agricultural water management 88 (2007) 147–158
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/agwat
Using plant water status to define threshold values for
irrigation management of vegetable crops using soil
moisture sensors
R.B. Thompson a,*, M. Gallardo a, L.C. Valdez b, M.D. Fernández c
a
Dpto. Producción Vegetal, Universidad de Almerı́a, La Cañada, 04120 Almerı́a, Spain
Dpto. Ciencias del Agua y Medio Ambiente, Instituto Tecnológico de Sonora, 5 de Febrero 818 Sur, Cd. Obregón, Sonora, Mexico
c
Cajamar ‘‘Las Palmerillas’’ Research Station, Apdo. 250, 04080 Almerı́a, Spain
b
article info
abstract
Article history:
Thresholds of soil matric potential (SMP) and available soil water content (AWC) required to
Accepted 10 October 2006
prevent water limitations between irrigations were determined for bell pepper, melon, and
Published on line 28 November 2006
spring and winter tomato grown in Mediterranean-type greenhouses on the south-eastern
coast of Spain. Thresholds were identified by measuring the divergence of leaf water
Keywords:
potential of un-watered plants from that of well-watered plants. Soil matric potential
Irrigation scheduling
thresholds were 58 kPa for pepper, 35 kPa for melon, and 38 to 58 kPa for tomato.
Soil matric potential
In general, SMP thresholds were more negative under lower evaporative demand conditions
Available water content
such as during autumn and winter months. Available soil water content thresholds, for a
Capacitance sensor
given crop and drying cycle, differed appreciably depending on soil depth and the method
EnviroSCAN
used to calculate the values. For the four crops studied, AWC thresholds calculated at
Volumetric soil water content
0–40 cm were 13–15% higher than those calculated at 0–20 cm. Each AWC threshold for
Tomato
0–20 cm depth was 21–29% lower when AWC was based on laboratory rather than field
Pepper
determinations of field capacity and permanent wilting point. For a given method of
Melon
calculating AWC, AWC threshold values were similar for different crops and drying cycles,
suggesting limited sensitivity of the AWC approach. Using the manufacturer’s calibration,
the capacitance sensor used for SWC measurements overestimated SWC by an average of
36%. An in situ calibration provided generally good agreement with the actual SWC between
0.15 and 0.22 cm3 cm3; however, for higher SWC values, the in situ calibration underestimated SWC. The results of this study demonstrated the uncertainty of using recommended fixed AWC threshold values for irrigation management, using SWC sensors,
because of issues related to the definition of rooting depth, measurement of FC and
PWP, sensor calibration, and sensor accuracy across the relevant range of water contents.
These data suggest that SMP thresholds are much more reliable than AWC thresholds for
scheduling irrigations in greenhouse-grown vegetable crops. Technical issues regarding onfarm measurement of SMP and SWC are discussed.
# 2006 Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: +34 950 015497; fax: +34 950 015939.
E-mail address: [email protected] (R.B. Thompson).
0378-3774/$ – see front matter # 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.agwat.2006.10.007
148
1.
agricultural water management 88 (2007) 147–158
Introduction
Intensive vegetable production systems are commonly located
in warm climatic regions, such as the Mediterranean coast,
because growing conditions are favourable for numerous
vegetable species. Irrigation is required to ensure a continuously adequate water supply. These regions commonly have
limited fresh water resources, and experience increasing
competition for limited water resources from tourism and
housing development. Many fresh water resources have been
degraded by agricultural activity, through over-exploitation,
contamination with nutrients and salinisation (Gardner, 1993;
Fereres et al., 2003). Consequently, there is considerable
pressure on vegetable growers, in these regions, to optimise
their use of irrigation water because of limited availability and
to minimise adverse effects on local fresh water resources.
The greenhouse-based vegetable production system
located along the south-eastern Spanish Mediterranean coast,
consisting of more than 37,000 ha of greenhouses (Castilla and
Hernández, 2005), with 27,000 ha in the province of Almerı́a
alone (Castilla and Hernández, 2005), provides an example of
the previously-mentioned issues. Currently, 80% of cropping is
conducted in soil, and drip irrigation is used on all farms
(Pérez-Parra and Céspedes, 2001). Most irrigation water is
obtained from over-exploited local aquifers (ITGME, 1991).
Other impacts on the local aquifer system include contamination with nitrate and pesticides, a rapidly rising water level in
a superficial aquifer, salinisation and saltwater intrusion
(Pulido-Bosch et al., 1997, 2000; Pulido-Bosch, 2005). Three of
the major crops in this horticultural system are fresh market
tomato, bell pepper and melon. Tomato and pepper are transplanted in summer/early autumn and grown through autumn
and winter, with tomato sometimes continuing until early
summer. Tomato and melon are grown as spring/summer crops.
Irrigation scheduling is generally based on experience
(Fereres, 1996; Fereres et al., 2003), and it is generally accepted
that farmers and vegetable growers over-irrigate to ensure
that water is not limiting production. The scientifically based
irrigation scheduling methods most suitable for vegetable
production are the FAO method of estimating crop water
requirements (Allen et al., 1998) and soil moisture sensors.
Given the variable nature of vegetable production through
differences in farm management, crop varieties and soil type,
the FAO method can be considered as providing general
guidelines. Soil moisture sensors potentially provide the
means to irrigate in accordance with the unique characteristics of a given crop in a given field. These sensors can be used
as a ‘‘stand-alone’’ method, or their use can be combined with
the FAO method, or they can be used to complement irrigation
management based on experience.
Soil moisture sensors measure either soil water matric
potential (SMP) or volumetric soil water content (SWC). While
tensiometers are probably the most commonly used SMP
sensor on commercial farms, granular matrix sensors are
widely available and have a number of favourable technical
characteristics for on-farm use (Thompson et al., 2006). A
variety of dielectric sensors, using time domain reflectometry
(TDR) (Ferré and Topp, 2002) or capacitance (also referred to
as frequency domain reflectometry or FDR) technologies
(Starr and Paltineanu, 2002; Fares and Polyakov, 2006), are
available for on-farm measurement of SWC (SOWACS, 2002);
with multiple depth capacitance sensors (Paltineanu and
Starr, 1997; Fares and Polyakov, 2006) probably being the most
widely used for on-farm applications. Tensiometers, granular
matrix sensors and capacitance sensors can be read manually
or logged continuously, and can be used to automatically
control irrigation systems. Critical to the use of both SMP and
SWC soil moisture sensors is the threshold or lower limit
value, which indicates the degree to which soil can dry before
irrigation is required. Generally, threshold values are selected
that ensure that crops do not experience water stress or a loss
in production.
