Prepping Beds Before the Rains: What the Evidence Actually Supports

Here, most of the year's rain falls between November and March, on beds that were built and watered by hand through a dry summer. This is what the research supports about getting a bed ready for that change, and where the measurements run out.

A Santa Clara County bed gets built in the dry months and then has to cope with the wet ones. In the 1991–2020 climate normals from the National Oceanic and Atmospheric Administration (NOAA), its 30-year station averages, November through March brings between 82 and 87% of the year's rain at the San Jose, Palo Alto, Gilroy and Los Gatos stations (NOAA National Centers for Environmental Information, 2021). It tends to come in a few big storms: across California, most of the year's precipitation falls on just 5 to 15 wet days (Dettinger et al., 2011).

Until the rain picks up in November, a fresh seedbed needs keeping damp by hand. Once it arrives, the job changes from keeping a bed damp to making sure it drains.

Go Easy on Turning It Over

A lot of bed-prep advice says to dig it in. Oregon State University Extension, for example, tells gardeners to spread a thin layer of compost on an existing bed each year and dig it into the soil 8 to 12 inches deep with a fork, spade or tiller (Miller & Mann, 2021).

The soil-structure research gives a reason to go easier than that, though it comes from farm fields, not garden beds. Panagea and colleagues analyzed seven long-term European arable experiments, running 8 to 54 years; three of them compared tillage intensities. Water-stable aggregates are the soil crumbs that hold together when they're soaked and sieved in water, and macroaggregates are the larger ones. The authors' conclusion: plowing that turns the soil over, which they call inversion tillage, reduced the mass of large water-stable macroaggregates in the top 6 inches (15 cm) and the organic carbon stored inside the aggregates. The more intense the tillage, the bigger the effect (Panagea et al., 2022).

There are two catches. The first is in the same concluding sentence: inversion tillage did not change the overall mass of macroaggregates, only the large ones. The second is scale. The conventional treatments were plowing on arable fields, and the samples went only 6 inches deep. A garden fork in a 4-foot bed is a different tool, so treat the direction as the useful part and the size of the effect as unknown.

Within those tillage experiments, the details vary by site. In the Czech trial, zero tillage had almost double the mean weight diameter of the aggregates (a measure of how big the stable crumbs are) compared with minimum or conventional tillage. In one Hungarian trial, shallow tillage had almost double the mean weight diameter of conventional plowing. In the UK trial, direct drilling (sowing straight into untilled soil) had significantly larger aggregates than conventional tillage (Panagea et al., 2022).

The same work has a finding about crop residues. Across the experiments, plowing in the previous crop's straw and stalks did not increase the mass of water-stable macroaggregates or the organic matter held inside them, and it had no consistent effect on carbon in the smaller fractions. At some sites, though, it did raise aggregate stability (Panagea et al., 2022). Those were field crop residues dug into arable soil, which is not the same as spreading finished compost on a bed.

Stay Off It Once It's Wet

Most gardeners know not to work wet soil. The research suggests heavy loads compact soil even before it gets that wet.

Field capacity is the fairly steady water content a soil settles at after about 48 hours of draining from saturation; soil at field capacity feels very moist in the hand (Shaxson & Barber, 2003). Bulk density is how tightly packed the soil is: the same volume weighing more means less pore space for air and water. On a sandy loam in Hebei, China, researchers drove an 11,600-pound (5.26-tonne) wheel loader, which they describe as heavier than normal field tractors in the region, over plots that were drier than field capacity. Bulk density in the 4 to 8 inch (10 to 20 cm) layer rose by about 9% after two passes and 13% after six, compared with plots that weren't driven on. The next year, when the same plots were driven over again, the increases were 14 to 16% (Nawaz et al., 2023).

A forest study in Italy compared dry and moist soil directly. On a loam soil in the Rincine forest northeast of Florence, five passes of a wheeled forestry tractor raised bulk density in the top 3 inches (8.5 cm) by 26% above plots that weren't driven on, on dry soil, and by 33% on moist soil. For the tracked tractor in the same study, soil moisture made no significant difference (Cambi et al., 2016). The authors concluded that even the worst case, the wheeled tractor on moist soil, did not push bulk density or penetration resistance (how hard it is to push a probe into the soil) to levels potentially harmful to the forest stand. They also say the measured values could limit plant growth to some degree.

