Soil moisture: what the station shows, when the field needs irrigating, and how it binds to evapotranspiration and VPD

Savvas Pastopoulos, Agronomist MSc, Pastopoulos Agronomics, Neos Mylotopos, Pella, Greece

The agrometeorological station shows a percentage for soil moisture, and the grower is asked to decide whether to irrigate. The percentage on its own is not a decision. The same "22%" means a saturated sand and a dry clay. And the point at which a field starts to run short of water is not fixed, because it moves with how much water the atmosphere is demanding that day. What follows is what the sensor measures, how the percentage becomes millimetres of water, where the irrigation limit sits, what evapotranspiration is and how it is computed step by step, what vapour pressure deficit is, and two calculators for the field.

1. Four different quantities, one word

The words "soil moisture" are used for four different quantities. Two say how much water is present. One says how tightly the soil holds it. The fourth is not measured at all, it is computed.

The distinction has a practical consequence. Two soils may both read 25%, one of them soaking and the other one wilting the plant. At the same percentage, the tension in a sand and in a clay is entirely different. The plant does not perceive amount, it perceives tension.

Sources: 2 3 30

2. What the percentage on the screen means

The stations installed in Greek fields carry sensors from different manufacturers. The word on the screen is the same. The quantity behind it is not.

The catalogue of a single manufacturer contains five soil moisture instruments returning four different units. Some return volumetric water content in m³/m³, one returns a percentage from 0 to 100, and two return tension in kPa [10]. All five are sold as "soil moisture sensors".

Until that answer exists, the reading remains useful as relative monitoring. It shows when the soil wets, how fast it dries, how deep the irrigation water reached, and when the curve stopped falling, which means the root stopped drawing. It remains useless as an absolute quantity, that is for computing millimetres of water or for comparison with a limit from a textbook table.

Sources: 10

3. How the sensor works and why it needs calibration

The capacitance sensor, the most widespread type in stations, does not measure water. It measures an electrical property of the mixture of soil, water and air, the apparent dielectric permittivity. Water is a strongly polar molecule, minerals and air are not, so the more water is present the higher the permittivity [3]. Between permittivity and water content sits a calibration equation, and that is where the problem is.

What shifts the curve

The factory equation is an average for mineral soils. Four properties of the particular soil displace it, bulk density, mineralogy, texture and salinity [3]. The manufacturer quotes an accuracy of 3 to 4 percentage points with the factory curve and 1 to 2 points after a calibration in the field [3]. The figures measured in real soil are worse.

The same was found in a Greek study. An evaluation of two commercial capacitance sensors found significant deviations from the behaviour the manufacturer states, and derived calibration equations for the particular soil [5].

What this means for the grower

It does not mean the sensor is useless. It means the relative change is reliable and the absolute value is not. The curve rises and falls correctly. The number on the vertical axis may be displaced by ten points.

The cheap correction is two reference points taken from the field itself. The upper point is taken two days after a deep irrigation or a heavy rain, once drainage has stopped. The lower one is taken at the end of a long dry period without irrigation, once the curve has gone flat. The range between them is the available water of that field, expressed in the language of its own sensor. After that, the percentage becomes interpretable even if what it measures is still unknown.

Sources: 3 4 5 10

4. Field capacity and permanent wilting point

The available water of a soil has two ends. The upper one is field capacity, the moisture that remains once the gravitational water has drained, one to three days after saturation. The lower one is the permanent wilting point, the moisture at which the plant does not recover even in a saturated atmosphere.

The tension values

The classic limits are -33 kPa for field capacity and -1,500 kPa for the permanent wilting point, that is one third of an atmosphere and fifteen atmospheres [2]. In the applied literature they are given as ranges, 10 to 33 kPa and 500 to 3,000 kPa [6].

Texture moves the upper end. For coarse-textured soils field capacity sits near -10 kPa, because the large pores empty at a much lower tension [7]. If -33 kPa is used on a sandy soil, the available water comes out smaller than the real one.

The volumetric values by texture

The table below gives the typical ranges by texture class, as FAO-56 lists them [1].

Values in m³ of water per m³ of soil. For a percentage, multiply by one hundred. The FAO-56 table has no rows for clay loam and sandy clay loam, which are common classes in the Greek plain.

For the missing classes, and as a second independent source, the following are mean values from American data, in percentages [6].

Averages from American soils, which have not been measured in Greek ones. The same source also gives measurements from real Oklahoma fields, where loam came out at 25% and 13% instead of 27% and 12%. The departure of the measured values from the averages is itself the argument for measuring in the field.

