Wicking Beds in Tucson: Water Savings, Failure Modes, and When Other Irrigation Methods Are Better

2. Abstract

Wicking beds store irrigation water below the root zone and move it upward through a porous growing medium. This arrangement can reduce runoff, deep drainage, and some surface evaporation while extending the interval between refills. It does not eliminate plant transpiration, prevent all evaporation, or guarantee lower total water use. The strongest controlled evidence comes from a small series of tomato, lettuce, and radish experiments in Adelaide, South Australia. In one tomato experiment, shallow wicking beds used 7.1% to 9.7% less irrigation water than carefully surface-watered controls in two configurations, while a third used 6.1% more. Yield was 59.6% to 72.8% higher in those shallow beds. A deeper wicking treatment used 22.1% less water but yielded 11.2% less, so its water-use efficiency did not improve. A separate observational study of South Australian home gardens found no statistically significant water-use or water-use-efficiency advantage for wicking beds, although gardeners refilled them much less often (Pollard et al., 2020; Semananda et al., 2016).

For Tucson, the most defensible conclusion is conditional. A wicking bed can be useful when a gardener already needs a raised, contained root zone and values several days or weeks of stored water. It is most promising for dense plantings of annual vegetables and flowers that prefer steady moisture. Its advantage is smaller when the alternative is well-managed drip irrigation in suitable soil. Tucson also creates specific risks: very high summer demand can exceed upward flow through the growing medium; warm, continuously wet root zones can lose oxygen; dissolved salts can accumulate near the surface; and monsoon inflow can saturate the bed if overflow fails. No controlled field trial located for this review measured wicking-bed water use, reservoir temperature, salinity, or reliability in Tucson. Local use should therefore be treated as a monitored irrigation choice, not a proven universal conservation measure.

3. Key findings for Tucson gardeners

• Water savings depend on the comparison. Wicking beds are likely to use less water than a conventional raised bed that is watered heavily from the surface and allowed to drain. Evidence does not show that they consistently beat carefully scheduled drip irrigation or precise surface irrigation.

• Longer refill intervals are the clearest benefit. Controlled tomato work required fewer than 26 reservoir additions, compared with 40 to 50 surface-watering events. An observational home-garden study also found much longer intervals between watering in wicking beds, even though total measured water use was not significantly lower (Pollard et al., 2020; Semananda et al., 2016).

• Capillary supply has a rate and height limit. Water does not rise equally well through every medium or depth. Fine pores can lift water farther but may transmit it slowly and hold too little air. Coarse pores admit more air but have less capillary reach. The plant, medium, bed depth, compaction, and evaporative demand must work as one system.

• Tucson summer heat remains a plant limit. A reservoir can dampen rapid temperature swings, but it cannot compensate for unsuitable planting season, excessive leaf temperature, or intense afternoon sun. The single controlled temperature observation located for this review came from an Adelaide glasshouse, not an exposed Tucson bed.

• Salt is a management cost. Evaporation and transpiration remove water but leave most dissolved salts behind. Bottom watering tends to move salts toward the upper soil layer. Periodic leaching may be necessary, but no research supports a universal flushing schedule for Tucson wicking beds. Any water used for flushing belongs in the system's total water budget.

• Root rot is not an automatic consequence of wicking. A functioning bed should have a saturated reservoir below an unsaturated, aerated root zone. Trouble begins when the overflow is too high or blocked, the medium stays saturated, oxygen diffusion becomes inadequate, or susceptible roots enter persistently warm and wet layers.

• The best-supported crops are limited. Tomatoes, lettuce, and radishes have direct controlled evidence. Chard, beans, and sunflowers performing well in one Tucson garden is useful local observation, but it does not establish general yield or water-saving effects.

• Drip irrigation is the strongest general alternative. It can serve crops with different moisture needs, works in both ground and raised beds, allows intentional leaching, and has mature, replaceable components. It still needs pressure control, filtration, inspection, flushing, and correct scheduling.

• Wicking beds solve access and storage problems better than they solve every irrigation problem. Their strongest case is a gardener who already wants a contained raised bed, grows moisture-responsive annuals, and will monitor salts, overflow, and refill behavior.

4. Research question and scope

The research question is: How can wicking beds reduce water use in Tucson gardens, and under what conditions might they perform poorly or create new problems?

This paper evaluates lined, reservoir-based wicking beds used for vegetables and flowers. It does not evaluate a particular commercial product, prescribe dimensions, or provide construction instructions. The comparison methods are conventional in-ground beds, freely draining raised beds watered from above, drip irrigation, and buried unglazed clay pots commonly called ollas.

“Water use” is separated into four ideas that are often blended together:

1. Applied irrigation: water added by the gardener.
2. Total water input: irrigation plus effective rainfall.
3. Crop water use: water evaporated from the soil and transpired by plants, often combined as evapotranspiration.
4. Water-use efficiency: a useful output, such as edible yield, divided by water input or crop water use. Studies do not always use the same denominator.

The paper uses three evidence labels. Measured evidence means a controlled experiment or monitored garden dataset. Indirect evidence means established soil, irrigation, or plant physiology applied to wicking beds without a direct Tucson trial. Practitioner observation means an experience that may identify a useful hypothesis but cannot isolate cause and effect. Numerical results are not transferred to Tucson unless the original crop, climate, medium, depth, watering rule, and study duration are sufficiently comparable. None of the published experiments found for this review meet that full standard.

5. Expert panel and selection rationale

The panel is a research framework based on the public work of three people. It is not a claim that they reviewed or endorsed this paper.