In practice, threshold SMP values are commonly recommended by Extension services, consultants or suppliers.
Recommended SMP values appear to be based on experience
and on a limited number of published scientific studies. Most
published SMP threshold values for vegetable species are from
agronomic studies comparing yields from treatments with
different threshold SMP values (e.g. Bower et al., 1975;
Smajstrla and Locascio, 1996). Most of these studies have
been conducted with open field crops, and, for a given species,
show a wide range of threshold SMP values, suggesting that
site-specific factors were influential. For fresh tomato,
reported SMP threshold values are 10 kPa in a fine sandy
soil (Smajstrla and Locascio, 1996), 20 kPa in a clay loam soil
(Bower et al., 1975), 30 kPa for loamy soils (Wang et al., 2005),
and 60 to 150 kPa (Taylor, 1965; Hanson et al., 2000a). For
pepper, reported SMP threshold values are 25 kPa (Smittle
et al., 1994; Beese et al., 1982), 45 to 65 kPa (Hedge, 1988) and
100 kPa (Tedeschi and Zerbi, 1984). For melon, reported SMP
threshold values are 35 to 40 kPa (Taylor, 1965; Hanson
et al., 2000a) and 50 to 75 kPa (Pew and Gardner, 1983).
For SWC sensors, the concept of available soil water content
(AWC) provides a practical framework for using SWC data for
irrigation management. Threshold values of AWC (AWCt) have
been estimated for many species and are listed in the relevant
FAO manuals as allowable depletion factors (100 AWCt)
(Doorenbos and Kassam, 1979; Allen et al., 1998). Recommended FAO threshold values for AWC are: 70% for bell pepper
and 60% for tomato and melon under conditions of medium
evaporative demand (Allen et al., 1998), with a reduction of
20% of AWC for conditions of low evaporative demand.
Published AWC threshold values, from agronomic studies,
for tomato, pepper and melon are generally higher than the
FAO recommendations; Hartz (1993) reported values >80% for
tomato, Jaimez et al. (1999) values of 68–80% for pepper, and
Pellitero et al. (1993) values of 85–90% for processing pepper.
Several authors (Sadras and Milroy, 1996; Sinclair et al., 1998;
Girona et al., 2002) have commented upon the uncertainty of
using fixed threshold values of AWC for irrigation scheduling
or modelling studies; because for a given species, many factors
such as soil characteristics, evaporative demand, or root
distribution can affect AWC threshold values.
Identification of threshold values of soil moisture for
irrigation management, either as SMP or AWC, using plant
water status ensures the maintenance of soil moisture
conditions that avoid physiological stress. This approach is
likely to be more sensitive than assessing differences in
production. To achieve statistically significant effects on yield
in field studies, it is likely that the stresses would need to be
agricultural water management 88 (2007) 147–158
appreciably larger than when first detected by plants.
Additionally, the threshold values obtained in agronomic
studies may be influenced by other factors influencing crop
yield and by the particular soil moisture values tested. Recent
studies examining the potential application of soil and plant
sensors for irrigation scheduling have used plant water status
to identify when crops initially experienced water stress
(Goldhamer et al., 1999; Fereres and Goldhamer, 2003; Intrigliolo
and Castel, 2004; Gallardo et al., 2006a,b). There are very few
published studies, conducted with vegetable crops, in which
threshold SMP and AWC values have been determined using
measurements of plant water status rather than agronomic
approaches. Sadras and Milroy (1996) reviewed AWC
threshold values obtained using physiological measures, but
no information was included for vegetable species.
Optimal irrigation scheduling using soil moisture sensors
requires accurate threshold values for individual crops in
given agricultural systems. The primary objective of this work
was to determine threshold values of soil matric potential
(SMPt) and available soil water content (AWCt), using leaf
water potential, for pepper, melon, spring tomato and winter
tomato crops in Mediterranean-type greenhouses. Additional
objectives were: (i) the evaluation of the accuracy of a
capacitance sensor using the manufacturer’s calibration and
using an in situ calibration developed for the experimental
soil; (ii) evaluation of the depth of soil used for AWC
calculation on AWC threshold values; (iii) evaluation of the
method of measurement of field capacity and permanent
wilting point values on AWC threshold values.
2.
Materials and methods
2.1.
Location and cropping details
The experimental work was conducted in four consecutive
crops grown in a greenhouse at the field research station ‘‘Las
Palmerillas’’, of the Cajamar Foundation, in El Ejido, Almerı́a
province, in south-eastern Spain (28430 W, 368480 N and
151 m elevation). The plastic greenhouse measured 28 m
long 22.5 m wide; it was unheated, and passively ventilated.
It was aligned with an east-west orientation. In each experiment, a vegetable crop was grown inside the greenhouse under
similar conditions to commercial vegetable production on the
south-eastern Mediterranean coast of Spain; the crops grown
were bell pepper, melon, winter tomato and spring tomato.
The soil within the greenhouse was an artificial layered
soil, as is commonly used in greenhouses in the region
(Castilla et al., 1986; Castilla and Hernández, 2005); it was
formed by placing a 20 cm layer of sandy loam soil, imported
from a quarry, over the original stony, loam soil. A 10 cm layer
of coarse river sand was placed over the imported sandy loam
soil as a mulch. The imported sandy loam soil had 1.0% organic
carbon and a pH (1:1, water) of 8.5. The original loam soil had
1.1% organic carbon and a pH (1:1, water) of 8.2. Soil bulk
density values were 1.5 and 1.6 g cm3 for, respectively, the 0–
10 and 10–20 cm depths in the imported soil, and 1.4 g cm3 for
the 20–30 cm depth in the underlying original soil. Electrical
conductivity (EC) measurements in saturated extract are
commonly 1.2–2.0 dS m1 for both soils. The dominant clay
149
mineral in each soil is illite. Drip irrigation tape was placed on
the surface of the sand mulch with 1.5 m spacing between drip
lines and 0.5 m spacing between emitters within drip-lines;
the emitters had a discharge rate of 2.6 L h1. The irrigation
water had an electrical conductivity of 0.4 dS m1. Nutrients
were applied through the irrigation system, in accordance
with local practice.
Bell pepper (Capsicum annuum L. cv. ‘Vergarsa’) was grown
from 18 July 2002 to 13 February 2003, melon (Cucumis melo L.
cv. ‘Sirio’) from 21 February to 25 June 2003, winter tomato
(Licopersicum esculetum L.; cv. Boludo) from 13 August 2003 to 16
January 2004, and spring tomato from 23 January to 6 July 2004.
