within three months. This is not a failure; it is the system working. The organic matter is being incorporated into the soil, exactly as intended. You simply need to apply it more frequently than gardeners in cooler climates do. Three times a year — March, July, November — is not excessive.
What kind of organic matter? The best is what you make yourself. A compost heap is not difficult to maintain. It requires: a balance of green material (kitchen scraps, fresh grass clippings, spent annuals) and brown material (dried leaves, shredded newspaper, cardboard, woody prunings); moisture (as damp as a wrung-out sponge); and air (turned every two to four weeks). The ratio is roughly two parts brown to one part green by volume. In Hong Kong’s warmth, a well-managed heap will produce usable compost in eight to twelve weeks — twice as fast as in temperate climates. The limiting factor is usually brown material; Hong Kong gardens produce far more green waste than brown. Stockpile autumn leaves. Hoard cardboard boxes. A bale of straw from a New Territories farm supply shop costs sixty dollars and will keep a compost heap balanced for months.
If making your own is not feasible, commercial compost is widely available in Hong Kong, though quality varies enormously. Look for compost that is dark brown, crumbly, and smells like a forest floor, not like ammonia or rot. The brand matters less than the feel: if it is still hot when you open the bag, it is not finished and will rob nitrogen from your soil as it continues to decompose. If it contains large recognisable chunks of wood, it needs another six months. The best compost in Hong Kong comes from Kadoorie Farm and Botanic Garden — their soil improver is made from on-site green waste and is consistently excellent, though it sells out quickly. Green Earth’s organic compost, available at most garden centres, is a reliable commercial alternative.
One final note on the subject of manure: it must be well-rotted. Fresh manure is too high in ammonia and will burn roots. Six months minimum, preferably a year. Horse manure is widely available from the riding schools in Beas River and Pok Fu Lam; it tends to contain weed seeds, so hot-compost it before use. Chicken manure is the most nutrient-dense but also the most likely to burn; use it sparingly and only after thorough composting. Cow manure is the gentlest and safest if you can find it.
The Mycorrhizal Underground: The Network Nobody Talks About
Beneath the soil surface, in every healthy garden, there exists a secondary root system that does not belong to your plants. It belongs to fungi. Specifically, to arbuscular mycorrhizal fungi — AMF — which form symbiotic associations with the roots of approximately eighty percent of terrestrial plant species. The fungi extend thread-like hyphae far beyond the reach of even the finest root hairs, mining phosphorus, nitrogen, zinc, and copper from soil pores that roots cannot access. In exchange, the plant provides the fungi with carbohydrates — sugars produced through photosynthesis. It is a trade network, ancient and elegant, that has been operating for at least 450 million years. You cannot buy it in a bottle, though many products claim otherwise.
Mycorrhizal networks do several things that synthetic fertilisers cannot. They improve drought tolerance: a plant connected to a fungal network can access water from a larger volume of soil. They suppress pathogens: the physical barrier of fungal hyphae around root tips, combined with the competitive exclusion of harmful organisms, reduces the incidence of soil-borne diseases. They improve soil structure: a protein called glomalin, secreted by AMF, binds soil particles into stable aggregates, creating the crumb structure that gardeners recognise as “good tilth.” And they enable something that sounds like science fiction but is well-documented in the literature: the transfer of nutrients and signalling compounds between plants of different species through a common mycorrhizal network. A nitrogen-fixing tree can share nitrogen with a neighbouring shrub. A plant under insect attack can send chemical warning signals through the fungal network to plants several metres away, which then upregulate their own chemical defences before the insects reach them. The ecologist Suzanne Simard, whose work on “mother trees” in Canadian forests has reshaped our understanding of plant communities, demonstrated this in meticulously controlled experiments. The garden is not a collection of individuals. It is a community, and the fungi are its communications infrastructure.
In Hong Kong gardens, mycorrhizal networks are often severely depleted or entirely absent. The combination of construction fill, soil compaction, chemical fertiliser use, and fungicide application can reduce AMF populations to near zero. And here is the trap: the very things gardeners do to compensate for poor soil often make the soil worse in the long run. Heavy applications of phosphorus fertiliser suppress mycorrhizal colonisation — the plant, finding abundant phosphorus freely available, reduces its carbohydrate investment in the fungal partnership. The fungi starve. The network collapses. The gardener, noticing that the plants look worse after the fertiliser wears off, applies more fertiliser. The cycle accelerates.