Niranjani P. K. Semananda, PhD

Semananda was selected because she is the principal investigator behind the most sustained peer-reviewed research program located on wicking beds. Her doctorate in water engineering at the University of South Australia examined wicking-bed irrigation, and she led controlled studies of tomatoes, lettuce, radishes, recycled water, water-use efficiency, soil temperature, and salinity. She is currently listed by the Institute of Technology, University of Moratuwa, with research interests in irrigation, agricultural water management, and urban agriculture (Institute of Technology, University of Moratuwa, n.d.; Semananda et al., 2016, 2018, 2020, 2021).

Baden R. Myers, PhD

Myers was selected as the second direct wicking-bed researcher. The University of Adelaide identifies his work in water resources, irrigation, and sustainable urban water management. He co-supervised Semananda's doctoral work and coauthored the controlled wicking-bed experiments and the review of capillary irrigation. His contribution is particularly relevant to water balance, comparative irrigation performance, and the limits of transferring a laboratory or glasshouse result into household practice (The University of Adelaide, n.d.; Semananda et al., 2016, 2018, 2020, 2021).

Ursula K. Schuch, PhD

Schuch was selected for Tucson and arid-horticulture context, not for direct wicking-bed research. The University of Arizona identifies her as a professor and extension specialist in environmental horticulture whose work includes drought and salinity tolerance, irrigation requirements, and specialty crops. She has also coauthored Arizona Extension guidance on drip irrigation. Her documented subject area supplies the local horticultural test that the Adelaide wicking-bed studies cannot provide (Schuch et al., 2016; University of Arizona, n.d.).

6. Research method and source-selection criteria

The search prioritized peer-reviewed experiments that measured irrigation input, yield, water-use efficiency, soil moisture, temperature, or salt behavior in wicking beds or closely related subirrigation systems. It then used university extension and government sources for Tucson climate, water quality, plant response, drip operation, and mosquito control. Older work was retained where it remains foundational, especially soil-water physics and buried clay-pot irrigation.

The central evidence base is small. Four peer-reviewed papers by Semananda and colleagues form most of the direct wicking-bed record. They include one tomato efficiency experiment, one shallow-rooted-crop experiment, one recycled-water and salinity experiment, and one review of capillary irrigation. A South Australian citizen-science study supplies a useful reality check from 34 home gardens and 93 garden areas. Controlled greenhouse subirrigation research supplies indirect evidence about salt and fertilizer concentration but is not treated as equivalent to a soil-filled outdoor wicking bed.

Sources were excluded when they made large savings claims without a measured baseline, described a construction method without performance data, or repeated another author's number without enough detail to reconstruct the comparison. Vendor pages and unsourced practitioner claims were not used as evidence of water savings. Search results were current through August 8, 2026.

The review has four important limitations. First, no Tucson field trial was found. Second, most direct experiments came from one research group and used small samples. Third, study definitions of water-use efficiency differ. Fourth, long-term liner life, mineral clogging, root intrusion, and maintenance are discussed widely in practice but have little controlled evidence. These limitations lower confidence in precise performance predictions but do not erase the physical mechanisms or the measured results.

7. Tucson climate and water context

Tucson International Airport's 1991–2020 climate normals show an annual average high of 84.0°F, an annual average low of 57.3°F, and 10.61 inches of precipitation. June's normal high is 101.2°F and its normal rainfall is 0.23 inch. July and August average highs near 100°F and receive 2.21 and 1.98 inches of rain, respectively. July through September together supply 5.51 inches, about 52% of normal annual rainfall. The airport averages 68 days each year at or above 100°F (National Weather Service, n.d.). A household site may be hotter or cooler because of elevation, walls, pavement, reflected sunlight, and nighttime heat storage.

These conditions create two different irrigation seasons. Late spring and early summer combine intense solar radiation, low humidity, wind, and little rain. Plant demand can rise faster than a small reservoir or capillary pathway can supply it. During the summer monsoon, atmospheric humidity rises and episodic storms can add more water in minutes than a bed can store. Autumn and winter lower evaporative demand, but shallow containers and exposed plumbing still experience larger temperature swings than deep in-ground soil. The University of Arizona's Pima County guidance notes that May is commonly the hottest and driest period and that vegetable and flower gardens may need daily watering under conventional management (University of Arizona Cooperative Extension, n.d.-a).

Tucson water is not uniform. Tucson Water blends and distributes sources differently across its service area, so pH, hardness, total dissolved solids, sodium, and electrical conductivity can vary by location and over time. The utility describes local water as hard to very hard in many areas and provides address-specific water-quality information (Tucson Water, 2025a, 2025b). Hardness, alkalinity, and salinity are related only indirectly. Hardness mainly describes calcium and magnesium. Alkalinity describes acid-neutralizing capacity. Salinity describes the total concentration of dissolved ions and is more directly estimated with electrical conductivity. A high pH by itself does not establish a plant salt hazard.

The practical implication is that a Tucson wicking bed should not be evaluated with a generic statement that the city has “alkaline water.” The relevant inputs are the actual water source, its electrical conductivity and sodium content, the fertilizer added, the crop's tolerance, and whether accumulated salts can leave the root zone. Rain can dilute salts, but a lined bed only exports them when water leaves through intentional leaching or overflow.

8. How wicking beds function

Storage and upward movement

A wicking bed separates water storage from most of the root zone. The lower reservoir or saturated layer holds water. Above it, unsaturated pores in the growing medium exert matric suction, meaning water adheres to particle surfaces and curved air-water interfaces. Water moves upward in response to a gradient created by drier soil, surface evaporation, and root uptake. Roots remove water locally; capillary flow partly replaces it.

This process is sometimes described as if the bed gives plants exactly what they request. It does not. The movement rate must be fast enough to match root extraction and evaporation. If demand is greater than the upward hydraulic flow, the upper root zone dries even while water remains below. If the medium transmits water too readily and the overflow maintains a high water table, too much of the root zone may remain saturated.