Sources: 1 2 6 7

5. From the percentage to millimetres, the available water

The percentage becomes useful the moment it is converted into millimetres. Millimetres are the common language of rainfall, irrigation and evapotranspiration. One millimetre of water equals one litre per square metre, that is ten cubic metres per hectare.

The two quantities

The p is the number that does the work. It is the fraction of the available water that can be depleted before water stress appears, and FAO-56 tabulates it by crop. It ranges from 0.30 to 0.70, with 0.50 as the common value [1].

From FAO-56 Table 22 [1]. Rooting depth is given as a range because it depends on the rootstock, the soil and the irrigation system. Under drip with closely spaced emitters the active root is shallower than the range of the table, and the available water smaller.

Why one sensor at one depth is not enough

The root zone does not empty evenly. The root draws first from the upper layers, where the water is given up most easily, and moves down as those dry. A sensor at 20 centimetres shows a fall long before the profile is empty. One at 60 centimetres shows stability while the upper part has already dried out.

The recommendation for tree crops is two to three measurements at different depths per position, typically around 30, 60 and 90 centimetres [8]. The deepest sensor is the one that judges the dose. It says whether the water reached that far, that is whether the dose was sufficient, or whether it passed through, that is whether it is draining below the root.

Sources: 1 8

6. When the soil is considered to need irrigating

There are three ways to answer the question, and they are not equivalent.

The first way, with tension

It is the simplest, because the soil need not be known. For orchards, the general recommendation is to begin irrigating at 30 kPa, that is 30 centibar [8]. In field and vegetable crops the limits are more differentiated.

Values from American applied literature [6]. No corresponding Greek table of tension limits was found, neither in Ministry guidance nor in a published Greek study. They are given as an order of magnitude and not as a recommendation for a Greek field, because the limit depends on the irrigation system and on how frequently it runs.

The same tension limit corresponds to an entirely different depletion depending on the soil. The tension at which half the available water is gone is 20 kPa in sand, 30 in loamy sand, 40 in sandy loam, 65 in loam and 90 in silt loam [8]. A limit of 30 kPa empties half the available water of a sand, and barely one fifth of a silt loam.

The second way, with the depletion fraction

This is the FAO-56 way. Irrigation starts when the depletion of the root zone reaches RAW, that is p times the available water. It requires the volumetric water content and both ends of the range to be known. It has one large advantage, though, the limit is not fixed, it adapts to the demand of the atmosphere. Section 11 develops this.

The third way, with the water balance

This uses no soil sensor. It starts from a full root zone, subtracts the crop evapotranspiration every day, adds rainfall and irrigation, and irrigates when the balance reaches the limit [9]. It is bookkeeping, and it works. Its weakness is that errors accumulate. Every rainfall the station mismeasured and every irrigation that went unrecorded shifts the balance for the rest of the season.

Sources: 1 6 8 9

7. Evapotranspiration, what it is exactly

The two flows

Water leaves a field by two routes that happen at the same time and are not easily separated in a measurement.

Evaporation is the conversion of liquid water into vapour from the soil surface, from the wet surfaces of leaves after rain or spraying, and from free water. It is purely physical. It is fuelled by energy, mainly solar radiation, and governed by temperature, by the wind that removes the saturated vapour from the surface, and by how wet the surface itself is. As the upper soil layer dries, evaporation falls sharply, because the water must now rise from deeper down.

Transpiration is the loss of water from inside the plant to the atmosphere, through the stomata. Water reaching the leaf evaporates from the cell walls inside the mesophyll, and the vapour leaves through the stoma. It is biological in the sense that the plant controls it: it opens the stomata to take in carbon dioxide and closes them when the loss of water becomes threatening. The important point is that transpiration is not waste. It is the unavoidable price of photosynthesis, because the same opening that lets carbon dioxide in lets water out.

In an early crop, with small plants and bare soil, most of the evapotranspiration is evaporation. In a fully developed canopy that shades the ground, most of it is transpiration. The same value of six millimetres means entirely different things in the two cases, and calls for entirely different measures: in the first, evaporation is reduced by covering the soil and by irrigating without wetting the surface; in the second, transpiration is not reduced without reducing yield as well.

The energy behind the number

Evaporating one kilogram of water takes about 2.45 megajoules of energy, which is the latent heat of vaporisation at ordinary temperatures [1]. That explains two things that look unrelated.

The first is that evapotranspiration has an upper limit set by the sun. As much energy as reaches the surface, that much water can at most leave. This is why a cloudy day has a low evapotranspiration however warm it is, and why December evapotranspiration in the Peloponnese is around one millimetre a day while August approaches five and a half [22].