Rebuilding a mycorrhizal network takes time and restraint. Step one: stop sterilising your soil. Put down the fungicide. Step two: stop over-fertilising, especially with phosphorus. Use slow-release organic fertilisers — blood and bone, seaweed extract, fish emulsion — rather than synthetic quick-release formulations. Step three: add organic matter, which feeds the saprophytic fungi that form the base of the soil food web. Step four: consider inoculation, though with realistic expectations. Commercial mycorrhizal inoculants — powders and granules containing AMF spores — are available in Hong Kong (Rootgrow and MycoApply are the most reliable brands), but their effectiveness depends on soil conditions. They work best when applied directly to roots at planting time, in soils that are not already high in phosphorus. Broadcasting inoculant over an established bed is unlikely to achieve much; the spores need root contact to germinate.
Step five, and perhaps the most counter-intuitive: disturb the soil as little as possible. Every time you dig, you sever fungal hyphae. The network rebuilds, but each rebuilding cycle sets it back. No-dig gardening, in which organic matter is applied as a surface mulch and allowed to be incorporated by soil organisms rather than by the spade, preserves mycorrhizal networks and, over several years, produces soil structure that no amount of digging can match. The no-dig method was developed and popularised by Charles Dowding in the UK, but it translates well to Hong Kong conditions, particularly in raised beds and perennial borders. The principle is simple: you feed the soil, and the soil feeds the plants.
Practical Soil Improvement for Hong Kong Gardens
Here, then, is a programme. It is not complicated. It does not require expensive products or specialist knowledge. It does, however, require patience — the kind of patience that is in short supply in a city where results are expected immediately and dissatisfaction sets in after the first week. Soil improvement operates on a timescale of seasons and years, not days. The gardener who understands this will succeed. The gardener who does not will continue buying replacement plants.
Year One: Stop doing harm. Cease all applications of synthetic quick-release fertilisers with NPK ratios higher than 5-5-5. Cease all fungicide applications unless you have a confirmed pathogen diagnosis (powdery mildew and black spot are managed by cultural practices, not chemicals). Cease all soil sterilisation — no Jeyes Fluid, no bleach, no “soil cleansers.” Test your soil: pH at three depths, texture by feel, organic matter by colour. Record the results. You cannot improve what you do not measure.
Apply a 5 cm layer of good-quality compost as mulch across all planted areas. This is your baseline. It will suppress weeds, conserve moisture, moderate soil temperature, and begin the slow process of adding organic matter. Do this in early spring (March) and again after the summer rains begin to ease (September). If you can manage a third application in late autumn (November), so much the better.
Year Two: Begin active amendment. Based on your soil test results: if pH is below 5.5, apply dolomitic limestone at the rate recommended on the bag — typically 100–200 grams per square metre, worked lightly into the top 5 cm. Do not exceed this rate; over-liming is harder to correct than under-liming. If pH is above 7.5, work with it rather than against it. Add ericaceous compost to planting holes for acid-loving species, but accept that most ornamentals will manage at pH 7.5 if organic matter levels are adequate.
Inoculate new plantings with mycorrhizal fungi. Sprinkle a teaspoon of inoculant powder directly onto the roots at planting time. Water in with a dilute seaweed solution — the trace elements and growth hormones in seaweed extract are thought to stimulate mycorrhizal colonisation, though the evidence is more anecdotal than experimental. It certainly does no harm, and seaweed extract is an excellent mild fertiliser in its own right.
Year Three: Transition to maintenance. By now, your soil organic matter should be visibly increasing — darker colour, better crumb structure, more earthworms. Reduce compost applications to twice-yearly rather than three times. Begin using green manures in any vegetable or annual beds that would otherwise lie bare between plantings: buckwheat (Fagopyrum esculentum) is fast-growing and tolerates poor soil; crimson clover (Trifolium incarnatum) fixes nitrogen and flowers attractively; phacelia (Phacelia tanacetifolia) produces masses of purple flowers that pollinators adore. Sow, let grow for six to eight weeks, then cut down and dig in — or, better, leave the tops as a mulch and let the roots decompose in situ. Each green manure cycle adds the equivalent of a modest compost application, with the added benefit of living roots feeding the mycorrhizal network throughout.
Throughout all three years, maintain a mulch layer. Mulch is not a one-time application; it is a permanent feature of a well-managed garden, like watering or weeding. The mulch layer should never be less than 3 cm deep. In Hong Kong’s climate, organic mulches — compost, shredded leaves, straw, cocoa shell — decompose rapidly, which is exactly what you want. The decomposition feeds the soil. Replenish as needed.