Capillary limits

Pore size creates a tradeoff. Smaller pores can support a taller water column, but water moves through them more slowly and they may leave less air space. Larger pores transmit water easily when saturated but cannot lift it as high. Particle-size distribution, organic matter, aggregation, compaction, hydrophobic drying, layer boundaries, and physical contact with the saturated zone all change performance (Hillel, 1998; Semananda et al., 2018).

There is no universal maximum soil depth for every wicking bed. In Semananda et al. (2016), tomato treatments with about 300 mm of soil performed differently from a treatment with about 600 mm in the same sandy medium. The deeper treatment used less applied water but also yielded less than its surface-watered comparison. That finding is evidence for a practical limit in that treatment, not a rule that every medium stops wicking at 300 mm.

Evaporation and drainage

Subirrigation can reduce three nonproductive losses: runoff from water applied faster than infiltration, deep drainage below the useful root zone, and evaporation caused by repeatedly wetting the top surface. A liner converts deep drainage into storage until the reservoir is full. Bottom watering also leaves the uppermost layer drier between rains than repeated surface watering often does.

Evaporation does not disappear. If capillary continuity keeps the surface moist, water can still evaporate directly. A dense canopy or mulch can reduce that pathway. Transpiration also continues and usually dominates once a vigorous canopy forms. Transpiration is productive in the sense that it supports plant cooling and carbon uptake, but it is still water consumption. A high-yielding wicking bed may therefore use as much or more total water than a stressed surface-watered bed while producing more crop per unit of water.

Moisture and oxygen

The intended profile contains a saturated storage zone, a capillary transition, and an unsaturated root zone containing both water and air. Oxygen diffuses through air-filled pores much faster than through water. When the root zone becomes continuously saturated, respiration consumes oxygen faster than it is replenished. Root function declines, and conditions may favor water-mold pathogens. “Constant moisture” should therefore mean a relatively stable unsaturated moisture level, not a completely water-filled root zone.

9. Evidence for reduced water use

Controlled tomato experiment

Treatment	Measured outcome	Change from matched control (%)	Soil depth (mm)	Wicking irrigation (L)	Control irrigation (L)	Wicking yield (kg)	Control yield (kg)
Shallow WB 1	Applied water change	-9.7	300	121.7	134.8	2.85	1.78
Shallow WB 1	Yield change	59.6	300	121.7	134.8	2.85	1.78
Shallow WB 2	Applied water change	-7.1	300	125.2	134.8	2.89	1.78
Shallow WB 2	Yield change	62.5	300	125.2	134.8	2.89	1.78
Shallow WB 3	Applied water change	6.1	300	143	134.8	3.08	1.78
Shallow WB 3	Yield change	72.8	300	143	134.8	3.08	1.78
Deep WB	Applied water change	-22.1	600	123.1	158.1	2.18	2.46
Deep WB	Yield change	-11.2	600	123.1	158.1	2.18	2.46

Semananda et al. (2016) compared wicking beds with precise surface irrigation in an Adelaide glasshouse during the 2014–2015 spring and summer. Tomatoes grew in a poorly graded sandy soil. Treatments varied reservoir depth, soil depth, and the connection between water and soil. Surface controls were watered every two to three days according to measured need.

For three treatments with approximately 300 mm of soil, irrigation totals were 121.7, 125.2, and 143.0 L per plant, compared with 134.8 L for the shallow surface-watered control. Those values correspond to 9.7% less, 7.1% less, and 6.1% more applied water. Yields were 2.85, 2.89, and 3.08 kg per plant, compared with 1.78 kg for the control, or 59.6%, 62.5%, and 72.8% higher. The large water-use-efficiency gains came mainly from higher yield, not from a dramatic reduction in water input.

The deeper comparison gives a different result. A wicking treatment with about 600 mm of soil used 123.1 L and yielded 2.18 kg per plant. Its surface-watered comparison used 158.1 L and yielded 2.46 kg. Applied water was 22.1% lower, but yield was 11.2% lower. The treatment therefore did not improve irrigation water-use efficiency. Reservoir depth of 150 versus 300 mm did not produce a significant efficiency difference in that experiment.

The wicking treatments required fewer than 26 water additions, while surface treatments required 40 to 50. Early plants could go as long as four weeks between reservoir additions and mature plants generally one to two weeks after an establishment period of surface watering. These intervals are measured results under glasshouse conditions and should not be used as a Tucson refill schedule.

Lettuce and radish experiment

A later Adelaide glasshouse study compared wicking and surface watering for cos lettuce and two radish cultivars, with and without mulch. Wicking treatments had higher yield, biomass, and water-use efficiency across those crops, but the size of the benefit depended on crop, root form, soil depth, and mulch (Semananda et al., 2020). This broadens direct evidence beyond tomato and shows that a shallow root system can use a wicking profile. It does not establish performance in Tucson heat or mineral-rich water.

Recycled-water and salinity experiment

Semananda et al. (2021) ran tomato treatments using fresh water or recycled water with an electrical conductivity of 1.42 dS/m, about four times that of the study's fresh water. Treatments varied watering interval, mulch, setting, and bed volume. Freshwater wicking treatments watered on a two-day schedule yielded 1.43 and 1.54 kg per plant and used 68.32 and 72.52 L. A comparable fresh-water surface treatment yielded 1.16 kg and used 68.32 L. The calculated irrigation water-use-efficiency values were 20.90 and 21.20 g/L for the two wicking treatments and 17.00 g/L for the surface treatment.

The broader treatment set did not show a simple wicking advantage. A seven-day fresh-water wicking treatment used 53.70 L and yielded 1.05 kg. A recycled-water wicking treatment on the same interval used 43.34 L and yielded 0.78 kg. Across all conditions, the clearest statistically significant contrast involved freshwater wicking and recycled-water surface irrigation, which changes both irrigation method and water quality. The experiment supports the possibility of good performance with stored water, but it also shows that water quality and scheduling can dominate the comparison.