The second is that transpiration cools the leaf. The energy spent evaporating the water no longer heats the tissue. A leaf transpiring freely is cooler than a leaf that has closed its stomata, and the difference can be several degrees. This is the basis of thermal detection of water stress, and it returns in section 10.

Sources: 1 22

8. Computing evapotranspiration, step by step

Nobody measures the evapotranspiration of a field directly. Direct measurement needs a lysimeter or an eddy covariance tower, and is done at research installations. In practice evapotranspiration is computed from meteorological data, in three steps.

8.1 First step, the reference surface

The problem to be solved is that evapotranspiration depends at once on the weather and on the crop itself. The solution was to define a hypothetical reference surface, the same everywhere, so that all the weight falls on the weather. FAO-56 defines it as a hypothetical reference crop with an assumed crop height of 0.12 metres, a fixed surface resistance of 70 seconds per metre and an albedo of 0.23 [1]. It is, in practice, a well-watered grass that is never short of water.

The evapotranspiration of that surface is called reference evapotranspiration and is written ET₀. It is a purely meteorological quantity. It describes no real crop, and that is the point: two fields with different crops in the same area have the same ET₀.

8.2 The Penman-Monteith equation

The structure of the equation says something worth reading without the mathematics. The first term of the numerator, the one with the radiation, is the energy side: how much energy there is to evaporate water. The second term, the one with the wind and the vapour pressure deficit, is the aerodynamic side: how fast the atmosphere removes the vapour. A day with strong sun and no wind is driven by the first. A day with a dry wind is driven by the second, and this is why the etesian winds dry a field faster than the temperature alone would justify.

8.3 Second step, the crop coefficient

The crop coefficient K_c absorbs all the differences between the real crop and the reference grass: height, albedo, canopy resistance, and how much soil stays exposed. It changes through the season, which is why it is given as three values, for the initial, the mid and the late stage.

From FAO-56 Table 12 [1]. FAO-56 does not separate peach from apricot and plum, it gives them as one group, and likewise cherry together with apple and pear. The pair of values in the late K_c refers to the condition of the crop at harvest.

8.4 Adjusting K_c for the Greek climate

The table values are calibrated for a sub-humid climate, with a minimum relative humidity around 45% and wind around 2 metres per second. The Greek summer is not that. FAO-56 gives an explicit adjustment.

The adjustment raises K_c whenever the air is drier than 45% or the wind stronger than 2 m/s, which is almost always the case in the Greek plain in July. For a 3-metre peach orchard with an active ground cover, a minimum relative humidity of 25% and a wind of 2 m/s, the term (h/3)^0.3 equals one and the adjustment gives:

At an ET₀ of 7 millimetres, the difference between 1.15 and 1.23 is half a millimetre a day, that is 150 cubic metres per hectare over a month. For the late K_c the same adjustment is applied only when the table value exceeds 0.45 [1].

8.5 Third step, the stress coefficient

This step is often skipped, and the error runs in one direction. Once the soil has dried, actual consumption is lower than ETc, because the plant has closed its stomata. A balance that subtracts ETc every day without the Ks coefficient overestimates depletion and leads to more irrigation than the real requirement.

Sources: 1

9. Greek evapotranspiration figures

The general FAO-56 classification places Northern Greece in July in the category "temperate region, arid and semi-arid, warm", with a range of 6 to 9 millimetres per day [1]. That is a generic bracket and not a measurement. The Greek measurements that were found are the following.

Projected increase

For the Thessaly plain, the six-model ensemble projects an increase in annual reference evapotranspiration of 6 to 7% by the 2021 to 2050 period and of 10.5% by 2071 to 2100, with the summer increase at 10.5 to 12% [21]. These are simulations and not observations.

How much water Greek crops consume

The published Greek figures are fewer than one would expect, and most of them concern field crops.

The simpler methods, what they give in Greece

When a station does not measure solar radiation or wind, the Penman-Monteith equation cannot be fully applied and simpler methods are used. Two Greek comparative studies give a picture.

At two stations in Western Macedonia, nine methods were compared against FAO-56. The Penman-type methods gave the closest agreement, with a coefficient of determination above 0.97, while Hargreaves-Samani gave 0.923 at Krania and 0.854 at Kozani. All methods perform better in winter and diverge most in the warm season, that is, exactly when the accuracy is needed [19].

At three Greek cultivation sites, ERA5-Land reanalysis data underestimated in situ evapotranspiration by about 35%, while among the empirical methods Valiantzas overestimated by 32 to 62% and Copais by 24 to 56%. Hargreaves-Samani was found to be the most efficient of the simple methods [29]. In Peloponnese Augusts, the same method underestimated systematically with a root mean square difference of 0.72 to 0.76 millimetres a day [22].