Pepper plants were grown in double-rows, with 0.5 m spacing
between adjacent plants within each row, 0.4 m between rows
in each double-row, and 1.1 m between the closest rows in
adjacent double-rows, giving a plant density of 2.6 plants m2.
Drip tape was positioned midway between the two rows of
each double-row of pepper, with one dripper adjacent to one
plant in each row. Melon and tomato plants were grown in
single rows 1.5 m apart and 0.5 m spacing between plants
giving a plant population of 1.33 plants m2. A drip irrigation
emitter was located adjacent to and 8 cm from each plant.
Plants were vertically supported by nylon cord guides, and
pruned and managed following local practices.
During warmer periods, whitewash (suspension of calcium
carbonate) was applied to the greenhouse roof to reduce the
temperature inside the greenhouse. Whitewash was applied
before transplanting the winter tomato and was removed on
16 September 2003. During the spring tomato crop, the
whitewash was applied on 8 April 2004, and a second
application was made on 17 May 2004; the whitewash was
maintained until the end of the crop. Whitewash was not used
with the pepper and melon crops. The use of whitewash in
these studies was consistent with local practice.
2.2.
Irrigation treatments and experimental design
One or two drying cycles were applied to each crop. Each
drying cycle consisted of a well-watered control treatment,
irrigated with 100% of estimated crop water requirements
calculated using a mathematical model of crop evapotranspiration (ETc) developed for vegetable crops grown in
Mediterranean-type greenhouses (Orgaz et al., 2005), and an
un-watered treatment in which no irrigation was applied. All
drying cycles were applied during the fruit production phase.
The dates of the drying cycles are given in Table 1. For the
pepper and melon crops, one drying cycle was applied
(Table 1). In each of the winter and spring tomato crops,
two drying cycles were imposed (Table 1). Before and after the
drying cycles, the crops were optimally irrigated as described
for the well-watered treatment.
The greenhouse was divided into 12 experimental plots,
with 6 plots on either side of a central passage. Plots on the
southern side of the greenhouse measured 10.5 m 4.5 m, and
plots on the northern side measured 8.5 m 4.5 m. There were
three lines of irrigation drippers in each plot. In the melon and
tomato crops, there were three rows of plants per plot, and in
the pepper crop, there were three double-rows of plants per
plot. Adjacent plots were partially hydraulically separated by
vertically placing plastic sheeting to a depth of 30 cm from the
150
agricultural water management 88 (2007) 147–158
Table 1 – Dates of drying cycles for the pepper, melon, winter tomato and spring tomato crops
Crop/drying cycle
Pepper
Melon
Winter tomato–autumn drying cycle
Winter tomato–winter drying cycle
Spring tomato–spring drying cycle
Spring tomato–summer drying cycle
First day of drying cycle
21
27
29
13
23
25
Last day of drying cycle
January 2003
May 2003
September 2003
November 2003
March 2004
May 2004
No. of days of drying cycle
9 February 2003
5 June 2003
14 October 2003
12 January 2004
19 April 2004
22 June 2004
10
15
61
28
29
The first day of a drying cycle was the day after that in which the last irrigation was applied, and the final day, was the day before irrigation
was resumed.
surface of the imported soil. In each crop, only 8 of the 12 plots
were used to apply the two irrigation treatments (4 plots for the
well-watered and 4 plots for the un-watered). The experimental
design for each treatment period was a randomised block
design within four blocks (two in the northern side, and two in
the southern side of the greenhouse), with the two irrigation
treatments randomly allocated to each block.
2.3.
Measurements
In the four crops, soil water status was measured in wellwatered and un-watered plots as soil matric potential (SMP)
and volumetric soil water content (SWC). Soil matric potential
(SMP) was measured with granular matrix sensors (model
Watermark 200SS, Irrometer Co., Riverside, CA, USA) every
30 s, with the data averaged and then recorded every 30 min
with a data logger (model CR10X, Campbell Scientific International, Logan, UT, USA). The Watermark sensors were
positioned so that the mid-point of the sensor body was at
10 cm soil depth relative to the surface of the imported soil. In
the pepper crop, each Watermark sensor was 14 cm from the
emitter (perpendicular to the drip line), and 8 cm from emitter
and plant in the direction parallel to the drip line. In melon and
the two tomato crops, the corresponding distances were 6 and
8 cm. Watermark readings were initially converted to SMP
using the equations of Allen (2000). Because of the inaccuracy
of the Watermark sensor in rapidly drying soil (Thompson
et al., 2006), SMP values determined with the Watermark, were
subsequently corrected in relation to SMP measured with
tensiometers (model SKT 600/IE, Skye Instruments, Llandrindod Wells, Wales, UK) that were recorded every 5 min and
logged every 30 min using a data logger (model Datahog 2,
Skye Instruments, Llandrindod Wells, Wales, UK). A unique
relationship between SMP measured simultaneously with the
Watermark sensors and with tensiometers was derived for
each drying cycle. The tensiometers were located at the same
depth and distances from the emitters and plants as the
Watermark sensor. In each replicated plot, one Watermark
sensor and one tensiometer were symmetrically located on
either side of the same irrigation emitter and plant, which
were both aligned. This correction procedure, for Watermark
data, enabled a wider range of SMP values to be used than
would have been possible if only tensiometers had been used.
Measurements of volumetric soil water content (SWC) were
made in all crops using capacitance sensors (model EnviroSCAN, Sentek Sensor Technologies, Stepney, South Australia, Australia). Access tubes were installed in each plot before
planting the pepper crop, in the central line of plants, at the
same distance from the plant and dripper as described
previously for Watermark sensors and tensiometers. Volumetric soil water content measurements were made and
recorded every 30 min on a data logger (model RT6, Sentek
Sensor Technologies, South Australia, Australia) from the 0–
10, 10–20, 20–30 and 30–40 cm soil depths relative to the
surface of the imported soil. Volumetric soil water content
(mm) was calculated for the 0–20 and 0–40 cm soil layers.
In the two drying periods in the spring tomato, SWC was
measured with a TDR system (model Trase 6005X1, Soil
Moisture Corp., Santa Barbara, CA, USA); measurements were
made every day from Monday to Saturday at 9:00 h. Twenty
centimeter-long TDR waveguides were installed vertically in
the 0–20 cm soil layer at an equivalent position from the plant
and dripper in each replicated plot as for the other sensors
used. The manufacturer’s standard calibration was used to
convert TDR readings to SWC as previous work had confirmed
its accuracy in the same soil (Fernández et al., 2004).