Home-garden evidence

Pollard et al. (2020) monitored 34 South Australian home food gardens containing 93 garden areas between November 2016 and June 2018. The sample was self-selected and management was not randomized. Wicking beds did not show a statistically significant advantage over other garden-area types in irrigation applied, total water input, or gross water-use efficiency. Median reported values for wicking areas were 95 L/m² per 30 days of total water, 52 L/m² per 30 days of irrigation, and 0.27 kg/m² per 30 days of harvest. The study's strongest practical difference was frequency: 43% of wicking areas were refilled weekly and 30% about every three weeks, while half of mixed-crop raised beds were watered daily.

This observational result does not prove that wicking beds save no water. Gardeners selected crops, planted at different densities, harvested differently, and may have overfilled reservoirs. It does show that a theoretical reduction in runoff and drainage does not automatically become lower household water use.

Interpretation for Tucson

The measured evidence supports three conclusions. First, water savings relative to careful surface irrigation can be modest, absent, or accompanied by lower yield. Second, water-use efficiency can improve when steady moisture increases yield. Third, refill frequency is more consistently improved than total water use.

A Tucson gardener replacing irregular hose watering in a freely draining raised bed could see a meaningful reduction because the starting method loses more water. A gardener replacing well-scheduled drip irrigation in mulched soil should expect a smaller and uncertain difference. Claims of 50%, 70%, or 90% savings cannot be responsibly applied to Tucson from the studies reviewed here.

10. Limitations and potential failure modes

Summer demand can exceed capillary supply

During Tucson's hot, dry pre-monsoon period, transpiration and surface evaporation can rise rapidly. The reservoir may contain water while the upper root zone becomes too dry because the medium cannot transmit water upward quickly enough. Tall crops with large leaf areas may be especially demanding. A refill record alone will not reveal this failure; soil moisture at root depth and midday plant response are more informative.

Reservoir and root-zone temperature

Semananda et al. (2016) observed lower maximum soil temperature in wicking treatments during hot periods and suggested that the water mass dampened temperature change. The response lagged air temperature by several hours. This was a limited observation in an Adelaide glasshouse, not a replicated Tucson outdoor temperature trial.

Stored water has thermal mass, so it can reduce a daytime peak and also retain heat after sunset. In an exposed Tucson bed, exterior material, color, shading, bed volume, wind, nighttime air temperature, and refill-water temperature could all matter. No study located for this review measured Tucson reservoir temperature through June heat or related it to root oxygen, disease, or yield. It is therefore reasonable to monitor temperature, but not to claim that the reservoir will necessarily cool roots.

Excess moisture, oxygen loss, and root disease

A wicking bed can become waterlogged when its overflow is blocked or too high, when stormwater enters faster than it leaves, when fine or compacted medium collapses air-filled pores, or when roots grow into the saturated zone. Oxygen stress can reduce nutrient uptake before obvious rot appears. Warm water also holds less dissolved oxygen than cool water.

Root rot is a group of diseases and physiological failures, not an inevitable outcome of subirrigation. Susceptibility depends on plant species, pathogen presence, temperature, saturated duration, and root-zone structure. A bed that maintains an aerated zone can support tomatoes, lettuce, and radishes in controlled trials. A bed that smells anaerobic, drains poorly after rain, or keeps the entire profile glossy-wet has departed from that condition.

Nutrient movement and fertilizer concentration

Nutrients dissolved in reservoir water move upward with the water. Plants remove some ions selectively. Evaporation removes none. Without adequate drainage, unused nutrients and other dissolved salts remain in the bed. Repeated fertilizer additions can therefore raise electrical conductivity even when plants show no immediate deficiency.

Closed greenhouse subirrigation provides indirect warning. In container tomato research, high nutrient-solution concentration produced substantially higher substrate salinity than drip irrigation and reduced root growth in the upper layer; lower concentrations reduced the severity of those effects (García-Santiago et al., 2019). That system recirculated nutrient solution and was not a soil wicking bed, so its numerical results should not be transferred. The direction of risk is relevant: a closed root zone makes fertilizer concentration more consequential.

Salt accumulation and flushing

Water taken up by roots or evaporated leaves most salts behind. Upward flow carries salts toward the evaporating surface, where they can accumulate. In Semananda et al. (2021), surface-layer electrical conductivity was lowest in fresh-water surface irrigation and highest in frequently supplied recycled-water wicking treatments. The highest reported surface value was about 2.63 dS/m in the recycled-water wicking treatment, compared with about 0.64 dS/m in the fresh-water surface treatment. Salinity in wicking beds generally decreased with depth, while surface watering carried more salt downward.

Those values came from one tomato experiment with defined water and fertilizer, not Tucson tap water. They demonstrate a transport pattern, not a local threshold. Plant damage depends on crop sensitivity and on the electrical conductivity of the saturated soil extract or another clearly specified test method.

Leaching requires enough relatively low-salinity water to dissolve and move salts beyond the active root zone and then out of the system. In a lined bed, that means deliberate discharge. General salinity research confirms that some drainage is necessary for long-term salt balance, although steady-state formulas can overestimate the amount needed under intermittent conditions (Letey et al., 2011). No validated calendar for flushing a Tucson wicking bed was found. A fixed monthly or seasonal rule would ignore water source, fertilizer, crop, rain, and accumulation rate. A defensible approach is to base the decision on repeatable electrical-conductivity measurements, crop symptoms, and known inputs. Flush water must be counted when comparing annual water use.

Monsoon overflow

July through September supplies roughly half of Tucson's normal annual rainfall, often in short storms. Rain first wets the surface and then adds to reservoir storage. Once storage is full, a functioning overflow should limit the water level. If the outlet is blocked, undersized for the inflow, or routed where water backs up, the root zone can remain saturated. If it works, the bed may discharge water containing fertilizer and salts.