When to Stop Amending and Start Accepting
There comes a point, in every soil improvement programme, when further amendment delivers diminishing returns. Recognising this point is a gardening skill as important as knowing when to water. The goal is not to turn Hong Kong soil into something it is not — not to replicate the deep, moisture-retentive loams of English cottage gardens, because you cannot, and you should not try. The goal is to make the soil good enough for the plants that belong here.
This is not defeatism. It is the gardener’s version of the serenity prayer: the wisdom to know the difference between what can be changed and what must be accepted. You can raise organic matter from 1% to 5%. You cannot turn decomposed granite into alluvial loam. You can buffer pH toward neutral with organic matter and modest liming. You cannot permanently alter the pH of a soil that sits on a bedrock of acid volcanic tuff. You can build a thriving mycorrhizal network over three to five years of careful management. You cannot maintain that network if you continue to apply high-phosphorus fertilisers and dig deeply twice a year.
The most successful Hong Kong gardens I know are not the ones with the most amended soil. They are the ones where the gardener looked at what they had, accepted the constraints, and chose plants that wanted to be there. A south-facing slope of thin, acid, free-draining soil is not a problem to be fixed; it is a Mediterranean garden waiting to happen. Plant it with lavender (Lavandula angustifolia ‘Hidcote’), rosemary (Salvia rosmarinus), santolina (Santolina chamaecyparissus), cistus (Cistus × purpureus), and euphorbia (Euphorbia characias subsp. wulfenii). Add a specimen olive (Olea europaea) in a large terracotta pot. Underplant with thyme (Thymus vulgaris) and oregano (Origanum vulgare). Water deeply once a week in summer, not at all in winter. This garden will look better, require less work, and live longer than any attempt to grow hydrangeas and hostas on the same site.
A shaded courtyard with compacted alkaline fill is not a disaster. It is a fern garden. Plant it with Nephrolepis cordifolia, Asplenium antiquum, and the indestructible Aspidistra elatior, which the Victorians called the cast-iron plant for good reason. Add Fatsia japonica for structure, Sarcococca confusa for winter scent, and a climbing Hydrangea integrifolia — the evergreen climbing hydrangea from Taiwan, which tolerates alkaline soil far better than its more famous cousin H. anomala subsp. petiolaris — on the north-facing wall. Mulch with leaf mould. These plants did not evolve in rich woodland loam. They evolved on thin forest soils and rocky slopes across East Asia. They are at home here.
A balcony container garden in Mid-Levels, exposed to wind and salt spray, with whatever potting mix the previous tenant left in the pots: this, too, is manageable. Refresh the potting mix with fifty percent new material (a good peat-free multipurpose compost, not the cheapest bag at the florist). Choose wind-tolerant species: dwarf agapanthus (Agapanthus ‘Peter Pan’), Pelargonium sidoides — the species pelargonium with silver-grey foliage and dark maroon flowers that laughs at drought — and the architectural Beschorneria yuccoides, a Mexican succulent that produces six-foot flower spikes in its third year and asks for nothing in return but a pot that won’t blow over. These plants do not need perfect soil. They need soil that drains, a modest feed once a month during the growing season, and a gardener who understands that neglect can be a form of care.
The soil first philosophy is, ultimately, a philosophy of attention. Pay attention to what is under your feet. Understand it. Work with it. Improve it where you can, accept it where you must, and never, ever blame the plant for what the soil failed to provide. The plant is not the problem. The plant is doing its best. The question is whether you are doing yours.
Further Reading
For those who want to go deeper — and I hope some of you do, because soil science is one of those subjects that rewards obsession — the essential texts are few and clear. Sir Albert Howard’s An Agricultural Testament (1940) is the founding document of organic soil management, written by a British scientist who developed his ideas while working in India, and it has lost none of its force. Charles Dowding’s No Dig (2022) is the practical modern manual. Jeff Lowenfels and Wayne Lewis’s Teaming with Microbes (2010) is the best introduction to the soil food web for gardeners who last studied biology in secondary school. For the Hong Kong context, G.A.C. Herklots’s The Hong Kong Countryside (1951), though long out of print, contains a chapter on local soils that remains surprisingly current. And the Kadoorie Farm and Botanic Garden publishes an irregular series of technical notes on soil management in subtropical conditions; their library is open to visitors by appointment.
Start with any one of these. Start with a soil test. Start, quite literally, from the ground up.