The performance of the manufacturer’s default calibration
equation for the capacitance sensors was evaluated during the
spring tomato–spring drying cycle by comparing daily measurements made with capacitance sensors and the TDR. The
manufacturer’s calibration is SF = 0.1957 SWC0.404 + 0.02852,
where SF is the scaled frequency which is calculated from
readings of individual sensors made in PVC access tubes in the
field, submerged in a bath of water or in the air, as described by
Paltineanu and Starr (1997). The capacitance sensors were
then calibrated for the 0–20 cm soil layer, using SWC measured
with the TDR probes as reference values, during the spring
tomato–spring drying cycle. The derived calibration equation
was based on a linear relationship of SWC to SF values. To
assess the accuracy of the derived in situ calibration equation,
SWC data measured with the capacitance sensors, using the in
situ calibration, were compared with TDR data measured
simultaneously in the spring tomato-summer drying cycle.
The derived in situ calibration equation was used for all SWC
data obtained with the capacitance sensors for the 0–20 and
20–40 cm depths for all drying cycles. A previous calibration
study, in a nearby greenhouse with the same soils, demonstrated that the same calibration equation for the capacitance
sensor could be used for both the 0–20 and 20–40 cm soil
depths (Fernández et al., 2004).
Available soil water content (AWC) was calculated for the
0–20 cm soil depth and for the 0–40 cm soil depth from SWC
data as
AWC ¼
SWCA SWCPWP
SWCFC SWCPWP
100
151
agricultural water management 88 (2007) 147–158
where SWC is the volumetric soil water content expressed in
millimeters, and the subscripts A, FC and PWP correspond to,
respectively, the actual measured water content, field capacity
and permanent wilting point. The two soil depths of 0–20 and
0–40 cm were evaluated because of conflicting information
regarding root depth of vegetable crops grown in these soils.
González (2003) reported that 85% of root length and mass
were in the 0–20 cm soil depth, which corresponded to the
layer of imported soil; whereas Castilla et al. (1986) reported
that 60% of roots were in 0–20 cm with an appreciably smaller
but significant proportion of roots in the 20–50 cm soil depth.
Visual observations from trench studies with tomato, in the
experimental greenhouse, suggested nearly all roots were in
the 0–20 cm soil depth (R.B. Thompson, unpublished data);
however, water uptake has been observed from the 20–40 cm
depth shortly after withholding irrigation (R.B. Thompson,
unpublished data).
Available soil water content was calculated in two different
ways: (i) using FC and PWP values determined in situ (AWC-in
situ) and (ii) using values determined in the laboratory (AWClab) with a pressure plate using tensions of 0.03 and 1.5 MPa
for FC and PWP, respectively, as described by Klute (1986). The
soil used for the pressure plate determinations was air-dried,
sieved to pass through a 2 mm mesh, and packed into 5 cm
diameter and 2 cm deep rings at the same bulk density as in
the field. Field capacity was determined in situ following the
method described by Cassel and Nielsen (1986). For that, the
sand mulch was removed from an area of soil of approximately 2 m2 in the experimental greenhouse, the top 40 cm of
soil was saturated and covered with plastic sheet, and the
gravimetric soil water content was measured at 12 h intervals
for four soil depths (0–10, 10–20, 20–30 and 30–40 cm) until
constant values were obtained. The field value of PWP was
obtained from previous work in the same soil (Fernández et al.,
2005); it was the SWC when an un-irrigated pepper crop
stopped extracting soil water.
For both methods of determining AWC, SWC values for FC
and PWP were calculated from gravimetric soil water content
determinations for the corresponding soil depth, and bulk
density determined for three soil depths (0–10, 10–20 and 20–
30 cm). Available soil water content based on FC and PWP
values determined in the field (AWC-in situ) was calculated for
the 0–20 and 0–40 cm soil depths, and AWC based on
laboratory determinations methods (AWC-lab) for 0–20 cm.
The bulk density value for 20–30 cm soil depth was used for the
20–40 cm soil depth.
Threshold values of soil matric potential (SMPt) and plant
available water content (AWCt) for each drying cycle were
determined as the value at which leaf water potential (Cleaf) in
un-watered plants started to diverge from that in wellwatered control plants using linear regression analysis (Meyer
and Green, 1981). Leaf water potential was measured during
each drying cycle and for several days afterwards following
the resumption of irrigation. Measurements of Cleaf were
made, with a pressure chamber (model 3000, Soil Moisture Co.,
Santa Barbara, CA, USA), on recently fully expanded leaves
that were well exposed to sunlight. These measurements were
made every day Monday–Saturday between 12:00 and 14:00 h
on four plants, one from each replicate plot.
Climatic parameters were continuously monitored within
the greenhouse. Air temperature and relative humidity were
measured inside the greenhouse with a ventilated psychrometer located immediately above the crop (model MTH-A1, ITC,
Almerı́a, Spain). These data were measured every 30 s, and
30 min average values were recorded with a data logger (model
CR10X, Campbell Scientific Inc., UT, USA). Solar radiation was
measured with a pyranometer (model SKS 1110, Skye
Instruments, Llandrindod Wells, Wales, UK) installed at 2 m
height, measurements were recorded every 30 min with a data
logger (model H08-008-04, Hobo, Onset Computer Co., Bourne,
Maine, USA). Reference evapotranspiration (ET0) was calculated for each drying cycle from solar radiation values using
the calibration of the FAO-radiation equation developed for
the plastic greenhouses of the south-eastern Mediterranean
coast of Spain (Fernández et al., 2001; Orgaz et al., 2005).
Statistical analyses were conducted with Statgraphics Plus
Version 4.1. (Manugistic Co., Rockville, MD, USA).
3.
Results
3.1.
Climatic conditions
There were considerable differences in climatic conditions
between the different drying cycles, both between crops and
between the two drying cycles in each of the two tomato crops
(Table 2), reflecting the different periods of the year when the
crops were grown and when the drying cycles were applied.
Table 2 – Climatic data averaged for each drying cycle for the four vegetable crops studied
Climatic parameter
Average daily mean
air temperature (8C)
Average daily mean
VPD (kPa)
Average daily ET0
(mm day1)
Pepper (pepper
drying cycle)
Melon (melon
drying cycle)
12.5
22.6
Winter tomato
Spring tomato
Autumn drying
cycle
Winter drying
cycle
Spring drying
cycle
Summer drying
cycle
21.9
13.3
16.1
23.2
0.36
0.75
0.72
0.24
0.40
0.96
1.3
4.4
2.2
0.8
2.2
3.1
Data are presented of: daily mean air temperature, daily mean vapour pressure deficit (VPD, kPa) and daily reference evapotranspiration (ET0)
inside the greenhouse.