No field study located for this review quantified storm capture, overflow water quality, or oxygen recovery in a Tucson wicking bed. The physical expectations are clear, but performance depends on the individual bed and storm. A bed should not be credited with capturing all monsoon rain simply because it contains a reservoir.

Mosquitoes and other pests

Mosquito larvae develop in standing water with little or no flow. A sealed reservoir does not provide habitat if adult mosquitoes cannot reach it. Fill, inspection, and overflow openings can create access. The Centers for Disease Control and Prevention recommends tightly covering stored water or screening openings with fine mesh (Centers for Disease Control and Prevention, 2024). Pima County similarly advises removing or excluding access to standing water (Pima County, n.d.). No comparative data show that functioning wicking beds create more mosquito problems than other water-holding garden objects. The risk is an access and maintenance problem, not an unavoidable feature.

Rodents, insects, roots, and sediment can also obstruct openings. Damp protected spaces may shelter ants or cockroaches. These risks are plausible and locally familiar, but no controlled wicking-bed incidence study was found.

Leaks, clogs, and hidden failures

A punctured liner converts stored water into uncontrolled drainage. Because the leak is buried, the first sign may be an unexpectedly short refill interval. Mineral precipitates, algae exposed to light, sediment, roots, or insects can obstruct small passages. Overflow failure is more serious because it can raise the water table during a storm. Regular refill volume, water-level behavior, and overflow response provide useful diagnostic information.

Published trials lasted months, not decades. They do not establish liner service life under Tucson ultraviolet exposure and heat, the rate of mineral clogging, or the probability of root intrusion. Long-term reliability claims are therefore practitioner estimates unless a manufacturer supplies relevant tested data for the actual materials and conditions.

Cost and labor

Pollard et al. (2020) estimated Australian retail setup costs of AUD 222.82/m² for wicking areas, compared with AUD 106.09/m² for conventional raised beds and AUD 18.91/m² for in-ground areas. These estimates reflected Australian materials and prices during the study period and excluded the gardener's labor. They are not Tucson quotations. The same study found median labor of 18 minutes/m² per 30 days for wicking areas, the highest among its compared categories, although that measure included all gardening work and reflected crop intensity as well as irrigation.

Wicking beds can reduce routine watering visits, but they add initial material, installation, monitoring, and eventual repair. A gardener who already requires an accessible raised bed may reasonably compare only the incremental reservoir cost. A gardener with usable ground should compare the entire structure against a drip line and mulch, not against another expensive raised bed.

11. Plant suitability in Tucson

Plant suitability depends on root aeration, rooting depth, salt tolerance, heat tolerance, season, and planting density. The irrigation method cannot make a cool-season crop tolerate Tucson's hottest weeks in full sun.

Strongest direct evidence

• Tomato: Multiple controlled wicking-bed experiments show that tomatoes can produce well with steady subsurface supply. Results varied with soil depth, water quality, schedule, and bed configuration (Semananda et al., 2016, 2021). Tucson summer fruit set and canopy heat remain separate limits.

• Cos lettuce: Controlled glasshouse evidence supports high yield and water-use efficiency in a shallow wicking profile (Semananda et al., 2020). In Tucson, lettuce remains mainly a cool-season crop. Constant moisture does not prevent heat stress or bolting.

• Radish: Two cultivars performed well in the shallow-rooted-crop experiment (Semananda et al., 2020). This directly contradicts the broad claim that all root crops are unsuitable. It does not prove that every long carrot, beet, or deep storage root will develop normally in every depth or moisture profile.

Likely compatible, with indirect rather than direct evidence

Leafy vegetables such as chard and many compact annual flowers are plausible candidates because they respond to a steady supply and commonly occupy the upper root zone. Basil, parsley, marigolds, zinnias, and other non-xeric annuals may also fit when their season and sunlight exposure are suitable. This is a physiological inference, not evidence that each crop saves water in a wicking bed.

The reported success of chard in the user's Tucson bed is consistent with this expectation. It shows that the particular combination of bed, season, water, and management supported chard. Without a side-by-side control and water measurements, it cannot show how much water was saved or which element caused the result.

Conditional crops

• Beans: Beans generally need even moisture during establishment, flowering, and pod fill but also require oxygen around roots. Germination can fail in excessively wet soil. Good performance in the user's bed suggests that its upper root zone was sufficiently aerated during that crop, not that beans prefer saturation. Variety, planting date, heat, and root-zone depth remain important.

• Sunflowers: Sunflowers tolerate drought through deep and extensive rooting, although irrigation can increase growth and yield. Strong growth in a wicking bed is plausible because continuous supply supports a large canopy. It does not demonstrate conservation. Tall plants may create high peak demand and anchorage problems in a shallow contained profile. Dwarf annual cultivars are a different case from full-size or perennial sunflowers.

• Peppers, eggplants, cucurbits, and many summer flowers: These can benefit from steady moisture but may have large seasonal demand. Their suitability depends on bed volume, capillary delivery rate, aeration, and heat management. A reservoir can run short or fail to deliver fast enough even when the crop is otherwise compatible.

Plants more likely to struggle or make poor use of the system

Plants adapted to drying soil and excellent drainage, including many cacti, succulents, desert natives, and Mediterranean herbs such as rosemary, thyme, and lavender, gain little from a continuously moist profile and may be more vulnerable to root disease. Trees, large shrubs, and long-lived woody perennials can outlast liners, enter the reservoir, and exceed the contained root volume. Deep-rooted crops may also reach a saturated layer or experience insufficient depth.

These are group-level cautions, not absolute prohibitions. A xeric plant in a bed with a dry upper zone may survive, but the reservoir and liner are unnecessary infrastructure for its normal water strategy. Drip irrigation or in-ground planting is usually a better match.