Sources: 1 19 20 21 22 23 24 25 26 27 29

10. Vapour pressure deficit

Air can hold a maximum amount of water vapour, and that maximum rises exponentially with temperature. The vapour pressure deficit, abbreviated VPD, is the difference between what the air can hold and what it already holds. It is, literally, how empty the air is.

10.1 The calculation

What matters in the first equation is the exp. Saturation vapour pressure does not rise in proportion to temperature, it rises exponentially. From 25 to 35 degrees, that is ten degrees, it nearly doubles. This is why a day at 38 degrees is not "a little worse" than one at 33, it is far worse.

Computed from the FAO-56 equations [1]. No published VPD measurements from a Greek field were found, in Greek or in English. The values in the table are computed and not measured, and are given so that the order of magnitude of a Greek summer afternoon is visible.

10.2 The error of using mean humidity

The usual way to compute a daily VPD is to take the mean temperature and the mean relative humidity of the day. It is wrong, and FAO-56 states so explicitly: using mean air temperature instead of daily minimum and maximum results in a lower mean saturation vapour pressure, a smaller deficit, and therefore an underestimation of reference evapotranspiration [1].

The correct way uses the extremes of the day.

Computed from the FAO-56 equations for three typical summer days of Northern Greece. The days are chosen as examples and do not come from a particular station.

The error is large and always in the same direction. The rough calculation underestimates the daily deficit by 23 to 26%. In the third row of the table, 1.70 kPa becomes 1.27 kPa, so the day moves from "above the point at which kiwifruit stomata are closing" to "comfortably below it". The decision changes because of the way the number was computed.

10.3 The leaf does not have the temperature of the air

The driving force of transpiration is not the deficit of the air but the deficit between the inside of the leaf, which is saturated at leaf temperature, and the air. A leaf in the sun is warmer than the air whenever the radiation it absorbs exceeds what transpiration and convection remove.

In measurements of leaf temperature in tree canopies, sunlit leaves were found to be 8 to 10 degrees warmer than the air at midday, with a maximum recorded difference of 12.06 degrees. The crossover sits near 25 degrees of air temperature: above it leaves run warmer, below it cooler [15].

Those measurements were made in forest canopies in Australia, in a temperate eucalypt woodland and in a tropical rainforest. No equivalent measurement in a Mediterranean orchard was found. They are carried over as the order of magnitude of the phenomenon and not as a value for a Greek tree.

The consequence is easy to compute. With air at 35 degrees and 25% relative humidity, the actual vapour pressure of the air is fixed at 1.406 kPa.

10.4 Where the stomata close

There is no single number for all species. There are measurements by species, and they are instructive.

For field crops, a review collected the deficit above which transpiration stops rising linearly, that is where the stomata begin to restrict: durum wheat 1.05 to 1.10 kPa, soybean 1.10 to 2.19, sorghum 1.17 to 4.42, maize 1.30 to 4.14, pearl millet 1.45 to 4.35, chickpea 1.69 to 3.07, wheat 1.86 to 3.89 [13]. The ranges are wide because they depend on genotype and conditions, but the lower end sits consistently around 1 kPa.

For trees, a census of five species in Sicily gave the deficit each species experiences during the growth of its fruit: peach, mean 2.1 and maximum 3.84 kPa, mango 1.11 and 1.97, olive 1.1 and 2.15, orange 1.1 and 1.8, loquat 0.6 and 1.4 [14]. Peach therefore develops its fruit at a deficit that is permanently above the limits of the field crops and above the point at which kiwifruit has already closed half its stomata.

10.5 The midday depression

The photosynthesis of a tree does not follow the sun. It peaks in the morning and falls towards midday, even when the soil holds plenty of water. In the Spanish olive orchard, assimilation fell to 6 to 11 units on days of high evaporative demand, against 15 units on a day of low demand, a depression of 27 to 60% [12]. In kiwifruit, the authors attribute the midday depression explicitly to the combination of the deficit arising inside the leaf from transpiration and the direct effect of the rising vapour pressure deficit [11].

Sources: 1 11 12 13 14 15 28

11. How the three quantities bind together

11.1 The same moisture, two different answers

The central finding of the recent literature is that the availability of soil water is not a property of the soil alone. It is defined jointly with the demand of the atmosphere. When the vapour pressure deficit is high, the transpiration rate required to satisfy the atmosphere creates a hydraulic gradient that the soil and root pathway cannot sustain. Stomata close at a soil water content that would have been perfectly adequate on a humid day [13].