152
agricultural water management 88 (2007) 147–158
Average daily mean air temperatures were approximately 10 8C
higher and average daily mean vapour pressure deficit (VPD)
was two to four times higher during the melon drying cycle,
winter tomato–autumn drying cycle and spring tomato–
summer drying cycle than during the drying cycles applied in
winter (pepper; winter tomato–winter drying cycle). The spring
tomato–spring drying cycle had intermediate values of temperature and VPD, and the spring tomato–summer cycle the
highest values. Values of ET0 reflected the period of the year
when the drying cycles were conducted and the presence/
absence of whitewash. The highest ET0 values were observed
for the drying cycle in the melon crop due to high values of solar
radiation (data not shown) and the absence of whitewash, and
were lowest for the winter tomato–winter drying cycle.
3.2.
Evaluation and calibration of the capacitance system
The relationship between volumetric soil water content (SWC)
measured with capacitance sensors using the manufacturer’s
calibration and measured with the TDR system, for the 0–
20 cm soil depth, is presented in Fig. 1a. Using this calibration,
the capacitance sensor considerably overestimated SWC, the
average overestimation being 36%. The slope of the regression
line of 0.65 indicated that the overestimation was relatively
larger at lower soil moisture contents and smaller at higher
soil moisture contents. The subsequent in situ calibration of
the capacitance sensors, during the spring drying cycle of the
spring tomato crop, produced the linear regression equation of
SWC = 1.1807 SF 0.6972, which was statistically significant
(R2 = 0.88, RMSE = 0.008 cm3 cm3, n = 20) (Fig. 1b). The evaluation of the in situ calibration by comparison with SWC
measurements made with the TDR system, during the
summer drying cycle of the spring tomato, showed generally
good agreement with respect to absolute values and trends
(Fig. 1c). At the highest SWC values, of >0.26 cm3 cm3, the
capacitance sensor, using the in situ calibration, appreciably
underestimated SWC.
3.3.
Relationships between plant and soil water status
The evolution of Cleaf and soil matric potential (SMP) for the
well-watered and un-watered treatments, and of available soil
water content (AWC) for the un-watered treatment, calculated
for both the 0–20 and 0–40 cm soil depths, for the drying cycle
applied to the pepper crop, in winter under low evaporative
conditions, is presented in Fig. 2. Differences in Cleaf, between
well-watered and un-watered plants, are apparent in Fig. 2a
from 5 days after irrigation ceased, and were statistically
significant (P < 0.05; ANOVA) from 8 days after irrigation
ceased until its resumption (Fig. 2a). The increase in Cleaf
between 7 and 10 February occurred during overcast days with
low solar radiation (data not presented). Following the
cessation of irrigation, SMP in the un-watered treatment
decreased progressively to 115 kPa at the end of the drying
cycle (Fig. 2b). Soil matric potential in the well-watered
treatment was generally maintained between 10 and
30 kPa (Fig. 2b). Available soil water content for the unwatered treatment also decreased progressively reaching
minimum values of 57% and 74% for, respectively, the 0–20
and 0–40 cm soil depths (Fig. 2c).
Fig. 1 – (a) Relationship between volumetric soil water
content (SWC) measured with a capacitance sensor and
with a TDR system; (b) relationship between SWC
measured with TDR and scaled frequency values of the
capacitance sensor; (c) evolution of SWC values with time
as measured with a TDR and with a capacitance sensor
that had been calibrated for the experimental soil. Data
presented in (a) and (b) are from the spring drying cycle of
the spring tomato crop, data in (c) are from the summer
drying cycle of the spring tomato crop.
The linear regression analyses of SMP and AWC against
relative Cleaf (Cleaf un-watered divided by Cleaf well-watered) to
determine threshold SMP and AWC values for the pepper crop
are presented in Fig. 3. The threshold SMP at which the
reduction in relative Cleaf was initially detected, using this
approach, was 58 kPa (Fig. 3a; Table 3). Threshold AWC
values were appreciably affected by the soil depth used to
determine AWC, with values of 71% for the 0–20 cm soil depth
and 86% for the 0–40 cm soil depth (Fig. 3b; Table 3). These
threshold AWC values were calculated using field measurement of field capacity (FC) and permanent wilting point (PWP).
agricultural water management 88 (2007) 147–158
153
Fig. 3 – For the pepper crop, the relationship between
relative leaf water potential (Cleaf un-watered/Cleaf wellwatered) and (a) soil matric potential (SMP) and (b)
available soil water content (AWC) for the 0–20 cm (circles)
and 0–40 cm soil (triangles) depths. The open symbols
represent relative Cleaf values of <100%, which were used
to fit the linear regression equations. The vertical arrows
indicate the SMP or AWC threshold values where relative
Cleaf began to decline from the 100% value.
Fig. 2 – For the pepper drying cycle, evolution of (a) leaf
water potential (Cleaf) in well-watered and un-watered
plants; (b) soil matric potential (SMP) in well-watered and
un-watered treatments; (c) plant available soil water
content (AWC) for both the 0–20 and 0–40 cm soil depths in
the un-watered treatment. Arrows in (a) indicate the last
irrigation prior to the drying cycle and the irrigation that
terminated the drying cycle. Asterisks indicate days in
which there were significant differences between
treatments (P < 0.05; ANOVA). Bars are Wpooled standard
error of the means of four replicate measurements.
Using AWC calculated with laboratory-determined FC and
PWP values gave a much lower AWC threshold value for the 0–
20 cm soil depth of 48% (Table 3).
In the conditions of much higher evaporative demand
during the summer drying cycle in the spring tomato crop
(Table 2), withholding irrigation had an earlier and stronger
effect on plant and soil water status than was observed with
the autumn-winter grown pepper crop. Differences in Cleaf
between un-watered and well-watered plants were apparent
from 3 days after irrigation ceased (Fig. 4a), and were
statistically significant (P < 0.05; ANOVA) from 1 day later
until the resumption of irrigation (Fig. 4a). Once irrigation was
withheld, SMP decreased progressively in the un-watered
treatment reaching minimum values of 154 kPa (Fig. 4b). In
the well-watered treatment, SMP was generally maintained
between 10 and 30 kPa with the exception of several days
when problems with the pump of the irrigation system
prevented intended irrigations (Fig. 4b). Available soil water
content decreased rapidly during the first 10 days of the drying
cycle; thereafter, there was a slower rate of decrease until the
end of the drying cycle when the lowest values of 37% and 50%
for, respectively, the 0–20 and 0–40 cm soil depths were
measured (Fig. 4c).