Flowers require the same separation of questions

“Flower” does not define a water response. A moisture-responsive annual grown for repeated blooms has different needs from a Sonoran Desert perennial. The decision should be based on rooting pattern, seasonal water demand, salt tolerance, and need for drying between irrigations. A mixed bed also creates compromise: the refill pattern suitable for chard or zinnias may be too wet for a drought-adapted herb.

12. Comparison with in-ground beds, conventional raised beds, drip irrigation, and ollas

Conventional in-ground beds

In-ground soil has the lowest infrastructure cost and usually the largest thermal mass. Roots can explore a larger volume, rainfall can infiltrate, and intentional deep watering can move salts below the root zone if drainage is adequate. University of Arizona vegetable-garden guidance likewise treats irrigation frequency as dependent on rooting depth and uses adequate watering to move salts away from roots (University of Arizona Cooperative Extension, n.d.-b). In Tucson, native soil may also be compacted, calcareous, shallow over caliche, poorly infiltrating, contaminated, or occupied by tree roots. Those conditions can justify a contained bed.

An in-ground bed with suitable amended soil, mulch, and well-scheduled drip is the strongest low-cost benchmark. It avoids liner failure and allows deep-rooted or long-lived plants. It can lose water below the crop if irrigation exceeds storage and rooting depth, but that loss is controllable through scheduling and may sometimes be required for salinity management.

Conventional raised beds with surface watering

A freely draining raised bed improves access, soil control, and drainage. In Tucson it also exposes more soil area to hot air and sunlight and often drains faster than the ground. Surface watering repeatedly wets the evaporation zone and can produce runoff or preferential drainage. Among the comparisons in this paper, this is the method a wicking bed is most likely to outperform in refill frequency and applied water.

The conventional raised bed has simpler failure modes. Water can be applied from above to leach salts, the drainage response is visible, and repairs do not require locating a buried reservoir leak. If a gardener needs rapid drying, grows plants with different moisture needs, or expects heavy monsoon inflow, free drainage can be an advantage rather than a defect.

Drip irrigation

Drip applies water slowly near selected plants and can limit surface wetting, runoff, and evaporation. Arizona Extension recommends a pressure regulator, filter, suitable emitters, periodic flushing, and regular inspection because lines can leak and emitters can clog (Schuch et al., 2016). The Food and Agriculture Organization uses about 90% as an indicative application-efficiency value for drip under appropriate conditions, but that is a planning benchmark rather than a measured Tucson home-garden result (Food and Agriculture Organization of the United Nations, n.d.). Poor scheduling can erase the advantage.

Drip is flexible. Different zones or emitter rates can serve crops with different requirements. It works in ground or raised beds, permits planned leaching, and uses standardized replaceable components. It requires a pressurized supply or elevated storage, exposed tubing can be damaged, and small emitters are vulnerable to mineral or sediment clogging. A timer reduces visits but does not store water at the bed.

Compared with drip, a wicking bed trades distribution hardware for a lined storage volume and capillary medium. It may provide longer autonomy during a supply interruption. It also makes the water level, leak condition, and salt pathway less visible. No direct Tucson trial shows that a wicking bed uses less annual water than correctly scheduled drip for the same crop and yield.

Ollas

An olla is a buried unglazed clay vessel filled with water. Water seeps through the wall as nearby soil dries, creating a localized wetted volume. Reviews and experiments support high efficiency in dry settings, but savings estimates vary widely and often compare ollas with watering cans or surface methods rather than modern drip (Bainbridge, 2001; Nickel & Brischke, 2021). Arizona Extension cites a 60% to 70% reduction from one maize comparison while noting that design, spacing, and crop evidence remain limited (Nickel & Brischke, 2021).

Ollas suit a few plants or small clusters, especially without a pressurized supply. Their finite wetted radius makes them less convenient for dense broadcast plantings or long rows. They must be refilled, can break, and may lose flow as mineral deposits or roots affect the porous wall. Aggressive or woody roots can encircle or damage them. Unlike a full-bed reservoir, each olla isolates a small zone; a failure affects less area and is easier to replace.

Comparative interpretation

No method is inherently efficient without reference to crop, yield, soil, schedule, and maintenance. Wicking beds reduce certain losses by storing drainage. Drip reduces certain losses by controlling placement and rate. Ollas combine local storage with porous release. In-ground soil supplies volume and thermal buffering. Conventional raised beds provide access and drainage at the cost of greater exposure and, often, more frequent irrigation.

13. Comparison table

The table below summarizes the best-supported relative tendencies for Tucson. “Low,” “medium,” and “high” are qualitative judgments, not measured scores. Actual performance can reverse with poor scheduling, unsuitable plants, a failed component, or a different soil and water source. The evidence behind the wicking-bed water rating is moderate for refill frequency and low to moderate for total annual savings in Tucson because no local field trial was found.

Method	Water-use potential	Plant suitability	Tucson heat	Salinity risk	Maintenance	Initial cost	Reliability	Best fit
Conventional in-ground bed	Variable; strong potential with mulch and correct scheduling	Broadest, including deep-rooted and long-lived plants	Best soil-volume and temperature buffer	Medium; can leach below roots if drainage is adequate	Low infrastructure; soil and weeds still require management	Low	High when soil and drainage are suitable	Usable native soil, low cost, large rooting volume
Conventional raised bed, surface watered	Usually weakest of these options in Tucson when unmulched or overwatered	Broad annual-crop range; moisture zones can be separated	Exposed soil dries and heats quickly	Medium; surface salts possible, but leaching is straightforward	Simple and visible; more frequent watering	Medium	High; few hidden components	Access and soil control where free drainage is desired
Drip irrigation	High potential when emitters and schedule match crop demand	Very broad; zones and emitter rates can differ	Does not cool the site; can deliver frequent small doses	Medium and manageable with planned leaching	Inspect leaks and emitters; filter, regulate pressure, and flush	Low to medium	High with maintenance and replaceable parts	General Tucson irrigation, mixed crops, existing beds
Ollas	High potential for a few localized plants; area-wide evidence is limited	Small clusters with roots inside the wetted radius	Buried storage is buffered; small volume can empty quickly	Medium; mineral clogging and salt near wetting front are possible	Manual refill; inspect for clogging, roots, and breakage	Medium per served area	Medium; simple but fragile and coverage is limited	A few plants, no pressure, localized manual watering
Wicking bed	Moderate potential versus surface-raised beds; uncertain versus good drip	Moisture-responsive annuals with compatible root depth	Reservoir may damp swings; Tucson summer data are absent	High upward-accumulation risk unless monitored and leached	Track refill, overflow, liner, moisture, and salts; failures may be hidden	High	Long-term field evidence is weak	Contained raised bed where fewer watering visits have high value