Both of those are heavy machines. I searched for a study that measured what a person standing on a garden bed does to the soil and found nothing, so we know the direction but not the size.

The practical test comes from extension advice. Dig a trowel of soil and squeeze it: if it crumbles through your fingers, it's ready to work, and if it forms a muddy ball, give it a few more days to dry (Lerner, 2007). Our own habit, whatever the squeeze test says: lay a plank across the bed to stand on, and keep permanent paths so you rarely have to make the call at all.

Compost: What the Rate Trials Found

Compost advice for an existing bed varies a lot. Oregon State recommends a layer a quarter-inch to 1 inch deep each year (Miller & Mann, 2021). North Carolina's Department of Agriculture caps maintenance applications at 1 inch a year (McGinnis & Myers, 2018). Utah State University Extension calls about 1 inch a year a general rule of thumb for most Utah soils, and adds that more may be added depending on the compost and a soil test (Stock et al., 2019). University of Minnesota Extension typically recommends about an inch or less a year over garden beds (Hoidal & Wilcox, 2023).

The one California source here goes further. The University of California (UC) Marin Master Gardeners say to apply a half-inch to 3-inch layer of compost to garden beds at least once a year, and for vegetable beds to apply it again in fall if you're growing edibles over winter (UC Marin Master Gardeners, n.d.). Followed to the top of the range, that's up to 6 inches a year.

Those figures are guidance, but compost rates have also been tested in garden beds and vegetable plots, mostly by one group at the University of St. Thomas in Minnesota. In one trial, raised-bed plots got municipal or manure compost at 15, 35 or 50% of the bed by volume. Phosphorus washed out in water draining through the beds (leachate) and removed in harvested crops added up to only 1 to 10% of the phosphorus applied, which the authors read as phosphorus building up in the soil (Small et al., 2018). In another set of raised beds, crop yield did not differ among the compost treatments, but crops recovered a smaller share of the phosphorus at the higher rates. Beds where compost was matched to the crops' nitrogen needs leached about twice as much phosphate as the other treatments (Shrestha et al., 2020). Seven years into that study, the higher-input compost beds had recovered about 30% of the phosphorus added in harvested crops, against more than 88% in the lower-input beds (Small et al., 2024).

A Connecticut Agricultural Experiment Station trial worked at the low end. Leaf compost was tilled into field plots at 10, 25 and 50 tons per acre on a fine sandy loam, with the top rate described as 1 inch on the surface. In carrots, every compost rate out-yielded the unamended plots, and going from 10 to 50 tons per acre raised yield by as much as 12% (Maynard, 2005). None of these trials tested a 2- to 3-inch layer spread on the surface each fall, so they don't settle the right depth. The Minnesota trials do agree on where the extra goes: into the soil, as phosphorus.

What the phosphorus says

One shorthand first: N, P and K stand for nitrogen, phosphorus and potassium. Compost carries all three. The question is whether it carries them in the proportions a vegetable bed uses.

Many composts have relatively little nitrogen for the phosphorus they carry, so applying enough to cover a crop's nitrogen often overshoots on phosphorus. In the Minnesota raised-bed trial, matching compost to the crops' expected nitrogen demand meant applying 3 to 4 times as much as matching it to their phosphorus demand (Shrestha et al., 2020). Before that trial started, the garden soil tested almost three times higher in phosphorus than a nearby reference lawn, 75 parts per million (ppm) against 26 (Shrestha et al., 2020). A companion study found that one form of phosphorus, the weakly held inorganic kind, was significantly elevated in the native soil beneath research garden beds, below about 14 inches (35 cm), compared with reference soil (Small et al., 2019).