The linear regression analyses of SMP and AWC against
relative Cleaf (Cleaf un-watered/Cleaf well-watered) to determine threshold SMP and AWC values for the summer drying
cycle of the spring tomato crop are presented in Fig. 5. The
threshold SMP value, at which the reduction in Cleaf of unwatered plants was detected, using this approach, was
43 kPa (Fig. 5a; Table 3). As with the drying cycle in the
pepper crop, the threshold AWC value was strongly influenced
by the depth of soil used when determining AWC, with values
of 70% for the 0–20 cm soil depth, and 84% for the 0–40 cm soil
depth (Fig. 5b; Table 3). The threshold AWC value for 0–20 cm
soil depth had a much lower value of 49% when estimated
154
agricultural water management 88 (2007) 147–158
Table 3 – Threshold values of soil matric potential (SMPt) and plant available water content (AWCt) for each crop and
drying cycle
SMPt (kPa)
AWCt-in situ, 0–20 cm (%)
AWCt-in situ, 0–40 cm (%)
Pepper
Melon
58
35
71
79
86
93
48
50
Winter tomato
Autumn drying cycle
Winter drying cycle
42
59
81
74
94
88
56
52
Spring tomato
Spring drying cycle
Summer drying cycle
39
43
73
70
86
84
51
49
AWCt-lab, 0–20 cm (%)
Threshold values correspond to when leaf water potential of un-watered plants began to diverge from that of well-watered plants using a
linear model. AWCt were determined for: (i) the 0–20 cm soil depth (AWCt-in situ, 0–20 cm); (ii) the 0–40 cm soil depth (AWCt-in situ, 0–40 cm)
using values of field capacity (FC) and permanent wilting point (PWP) determined in situ; (iii) for the 0–20 cm soil depth using FC and PWP
values determined in the laboratory.
using AWC calculated with laboratory-determined FC and
PWP values compared to field-determined values (Table 3).
Threshold values of SMP (SMPt) and of AWC (AWCt-in situ)
determined for both the 0–20 and 0–40 cm soil depths, for each
of the drying cycles conducted during the pepper, melon,
winter tomato and spring tomato crops, are presented in
Table 3. Considering all the drying cycles and crops, SMPt
values were between 35 and 59 kPa. Generally, the highest
SMPt values were obtained under relatively high evaporative
conditions, as with the melon and the spring tomato–summer
drying cycle, and the lowest SMPt values corresponded to
relatively low evaporative demand conditions, as with the
pepper and winter tomato–winter drying cycle.
There were large consistent differences between AWC
threshold values calculated for the 0–20 and 0–40 cm soil
depths, the differences being 13–15% of AWC (Table 3).
Available soil water content threshold values for the 0–20 soil
depth differed with the method used to measure FC and PWP,
AWCt-in situ calculated with FC and PWP determined in the
field was consistently much higher than AWCt-lab calculated
using laboratory-determined values. For all drying cycles and
crops, for the 0–20 cm soil depth, AWCt-in situ was between
70% and 81%, which corresponded to depletion factors of 19–
30%, while AWCt-lab was between 48% and 56%, which
corresponded to depletion factors of 44–52%. In general, each
method of calculation of AWCt gave similar values for the six
drying cycles examined (Table 3), suggesting that, unlike SMPt,
AWCt was relatively insensitive to climatic conditions
(Tables 2 and 3).
4.
Discussion
The results of this study demonstrated some important
considerations regarding the use of volumetric soil water
content (SWC) sensors for on-farm irrigation scheduling when
using the conceptual framework of available soil water
content (AWC). For the experimental soil, the manufacturer’s
calibration, for the capacitance sensor used, substantially
over-estimated SWC. Threshold AWC values were appreciably
influenced by the depth of soil selected for AWC calculation,
and by the method used to measure field capacity (FC) and
permanent wilting point (PWP). These issues have major
implications for on-farm use of SWC sensors with recommended fixed threshold values of AWC, because site-specific
in situ calibrations are inappropriate for most commercial
farmers, and on commercial farms there are likely to be
uncertainties regarding rooting depth and site-specific FC and
PWP values.
Using the manufacturer’s calibration, the capacitance
sensor appreciably overestimated SWC. There are differing
reports regarding the general applicability of a single calibration equation with this capacitance sensor. Paltineanu and
Starr (1997) developed a calibration equation that was suitable
for several soils with appreciably different textures. In several
studies conducted in the laboratory (Gardner et al., 1998;
Baumhardt et al., 2000) or in the field (Morgan et al., 1999;
Hanson and Peters, 2000b; Leib et al., 2003), capacitance
sensors appreciably overestimated SWC using the manufacturer’s calibration. A field study conducted in the same soil, as
the present study, reported that the manufacturer’s calibration over-estimated SWC by an average of 47% (Fernández
et al., 2004).
Several factors such as soil salinity, soil organic matter,
swelling 2:1 clays, bulk density and changes in soil temperature are known to affect the calibration of this capacitance
sensor (Mead et al., 1995; Paltineanu and Starr, 1997;
Baumhardt et al., 2000). We cannot fully explain why the
manufacturer’s calibration was so inaccurate in the present
study. The moderately high soil salinity levels characteristics
of vegetable crops continuously irrigated with nutrient
solution may have contributed; soil solution electrical conductivity (EC) values were generally 2.0–3.0 dS m1, suggesting
saturated extract values of approximately 1.0–1.5 dS m1
(assuming a two-times dilution). Within these soil solution
EC values, the capacitance sensor can be influenced by
changes in soil salinity (Thompson et al., 2004). It is generally
recommended that for research applications of dielectric
sensors that in situ calibrations be conducted to enhance the
accuracy of measurement (Starr and Paltineanu, 2002), and it
has been suggested that for on-farm irrigation management
applications the manufacturer’s calibration may suffice (Leib
et al., 2003). The present study, together with several other
studies, suggests that substantial measurement errors can
agricultural water management 88 (2007) 147–158
155
Fig. 5 – For the spring tomato crop–summer drying cycle,
relationship between relative leaf water potential (Cleaf unwatered/Cleaf well-watered) and (a) soil matric potential
(SMP) and (b) available water content (AWC), for the
0–20 cm (circles) and 0–40 cm soil (triangles) depths.