14. Circumstances in which a wicking bed is the best option

A wicking bed is most defensible when several of the following conditions occur together:

• A raised, contained root zone is already needed because of access, contaminated or unusable soil, shallow caliche, severe root competition, or the need to control the growing medium.

• The planting consists mainly of dense annual vegetables or flowers that prefer steady moisture and have roots compatible with the available unsaturated depth.

• The gardener's main constraint is watering frequency or short-term autonomy rather than proving the smallest possible number of gallons per pound of crop.

• The current alternative is hand or hose watering in a freely draining raised bed, especially when that method produces runoff, deep drainage, or repeated wetting of bare surface soil.

• The water source has manageable salinity, fertilizer additions are measured, and the gardener is willing to track electrical conductivity or another consistent salt indicator.

• Overflow can discharge safely during monsoon storms and its function can be checked.

• Refill volume and interval can be recorded. A sudden change can then identify a leak, clog, crop-demand shift, or weather effect.

• Planting season, canopy, and site exposure are managed for Tucson. The bed is not expected to make lettuce tolerate June sun or to keep an exposed reservoir cool without evidence.

In this setting the reservoir is doing useful work: replacing frequent irrigation visits, catching water that would otherwise drain immediately, and maintaining a steadier root-zone supply. The benefit remains conditional on salt export and aeration.

15. Circumstances in which it is not the best option

A wicking bed is usually not the best choice when one or more of these conditions dominate:

• Existing in-ground soil is suitable or can be improved at reasonable cost, and the gardener can use mulch and correctly scheduled drip.

• The planting is dominated by cacti, succulents, desert natives, Mediterranean herbs, trees, shrubs, or other plants that benefit from deep rooting or drying between irrigations.

• The primary goal is the lowest documented water use for a given yield. Current evidence cannot show that a Tucson wicking bed beats well-managed drip.

• Irrigation water or fertilizer creates a high salt load and there is no acceptable place to discharge leachate.

• The site receives concentrated roof runoff or intense monsoon inflow but overflow cannot be verified and routed safely.

• The gardener cannot inspect the bed, record abnormal refill behavior, or respond to a blocked overflow. A hidden reservoir is not maintenance-free.

• The bed will be exposed to maximum summer sun and reflected heat while crops have high demand, but there is no plan to observe root-zone moisture and temperature. Stored water is not proof of adequate delivery.

• Initial cost, future liner replacement, or difficult access to buried components outweighs the value of reduced watering visits.

In those cases, drip is generally the most adaptable alternative. Ollas can be preferable for a few isolated plants without pressure. A conventional raised bed can be preferable when free drainage and easy salt leaching matter more than stored water. In-ground planting is preferable when low cost, thermal stability, large rooting volume, and long service life matter most.

16. Evidence gaps and unanswered questions

The largest evidence gap is geographic. No peer-reviewed or university field trial located for this review compared wicking beds, drip, ollas, and conventional beds in Tucson or a closely matched hot-desert city using the same crops and yield measurement.

Specific unanswered questions include:

1. How much irrigation does a wicking bed save over correctly scheduled drip in Tucson when crop, planting density, mulch, shade, soil depth, and yield are held constant?

2. What upward flow rates can common Tucson bed media sustain during June afternoons, and at what root-zone depth does supply fail to match tomato, sunflower, cucurbit, or flower demand?

3. How hot do reservoirs and root zones become in exposed beds of different materials and volumes, and do they cool enough overnight to protect roots and oxygen supply?

4. How quickly do Tucson water sources and common fertilizers raise electrical conductivity in the surface and root zone? How much water is then required to restore an acceptable profile?

5. What happens during representative monsoon storms? The needed measurements include inflow, overflow, water-table duration, oxygen recovery, salt movement, and nutrient loss.

6. How long do liners, fittings, wicks, and overflow pathways remain reliable under heat, minerals, roots, rodents, and normal maintenance? Existing crop trials are too short to answer.

7. Which flowers and crop groups respond best when water use is measured per marketable yield or per flowering duration rather than only by biomass?

8. Do the reported successes of sunflowers, beans, and chard persist in paired beds with measured inputs? These observations are strong candidates for a local trial because they represent different root forms and water strategies.

A useful Tucson experiment would use replicated beds, identical crop densities, meters on every irrigation method, local weather and rainfall, electrical-conductivity profiles, multiple root-zone temperatures, overflow collection, marketable yield, and at least two years of operation. A single season would still miss component aging and year-to-year weather variation.

17. Conclusion

Wicking beds can reduce water use by storing drainage and supplying roots from below. Controlled research shows that they can increase yield and water-use efficiency for tomatoes, lettuce, and radishes, but the size and direction of the water difference are not consistent. In the strongest tomato comparison, shallow beds saved only 7.1% to 9.7% in two configurations, used 6.1% more in another, and produced much higher yields in all three. A deeper treatment saved 22.1% but yielded 11.2% less. Home-garden monitoring found longer watering intervals without a significant total water-use advantage. These findings do not support a universal savings percentage for Tucson.