The mechanism is hydraulic. As the rhizosphere dries, the hydraulic conductance of the soil and of the soil to root interface collapses, and that collapse interacts with the flow rate the deficit demands. It is not that the plant "senses" less water. It is that the water does not arrive in time.

11.2 The controlled experiment

The cleanest demonstration that the two act independently comes from a greenhouse, where one can be held constant while the other changes. Tomato and cucumber were tested at two levels of soil moisture, 100% and 60% of field capacity, crossed with two levels of vapour pressure deficit, high at 3 to 6 kPa and low at 1.5 to 2.5 kPa [16].

Three findings matter for the field.

• At the same water shortage, the low deficit rescued the plant. At 60% of field capacity, the relative water content of the leaves was 11% higher at the low deficit than at the high one [16]. Same soil, different atmosphere, different plant water status.
• The low deficit restored photosynthesis. The reduction in stomatal conductance caused by drought was largely reversed when the deficit was kept low [16].
• Soil moisture remained the dominant factor for biomass. The fresh and dry weights of tomato at 100% of field capacity were 40.25% and 36.79% higher than at 60% [16]. The low deficit helps, but it does not replace water.

The experiment was carried out in a greenhouse, in pots, in China, with fogging systems controlling humidity. It does not transfer as a practice to open field crops, where the vapour pressure deficit is not controlled. It transfers as a mechanism, that is, as proof that the two factors are separate and additive.

11.3 Why the two coincide more often than chance would suggest

The two adverse conditions do not arrive independently. High atmospheric demand accelerates evapotranspiration, which reduces soil moisture, and the continuing loss of soil water in turn limits evapotranspiration, increasing sensible heat and making the near-surface air hotter and drier still [17].

In an analysis of 37 years over Chinese territory, the probability of high vapour pressure deficit and low soil moisture occurring together was found to be significantly greater than the product of the two separate probabilities, and the difference grew as the stress threshold became more severe. The coupling of the two was strongest in the semi-arid and sub-humid regions, that is, in the transitional zones, and weakest in the extremely arid and in the humid ones [17].

The study concerns China and rests on remote sensing and reanalysis data, with soil moisture representing the surface layer of 0 to 7 centimetres. The percentages and the regions do not transfer. What transfers is the structure of the phenomenon, and the fact that the Greek agricultural zone belongs climatically to exactly the transitional categories where the coupling was strongest.

11.4 The quantitative expression, inside FAO-56 itself

FAO-56 does not leave the matter in qualitative terms. The depletion fraction p is tabulated for an evapotranspiration of about 5 millimetres a day, and is explicitly adjusted for demand.

That single line is the translation of this whole section into practice. The irrigation limit moves with the demand of the atmosphere, without anything changing in the soil.

The peach orchard example of section 5, with an available water of 144 millimetres, gives a readily available water of 83.5 millimetres on the mild day and 54.7 in the heatwave. In days the difference is larger still, because the heatwave empties a smaller reserve faster: twenty-eight days against seven.

For stone fruit there is also the reverse case, where controlled water stress is deliberate rather than a mistake. The subject is developed in the article on deficit irrigation in peach and nectarine, where the stress is applied at a specific phenological stage and for a specific purpose.

Sources: 1 11 13 16 17 18

12. The two calculators

The first converts the sensor reading into millimetres of water and into days. The second computes the vapour pressure deficit from temperature and relative humidity.

13. What the station provider needs to be asked

The questions that follow are not technical detail. Without their answers, the soil moisture reading cannot enter the calculations of this article.

14. The points that remain open

Writing this article brought up gaps that are not covered by estimation and are stated as they are.

• The unit, the depth and the calibration of the soil moisture sensors of the agrometeorological station networks operating in Greece are not published. Without them, the reading does not convert into millimetres.
• No published reference evapotranspiration series was found for Pella, Imathia or Thessaloniki with a daily July value. The nearest one is Kozani, at 625 metres of elevation.
• No Greek table of tension limits for starting irrigation by crop was found, neither in Ministry guidance nor in a published study. The values in circulation come from American sources.
• No measured seasonal crop evapotranspiration was found for peach, cherry or kiwifruit in Greece, by lysimeter or eddy covariance. For kiwifruit there are applied irrigation amounts, which is not the same thing.
• No published vapour pressure deficit measurements from a Greek field were found. The values given here are computed from the equations.
• No measurement of the leaf to air temperature difference in a Mediterranean orchard was found. The values used come from forest canopies in Australia.
• No measurement of the midday depression of photosynthesis in a Greek orchard was found. The nearest ones are from Spain and Italy.