The open symbols represent relative Cleaf values of <100%,
which were used to fit the linear regression equations.
The vertical arrows indicate the SMP or AWC threshold
values where relative Cleaf began to decline from the 100%
value.
Fig. 4 – For the spring tomato–summer drying cycle,
evolution of: (a) leaf water potential (Cleaf) in well-watered
and un-watered plants; (b) soil matric potential (SMP) in
well-watered and un-watered treatments; (c) plant
available soil water content (AWC) for both the 0–20 and 0–
40 cm soil depths in the un-watered treatment. Arrows in
(a) indicate the last irrigation prior to the drying cycle and
the irrigation that terminated the drying cycle. Asterisks
indicate days in which there were significant differences
between treatments (P < 0.05; ANOVA). Bars are Wpooled
standard error of the means of four replicate
measurements.
occur, with this capacitance sensor, when using the manufacturer’s calibration. Should such measurement errors occur
on a commercial farm, they could have major implications for
irrigation scheduling when using recommended threshold
values of AWC or SWC to initiate irrigation.
Once the capacitance sensor was calibrated for the
experimental soil, measurement of SWC was generally
accurate between 0.15 and 0.22 cm3 cm3; however, for higher
SWC values between 0.26 and 0.30 cm3 cm3, SWC was
markedly underestimated, suggesting a reduced response to
more moist soil conditions. Similar observations were
reported by Tomer and Anderson (1995) and Moreno (2003).
Irrigation management using SMP sensors is based on the
use of recommended threshold SMP values to indicate when
irrigation is required. The SMP threshold values determined in
the present study were 35 kPa for melon, 58 kPa for pepper,
and 39 to 59 kPa for tomato. A relatively high value is
expected for melon because it is grown when evaporative
demand is relatively high. The value for melon, from the
present study, is consistent with the 35 to 40 kPa range
recommended by Taylor (1965), and subsequently by Hanson
et al. (2000a). The SMP threshold values for tomato obtained in
the present study are higher than the 60 to 150 kPa range
recommended by Taylor (1965), and subsequently by Hanson
et al. (2000a), for field-grown tomato, but slightly lower than
the values of 20 to 30 kPa reported for agronomic studies
conducted in medium-textured soils in open field crops
(Bower et al., 1975; Wang et al., 2005). Differences between
SMP threshold values of the present and other published
156
agricultural water management 88 (2007) 147–158
studies can be attributed to site-specific factors such as the
shallow rooting depth in the present work (González, 2003),
the lower evaporative demand of greenhouse compared to
field-grown crops (Montero et al., 1985), and the higher
sensitivity of leaf water potential compared to agronomic
differences to detect differences. The threshold SMP value for
pepper, in the present study, is within the range of values
reported by Beese et al. (1982), Tedeschi and Zerbi (1984),
Hedge (1988) and Smittle et al. (1994).
Threshold SMP for a given species is influenced by growth
stage and evaporative conditions (Hanson et al., 2000a). In the
present work, the threshold SMP values are from the fruit
production phase. The fresh market vegetable crops grown in
the present study, e.g. fresh market tomato, are indeterminate
plants with a very short vegetative phase (of several weeks
after cotyledon expansion) and a prolonged phenological stage
of simultaneous vegetative growth, fruit formation and fruit
growth that continues for much of the period of crop growth
(Kinet and Peet, 1997). In the present study, an effect of
evaporative demand upon threshold SMP values was
observed. For tomato, threshold SMP was approximately
60 kPa under very low evaporative conditions (ET0 of
0.8 mm day1) and was 40 kPa under appreciably higher
evaporative conditions (ET0 around 2–3 mm day1) (Tables 1
and 2). The effect of climate on SMP is well established, with
higher SMP threshold values being recommended for higher
evaporative conditions (Hanson et al., 2000a). Soil texture can
also influence threshold SMP values (Hanson et al., 2000a); in
lighter textured soils, SMP threshold values would be somewhat higher than reported here.
The threshold AWC values determined depended on the
method used to calculate AWC. The soil depth used to
calculate AWC, and the method used to determine FC and
PWP, had large effects on threshold AWC values. Considering
all of the drying cycles examined, the threshold AWC
calculated for the 0–40 cm soil depth was 13–15% higher than
for the 0–20 cm soil depth. In many cropping situations, there
is likely to be uncertainty about the depth of roots in the soil
profile. Additionally, there will be uncertainty as to how to
define the rooting depth, e.g. ‘‘what proportion of root mass or
length in a given increment of soil depth is required for that
increment to be included in the calculation of AWC?’’.
Working with peach trees, Girona et al. (2002) demonstrated
the effect of using different soil depths on the calculation of
threshold AWC; when they used the maximum rooting depth,
they obtained an unrealistic AWC threshold value of 94%.
Similarly, in the present work, using an AWC value calculated
using the maximum rooting depth of 40 cm gave threshold
AWC values that are very high compared to published values
for herbaceous crops (Sadras and Milroy, 1996).
In the present study, threshold AWC values estimated
using AWC calculated from laboratory-determined FC and
PWP values were much lower than when using FC and PWP
values obtained in undisturbed field soil. Such differences in
AWC values, for a given soil, will clearly affect the application
of recommended threshold AWC values. The occurrence of
appreciable differences between laboratory and field determined values of FC and PWP, and consequently of AWC is well
established (Sadras and Milroy, 1996; Girona et al., 2002). The
method used to obtain AWC is an important consideration
when the conceptual framework of AWC is used to apply SWC
data to irrigation management. Considering that determination of FC and PWP in the laboratory often involves
considerable disturbance of soil, field determinations of FC
and PWP appear to be more representative of cropping
conditions.
Threshold AWC values, based on AWC calculated for the 0–
20 cm soil depth and using field-determined FC and PWP
values, were 70–81%, which correspond to depletion factors of
19–30% of AWC. The threshold AWC values determined in the
present study are generally higher than the general recommendations for vegetable crops in the FAO manuals (Doorenbos and Kassam, 1979; Allen et al., 1998). For example, in the
FAO manuals, recommended AWC threshold values for
tomato grown in winter with ETc rates of 1–1.5 mm day1
are 45%, and approximately 50% in late spring with ETc of
3 mm day1 (Allen et al., 1998). The difference between the
threshold values obtained in the present study and the FAO
recommendations may be attributable to the general nature of
the FAO recommendations, and that the methodology used to
identify thresholds in the present study, of using plant water
status, is presumably more sensitive than the agronomic
approach behind the FAO recommendations.