The decision should begin with the problem being solved:

• Choose a wicking bed when a contained raised bed is already justified, crops prefer steady moisture, fewer watering visits have real value, and salt, overflow, and hidden failures will be monitored.

• Choose drip irrigation when crops have different water needs, existing soil is usable, precise scheduling and planned leaching matter, or the goal is a mature and repairable general irrigation system.

• Choose ollas when a small number of plants need localized stored water without pressure and manual refilling is acceptable.

• Choose a freely draining raised bed when access and soil control are needed but rapid drainage, simple inspection, and easy leaching are more important than stored water.

• Choose in-ground planting when soil quality, rooting volume, low cost, thermal buffering, and long-term reliability are favorable.

For Tucson, a wicking bed is best treated as a useful specialty system rather than the default conservation method. Its most certain advantage is reduced irrigation frequency. Its probable water advantage is greatest against poorly controlled surface watering and least certain against competent drip. The deciding conditions are crop moisture strategy, usable root depth, summer exposure, water salinity, a safe path for leaching and monsoon overflow, willingness to monitor the hidden reservoir, and the value of time between refills.

18. References

Bainbridge, D. A. (2001). Buried clay pot irrigation: A little known but very efficient traditional method of irrigation. Agricultural Water Management, 48(2), 79–88. https://doi.org/10.1016/S0378-3774(00)00119-0

Centers for Disease Control and Prevention. (2024). Mosquito control at home. https://www.cdc.gov/mosquitoes/mosquito-control/mosquito-control-at-home.html

Food and Agriculture Organization of the United Nations. (n.d.). Annex I: Irrigation efficiencies. https://www.fao.org/4/t7202e/t7202e08.htm

García-Santiago, J. C., Valdez-Aguilar, L. A., Cartmill, D. L., Cartmill, A. D., Juárez-López, P., & Alvarado-Camarillo, D. (2019). Subirrigation of container-grown tomato II: Physical and chemical properties of the growing medium. Water, 11(11), 2211. https://doi.org/10.3390/w11112211

Hillel, D. (1998). Environmental soil physics. Academic Press.

Institute of Technology, University of Moratuwa. (n.d.). Dr. (Mrs.) N. P. K. Semananda. https://itum.mrt.ac.lk/staff/dr-mrs-n-p-k-semananda

Letey, J., Hoffman, G. J., Hopmans, J. W., Grattan, S. R., Suarez, D., Corwin, D. L., Oster, J. D., Wu, L., & Amrhein, C. (2011). Evaluation of soil salinity leaching requirement guidelines. Agricultural Water Management, 98(4), 502–506. https://doi.org/10.1016/j.agwat.2010.08.009

National Weather Service. (n.d.). Tucson monthly and annual normals and extremes. National Oceanic and Atmospheric Administration. https://www.weather.gov/twc/TucsonMonthlyNormalExtremes

Nickel, A., & Brischke, D. (2021). Irrigating with ollas. University of Arizona Cooperative Extension. https://extension.arizona.edu/publication/irrigating-ollas

Pima County. (n.d.). Mosquitoes. https://www.pima.gov/2100/Mosquitoes

Pollard, G., Roetman, P., & Ward, J. (2020). Productivity, resource efficiency and financial savings: An investigation of the current capabilities and potential of South Australian home food gardens. PLOS ONE, 15(4), e0230232. https://doi.org/10.1371/journal.pone.0230232

Schuch, U. K., Allen, L. N., & Lohr, V. I. (2016). The basics of drip irrigation for landscaping. University of Arizona Cooperative Extension. https://extension.arizona.edu/publication/basics-drip-irrigation-landscaping

Semananda, N. P. K., Ward, J. D., & Myers, B. R. (2016). Evaluating the efficiency of wicking bed irrigation systems for small-scale urban agriculture. Horticulturae, 2(4), 13. https://doi.org/10.3390/horticulturae2040013

Semananda, N. P. K., Ward, J. D., & Myers, B. R. (2018). A semi-systematic review of capillary irrigation: The benefits, limitations, and opportunities. Horticulturae, 4(3), 23. https://doi.org/10.3390/horticulturae4030023

Semananda, N. P. K., Ward, J. D., & Myers, B. R. (2020). Experimental investigation of wicking bed irrigation using shallow-rooted crops grown under glasshouse conditions. Irrigation Science, 38, 117–129. https://doi.org/10.1007/s00271-019-00651-5

Semananda, N. P. K., Ward, J. D., & Myers, B. R. (2021). Assessing reliability of recycled water in wicking beds for sustainable urban agriculture. Earth, 2(3), 468–484. https://doi.org/10.3390/earth2030028

The University of Adelaide. (n.d.). Baden Myers. https://researchers.adelaide.edu.au/profile/baden.myers

Tucson Water. (2025a). 2025 annual water quality report: Main system. City of Tucson. https://www.tucsonaz.gov/files/sharedassets/public/v/1/city-services/tucson-water/water-quality/report-archive/ccr_mainsystem_2025.pdf

Tucson Water. (2025b). Hard water. City of Tucson. https://www.tucsonaz.gov/Departments/Water/Water-Quality/Aesthetics/Hard-Water

University of Arizona. (n.d.). Ursula K. Schuch. https://profiles.arizona.edu/person/uschuch

University of Arizona Cooperative Extension. (n.d.-a). May monthly gardening guide for Pima County. https://extension.arizona.edu/publication/may-monthly-gardening-guide-pima-county

University of Arizona Cooperative Extension. (n.d.-b). Ten steps to a successful vegetable garden. https://extension.arizona.edu/publication/ten-steps-successful-vegetable-garden