15. Sources

The numbers in the text refer to the list below.

• [1] Allen, R.G., Pereira, L.S., Raes, D. and Smith, M. (1998). Crop evapotranspiration, guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper 56, FAO, Rome. Source for the definition of the reference surface with a height of 0.12 m, a resistance of 70 s/m and an albedo of 0.23; for the Penman-Monteith equation and all its symbols; for the saturation and actual vapour pressure equations and the explicit warning about mean temperature; for Table 12 of crop coefficients and their climate adjustment; for Table 19 of soil water characteristics; for Table 22 of rooting depth and depletion fraction and its adjustment for evapotranspiration; and for the equations of total and readily available water and of the stress coefficient. Full text used.
• [2] Easton, Z.M. and Bock, E. Soil and Soil Water Relationships. Virginia Cooperative Extension, BSE-194P. Source for the definitions of volumetric and gravimetric water content, the conversion between them through bulk density, and the field capacity and permanent wilting point values in atmospheres. Full text used.
• [3] METER Group. The researcher's complete guide to soil moisture. Source for the distinction between extensive and intensive quantities, for the working principle of the capacitance sensor through dielectric permittivity, for the accuracy of 3 to 4 points with factory calibration against 1 to 2 points with soil-specific calibration, and for the four properties that displace the curve. Full text used. This is educational material from a sensor manufacturer and not a peer-reviewed paper.
• [4] Zawilski, B.M., Granouillac, F., Claverie, N., Lemaire, B., Brut, A. and Tallec, T. (2023). Calculation of soil water content using dielectric-permittivity-based sensors, benefits of soil-specific calibration. Geoscientific Instrumentation, Methods and Data Systems 12:45-56. Source for the relative errors of up to +115% and +245% with factory settings on a clay soil, for the mean overestimation of 0.10 m³/m³ at the dry end, and for the reduction of the error below 0.05 m³/m³ after soil-specific calibration. Full text, open access.
• [5] Fragkos, A. Evaluation and calibration of soil moisture sensors. Doctoral thesis, Agricultural University of Athens. In Greek. Source for the deviations of two commercial capacitance sensors from their stated behaviour and for the criticism of accepting factory calibrations as a source of systematic error, in Greek soils. Full text used.
• [6] Datta, S., Taghvaeian, S. and Stivers, J. (2018). Understanding Soil Water Content and Thresholds for Irrigation Management. Oklahoma State University Extension, BAE-1537. Source for the ranges of field capacity at 10 to 33 kPa and of the wilting point at 500 to 3,000 kPa, for the table of field capacity and wilting point by texture in percentages, for the Oklahoma measurements that depart from the averages, and for the table of tension limits by crop. Full text used. The data are American.
• [7] Campbell, G. (2015). Field Capacity and Permanent Wilting Point, do standards need to be changed? Environmental Biophysics. Source for the -10 kPa value of field capacity in coarse-textured soils, for the criticism of the fixed -33 and -1,500 kPa values, and for the doubled water retention of a layered profile with a fine horizon over a coarse one. Full text used. This is an opinion piece by an experienced researcher and not a peer-reviewed paper, and it is used as a caveat and not as a source of numbers.
• [8] Gohil, H. and Carleo, J. (2019). Irrigating Your Orchard, Water Monitoring with a Tensiometer. Rutgers New Jersey Agricultural Experiment Station, FS1306. Source for the irrigation onset limit of 30 centibar in orchards, for the table of tension corresponding to 50% depletion by soil texture, and for the recommendation of two to three measurements at different depths per position. Full text used. The data are American.
• [9] Henry, C., Francis, P.B. and Espinoza, L. (2020). How to Use Watermark Soil Moisture Sensors for Irrigation. University of Arkansas Division of Agriculture, Cooperative Extension Service, FSA58. Source for the operating limits of 60 centibar in silt loams and clays and 20 to 35 centibar in sandy soils, and for the allowable depletion fractions by irrigation system. Full text used. The data are American.
• [10] Pessl Instruments, METOS. Soil moisture sensors. Source for the fact that five instruments from one catalogue return four different units for what is called soil moisture, and for the depths and measurement ranges of each type. A commercial catalogue, used as evidence of the variety of units and not as a scientific source.