A survey of working gardens shows where that ends up. In October 2022, University of Minnesota Extension educators tested soil at 20 Ramsey County community gardens and urban farms, three samples per garden. All 20 had too much phosphorus. The lowest sample read 27 ppm, the median 133 and the highest 244, against the 25 ppm the educators call very high. Asked what else they had applied, almost none of the gardeners had used anything but compost (Hoidal & Wilcox, 2023).

Oregon State's guidance names the risk: if too much compost is added, nutrients including phosphorus and potassium can leach below the root zone and out of the garden (Miller & Mann, 2021).

So let the soil test set the rate. If phosphorus isn't excessive, Minnesota's educators say to go ahead and add compost, and they typically recommend about an inch or less a year (Hoidal & Wilcox, 2023). A bed that tests high needs less compost, or a break from compost for a few years (Hoidal & Wilcox, 2023).

A new bed is a different case. Oregon State suggests 3 to 4 inches of compost for a new vegetable bed against a quarter-inch to 1 inch a year for an existing one (Miller & Mann, 2021). North Carolina's guidelines never go above a 3-inch layer tilled to 6 inches at any one time, and set the rate for a new bed by the compost's chemistry (McGinnis & Myers, 2018). Building a bed allows more at the start, but test the soil and the compost first.

Raised Beds Are Mainly Drier

Raised beds are often recommended because they drain better and warm up faster. The drainage part holds up better than the warmth.

The closest test we found ran in Fort Albany First Nation in subarctic Ontario. Researchers filled ground-level beds, 10-inch (0.25 m) raised boxes and taller hügelkultur-style boxes (raised boxes with woody debris and leaf litter in the bottom) with the same soil and logged soil climate at 4 inches (10 cm) deep from late June to early October. Over that period, the standard raised boxes averaged no warmer than the ground: the mean difference was zero, give or take about 1 °F (0.0 ± 0.5 °C). The differences ran from about 1.4 °F cooler to 3.4 °F warmer (−0.8 to +1.9 °C) (Wilton et al., 2023). Early in the season both box types ran warmer than the ground, by about 1 to 4.5 °F (0.5 to 2.5 °C) in the authors' summary, and late in the season the ground ran very slightly warmer than either box. The taller hügelkultur boxes stayed about 2 °F (1.0 °C) warmer on the seasonal average.

That's one small study a long way from here: four plots per treatment, one season, no statistical tests on the soil-climate data, and landscape fabric under the boxes that the ground beds lacked. An older Ohio strawberry trial on a single soil found raised beds drier than flat ones. At 6 inches (15 cm) their soil temperatures swung more widely, and at 4 inches (10 cm) there was no difference (Goulart & Funt, 1986). Neither study shows a raised bed reliably running warmer through a season.

The moisture difference in the Ontario study was large and held through the whole logging period. At 4 inches (10 cm) deep, water made up 36% of the soil's volume in the ground beds against 21% in the standard raised boxes (0.36 against 0.21 m³/m³), and the daily values in the two never overlapped (Wilton et al., 2023). That's the reliable part: a raised bed is a drainage tool.

So its value flips with the season here. The same drainage that's a liability in a rainless summer, when a raised bed holds less water than the ground, is the point in a soggy January. At the San Jose airport station, May through October brings about 1.2 inches of rain in total in the 1991–2020 normals (NOAA National Centers for Environmental Information, 2021). If your beds hold water through winter, raising them is a reasonable response. If they don't, you'd be taking on extra watering through the dry half of the year to fix a drainage problem you may not have.

Whether a raised bed improves the soil under it is a separate question, and two studies point different ways. Saturated hydraulic conductivity, which this study estimated for the surface 4 inches (0 to 10 cm), is reported in millimeters per hour. At the research station of the International Crops Research Institute for the Semi-Arid Tropics near Hyderabad, India, the surface 4 inches (0 to 10 cm) of a deep clay soil under raised beds had an estimated saturated hydraulic conductivity of about 5.5 inches per hour (141 mm/h) against 2.2 inches per hour (57 mm/h) under flat beds (Garg et al., 2022). Those raised beds had been left undisturbed for thirty years, with only the furrows reopened each season. In a five-year trial in the Eastern Himalaya, where the bed layouts were rebuilt every year, bulk density did not differ significantly between flat beds, ridge-and-furrow and broad beds after five cropping cycles (Babu et al., 2020).