• [11] Gucci, R., Massai, R., Xiloyannis, C. and Flore, J.A. (1996). The Effect of Drought and Vapour Pressure Deficit on Gas Exchange of Young Kiwifruit Vines. Annals of Botany 77(6):605-613. Source for the steady decline of kiwifruit stomatal conductance from 0.8 to 2.5 kPa, for the point at which conductance falls to half at 2.1 to 2.2 kPa, for the net daily carbon gain of 4.7 against 2.7 grams per square metre, and for the attribution of the midday depression to the combination of the leaf deficit and the direct effect of the vapour pressure deficit. Abstract and extracts used. The work was carried out in Italy.
• [12] Moriana, A., Villalobos, F.J. and Fereres, E. (2002). Stomatal and photosynthetic responses of olive leaves to water deficits. Plant, Cell and Environment 25:395-405. Source for the linear decline of olive conductance with the vapour pressure deficit and the four equations by water potential class, for the slope of 20.7 units per kPa in well-watered trees, and for the midday depression of assimilation from 15 to 6 to 11 units. Full text used. The work was carried out at Córdoba, Spain.
• [13] Koehler, T., Wankmüller, F.J.P., Sadok, W. and Carminati, A. (2023). Transpiration response to soil drying versus increasing vapor pressure deficit in crops, physical and physiological mechanisms and key plant traits. Journal of Experimental Botany 74(16):4789-4807. Source for the table of transpiration breakpoint deficits by crop, and for the statement that the same soil water content is adequate at low vapour pressure deficit and limiting at high, with the hydraulic mechanism that explains it. Full text used. This is a review collecting work from many countries.
• [14] Carella, A., Massenti, R. and Lo Bianco, R. (2023). Testing effects of vapor pressure deficit on fruit growth, a comparative approach using peach, mango, olive, orange, and loquat. Frontiers in Plant Science 14:1294195. Source for the mean and maximum vapour pressure deficit each species experiences during fruit growth, with peach at 2.1 and 3.84 kPa and olive at 1.1 and 2.15 kPa. Full text, open access. The measurements were made in Sicily.
• [15] Crous, K.Y. et al. (2023). Similar patterns of leaf temperatures and thermal acclimation to warming in temperate and tropical tree canopies. Tree Physiology 43(8):1383-1399. Source for the 8 to 10 degree difference between a sunlit leaf and the air at midday, for the maximum recorded difference of 12.06 degrees, and for the crossover near 25 degrees. Full text used. The measurements were made in forest canopies in Australia and not in an orchard.
• [16] Song, X., Miao, L., Jiao, X., Ibrahim, M. and Li, J. (2022). Regulating Vapor Pressure Deficit and Soil Moisture Improves Tomato and Cucumber Plant Growth and Water Productivity in the Greenhouse. Horticulturae 8(2):147. Source for the 11% increase in relative leaf water content at the low vapour pressure deficit under the same water shortage, for the recovery of stomatal conductance, and for the 40.25% and 36.79% difference in the fresh and dry weight of tomato between 100% and 60% of field capacity. Full text, open access. The experiment was carried out in a greenhouse, in pots, in China.
• [17] Nie, C. et al. (2024). The Coupling Characteristics of Vapor Pressure Deficit and Soil Moisture in China. Remote Sensing 16:4387. Source for the land and atmosphere interaction mechanism that links the vapour pressure deficit to soil moisture, and for the finding that the probability of the two occurring together is greater than the product of the separate probabilities, with the coupling strongest in semi-arid and sub-humid regions. Full text, open access. The study concerns China, with remote sensing and reanalysis data, and soil moisture of the 0 to 7 centimetre layer. The percentages do not transfer.
• [18] Roby, M.C., Scott, R.L. and Moore, D.J.P. (2020). High vapor pressure deficit decreases the productivity and water use efficiency of rain-induced pulses in semiarid ecosystems. Journal of Geophysical Research, Biogeosciences 125:e2020JG005665. Source for the reduction of gross photosynthesis by 7 to 27% and of surface conductance by 22 to 34% under high vapour pressure deficit, in periods with adequate soil moisture. Full text used. Four sites in Arizona.
• [19] Efthimiou, N., Alexandris, S., Karavitis, C. and Mamassis, N. (2013). Comparative analysis of reference evapotranspiration estimation between various methods and the FAO56 Penman-Monteith procedure. European Water 42:19-34. Source for the annual reference evapotranspiration of 952 millimetres at Kozani and 822.8 at Krania, for the July values of 166.8 and 143.9 millimetres, and for the comparison of nine methods against FAO-56 at two stations in Western Macedonia. Full text used. The error magnitudes of that study are not reproduced here, because in retrieval its two error columns appeared mutually inconsistent.
• [20] Karamanos, A., Aggelides, S. and Londra, P. (2007). Water use efficiency and water productivity in Greece. Options Méditerranéennes, Series B 57:91-99, CIHEAM, Bari. Source for the annual reference evapotranspiration of 1,058 millimetres in Northern Greece and 2,015 millimetres in the south-east. Full text used.
• [21] Nastos, P., Kapsomenakis, J., Kotsopoulos, S. and Poulos, S. (2015). Present and future projected reference evapotranspiration over Thessaly plain, Greece, based on regional climate models simulations. European Water 51:63-72. Source for the annual reference evapotranspiration of 900 to 1,100 millimetres over the Thessaly plain, for the summer value of 420 to 540 millimetres, and for the projected increase of 6 to 7% by 2050. Full text used. These are simulations with climate models and not station measurements.
• [22] Dimitriadou, S. and Nikolakopoulos, K.G. (2021). Reference Evapotranspiration Methods Implemented as ArcMap Models with Remote-Sensed and Ground-Based Inputs, Examined along with MODIS ET, for Peloponnese, Greece. ISPRS International Journal of Geo-Information 10(6):390. Source for the mean daily reference evapotranspiration of 5.42 to 5.62 millimetres in August and 1.13 to 1.37 millimetres in December at 62 stations in the Peloponnese, and for the systematic underestimation by Hargreaves-Samani with a root mean square difference of 0.72 to 0.76 millimetres per day. Full text, open access.
• [23] Kokkora, M., Koukouli, P., Karpouzos, D. and Georgiou, P. (2023). Model Application for Estimation of Agri-Environmental Indicators of Kiwi Production, A Case Study in Northern Greece. Environments 10(4):69. Source for the applied irrigation amounts of 599 to 1,350 millimetres on two kiwifruit plots in Pieria, and for the authors' conclusion that in most cases the water applied exceeded the needs, in certain months by more than double. Full text, open access.
• [24] Kanakas, V., Doulgeris, C., Fountas, S., Danalatos, N. et al. (2018). Rational irrigation practices in the agricultural sector. In Greek. Source for the total consumption of cotton at around 700 millimetres with a peak of about 10 millimetres per day at flowering and boll set. Full text used.
• [25] Paço, T.A., Conceição, N. and Ferreira, M.I. (2004). Measurements and Estimates of Peach Orchard Evapotranspiration in Mediterranean Conditions. Acta Horticulturae 664:505-512. Source for the measured peach orchard evapotranspiration of 1.2 to 3.6 millimetres per day with a mean of 2.5, and for the measured crop coefficient of about 0.5 against the 0.9 of the table, that is an overestimation by FAO-56 of about 35% in a sparse canopy. Full text used. The orchard is in central Portugal.
• [26] Roussos, P.A. Irrigation of the olive. Teaching material, Agricultural University of Athens. In Greek. Source for the annual water needs of the olive at 2,000 to 4,000 cubic metres per hectare in Greece and for the crop coefficient of 0.65 to 0.75 under Mediterranean conditions. Full text used.
• [27] Greek Ministry of Rural Development and Food, Directorate of Plant Production Protection. Integrated pest management guidance for the maize crop. In Greek. Source for the total maize requirement of 500 to 800 millimetres, of which about 300 from irrigation, in five to six applications. Full text used. The document gives no soil moisture or tension threshold for starting irrigation.
• [28] Shamshiri, R.R. et al. (2018). Review of optimum temperature, humidity, and vapour pressure deficit for microclimate evaluation and control in greenhouse cultivation of tomato. International Agrophysics 32:287-302. Source for the symptom thresholds in the greenhouse, that is, below 0.3 kPa nutrient deficiencies and fungal infection, above 1.5 kPa wilting and leaf roll, and above 2.2 kPa raised risk. Full text used. Note: the range of 0.45 to 1.25 kPa that circulates widely as the greenhouse optimum is attributed by the review itself to a commercial weblog and not to a research study, and for that reason it is not used in this article.
• [29] Gourgouletis, N., Gkavrou, M. and Baltas, E. (2023). Comparison of Empirical ETo Relationships with ERA5-Land and In Situ Data in Greece. Geographies 3(3):499-521. Source for the underestimation of ERA5-Land data by about 35% against in situ measurements at three Greek sites, for the overestimation by the Valiantzas method of 32 to 62% and by Copais of 24 to 56%, and for Hargreaves-Samani being found the most efficient of the simple methods. Abstract used.
• [30] University of New England. Oz Soils, Water potential, water retention and potential. Source for the definition of pF as the base-ten logarithm of the tension in centimetres of water column, and for the correspondence of pF 2 with field capacity and pF 4.2 with the permanent wilting point. Full text used.