One set of beds was left alone for thirty years, the other rebuilt every season for five, and neither study was set up to separate the shape of the bed from how long it was left alone, so we can't say which one did the work. Our reading is that the permanent part of a permanent raised bed may matter more than the raised part, and we'd like to see someone test that.

Pots and Planters

A container isn't a small bed. A perched water table, a saturated layer at the bottom of the mix, forms in container soil even when the container drains freely (Spomer, 1980), so the bottom of a pot stays wetter than the rest no matter how good the drainage hole is. As Rowe summarizes the standard physics, the depth of that saturated layer depends on the mix, not the pot's size or shape (Rowe, 2025). Our substrate primer goes through what roots need from a mix, so we won't repeat it here.

Three things matter for a container going into a wet winter.

Bed soil usually doesn't work in a pot. Spomer warns that an excellent garden or field soil put in a container will probably stay saturated after watering and drainage, which leads to poor aeration and poor growth (Spomer, 1980). If you're prepping beds and pots the same weekend, it's tempting to fill the planters from the same pile. Don't.

Size does more than you'd think. The best number we have comes from research pots, not garden planters. Across a meta-analysis of 65 studies, plant mass rose on average 43% for every doubling of pot size, and herbaceous and woody plants responded about the same (Poorter et al., 2012). That's total plant mass in experiments, not vegetable yield, and it's about volume, not depth. Our advice from it is simple: when in doubt, go one container size up.

We skip the gravel layer, by choice. Some researchers have warned that a gravel layer holds water up in the mix above it. When a 2025 study tested drainage layers in pots, a layer usually lowered the water held in two peat-free mixes and mostly made no difference in a loam-based one. The author calls drainage layers "a potentially useful tool" (Rowe, 2025). Our Myths Demystified entry on drainage rocks has the full account. We still don't add one: we'd rather give that space to the mix.

One more thing for the cold months. An Oregon State horticulture article for nursery growers on overwintering container plants points out that their roots are typically poorly buffered against temperature swings because they're exposed and surrounded by only a small amount of mix. Low temperatures make roots hardier, warm ones reverse it at any point in winter, and roots can lose all their hardiness within 24 hours (Green & Fuchigami, 1985). We didn't find measurements of root-zone temperatures in containers through a winter as mild as ours, so we can't put a number on what a warm January spell costs a potted plant here.

Two Things We're Not Settling Here

Mulch. Wood chips and nitrogen deserve more than a paragraph, so we're giving them a full entry.

Cover crops. These get their own treatment too, but one caution belongs here because it's easy to get backwards in a dry-summer climate. In a synthesis of Californian and Mediterranean arable farmland, 23 experiments found that plots with cover crops held 13% less soil water in spring, before the food crop was planted, than plots without (Shackelford et al., 2019). A meta-analysis of dryland trials found the same direction, with soil water 18% lower on average after cover crops (Garba et al., 2022). Over the winter itself, a three-year study at ten California tomato and almond field sites found that soil moisture generally didn't differ significantly between fields with and without a cover crop (DeVincentis et al., 2022). If you're sowing a cover crop partly to save moisture for spring, the evidence doesn't back that.

What to Do Before the Rains

Everything above, in one place.

• Go easy on turning the bed over. Cut the spent crop at the base, leave the roots, add on the surface. Fork to crack compacted ground without turning it.
• Get a soil test before you buy compost. The phosphorus number should set your rate. Ask the lab which extraction method it uses.
• If phosphorus comes back high, use less compost, or take a break from it for a few years.
• A new bed can take more at the start. Test the soil and the compost first, and go easier once it's established.
• Stay off it once the rain starts. Squeeze test, a plank, or permanent paths.
• Raise the bed if it holds water in winter, not because you expect it to warm up. Expect to water it more from May through October.
• Don't fill pots with bed soil, go one container size up, and leave out the gravel layer.