In the current study, differences in SMP threshold values
were observed with different evaporative conditions for the
same species; however, such differences were not observed
with the AWC threshold values. The lack of such an effect is
inconsistent with the FAO manuals (Doorenbos and Kassam,
1979; Allen et al., 1998). The reason may be the characteristics
of the soil retention curve of the soil used, in which for the
range of soil moisture examined, relative changes in SMP were
larger than the corresponding relative changes in SWC,
suggesting that SWC using AWC was a less sensitive measure
of soil water status. The apparently reduced sensitivity of the
capacitance sensor, in this soil, at higher soil moisture
contents, may have been a contributing factor. Additionally,
the larger volume of soil with a heterogeneous distribution of
soil water measured by the capacitance sensor may contribute
to an apparent reduced sensitivity compared to SMP sensors,
which are making ‘‘point’’ measurements.
The current work was conducted with drip irrigation,
which is commonly used with intensive vegetable production,
particularly where water resources are limited. The location of
soil moisture sensors in relation to both the dripper and plant
will influence the actual SMP or SWC values measured, which
has implications for the general applicability of threshold
values from individual studies such as this and of recommended fixed values such as those in the FAO manuals and
Hanson et al. (2000a).
In summary, the work reported in this paper has
demonstrated uncertainties regarding the use of recommended fixed AWC threshold values for irrigation management on account of the effect on measured AWC values of: (i)
the depth of soil used to calculate AWC; (ii) the method used to
measure FC and PWP. Additionally, there can be appreciable
errors associated with measurement of SWC using soil
moisture sensors; a requirement for site-specific calibration
would considerably increase the complexity of using SWC
sensors for irrigation management in commercial farms. An
alternative approach may be to use information on soil water
agricultural water management 88 (2007) 147–158
dynamics, derived from continuously measured SWC data, to
determine in situ SWC threshold values for individual crops
and fields. Such an approach would avoid the uncertainties
associated with applying recommended fixed threshold
values that have been obtained elsewhere, and with sensor
calibration. Currently, there are no published studies that
assess such an approach. The use of SMP threshold values for
irrigation management appeared to be more straightforward;
however, there are issues relating to sensor performance that
affect the practical application of these sensors. The limited
working range of tensiometers (approximately 0 to 80 kPa)
can be a limitation for crops in rapidly drying soil, and
growers tend to dislike them for their preparation and
maintenance requirements. Granular matrix sensors overcome the mentioned shortcomings of tensiometers; however,
their calibration is an important issue (Thompson et al., 2006),
and they appear to respond slowly in rapidly drying soil
(Thompson et al., 2006). As a general observation, the current
state of the use of soil moisture sensors for irrigation
management, appears to be that SWC sensors have favourable technical characteristics but that there are major issues
associated with applying them to on-farm use using the
approach of recommended threshold AWC values, and that
SMP sensors can be readily used with recommended threshold values, but that there are technical issues that influence
their suitability for on-farm use.
Acknowledgements
This work was part of project AGL2001-2068 funded by the
Spanish Ministry of Science and Technology and FEDER. We
thank the Cajamar ‘‘Las Palmerillas’’ research station for the
provision of the facilities to undertake this work and for
assistance throughout this study. We would particularly like
to thank Francisco Bretones of the same research station, for
his invaluable assistance and ready availability to help with
electrical matters. The helpful and constructive comments of
the two anonymous referees are acknowledged.
references
Allen, R., 2000. Calibration for the Watermark 200SS soil water
potential sensor to fit the 7-19.96 ‘‘calibration #3’’ table from
Irrometer, University of Idaho, Kimberley, http://
www.kimberly.uidaho.edu/water/swm/
calibration_Watermark2.html, confirmed June 19, 2006.
Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop
evapotranspiration, Guidelines for computing crop water
requirements, FAO Irrigation and Drainage Paper 56, FAO,
Rome, Italy.
Baumhardt, R.L., Lascano, R.J., Evett, S.R., 2000. Soil material,
temperature, and salinity effects on calibration of
multisensor capacitance probes. Soil Sci. Soc. Am. J. 64,
1940–1946.
Beese, F., Horton, R., Wierenga, P.J., 1982. Growth and yield
response of chile pepper to trickle irrigation. Agron. J. 74,
556–561.
Bower, C.A., Kratky, B.A., Ikeda, N., 1975. Growth of tomato on a
tropical soil under plastic cover as influenced by irrigation
practice and soil salinity. J. Am. Soc. Hort. Sci. 100, 519–521.
157
Cassel, D.K., Nielsen, D.R., 1986. Field capacity and available
water capacity. In: Klute, A. (Ed.), Methods of Soil Analysis.
Part 1. Physical and Mineralogical Methods. 2nd ed.
American Society of Agronomy, Soil Science Society of
America, Madison, WI, USA, pp. 901–926.
Castilla, N., Giménez, C., Fereres, E., 1986. Tomato root
development on sand mulch, plastic greenhouse in Almeria
(Spain). Acta Hort. 191, 113–121.
Castilla, n., Hernández, J., 2005. The plastic greenhouse industry
of Spain. Chron. Hort. 45 (3), 15–20.
Doorenbos, J., Kassam, A.H., 1979. Yield response to water, FAO
Irrigation and Drainage Paper 33, FAO, Rome, Italy.
Fares, A., Polyakov, V., 2006. Advances in crop water
management using capacitance sensors. Adv. Agron. 90,
43–77.
Fereres, E., 1996. Irrigation scheduling and its impact on the 21st
century. In: Camp, C.R., Sadler, E.J., Yoder, R.E. (Eds.),
Evapotranspiration and Irrigation Scheduling.
Proceedings of the International Conference on American
Society of Agricultural Engineers, San Antonio, TX, USA, pp.
547–553.
Fereres, E., Goldhamer, D.A., 2003. Suitability of stem
diameter variations and water potential as indicators for
irrigation scheduling of almond trees. J. Hortic. Sci.
Biotechnol. 78, 139–144.
Fereres, E., Goldhamer, D.A., Parsons, L.R., 2003. Irrigation
water management of horticultural crops. Hortscience 38,
1036–1042.
Fernández, M.D., Orgaz, F., Fereres, E., López, J.C., Céspedes, A.,
Pérez, J., Bonachela, S., Gallardo, M., 2001. Programación del
Riego de Cultivos Hortı́colas bajo Invernadero en el
Sudeste Español. Cajamar (Caja Rural Intermediterránea)
Almerı́a, Spain.
Fernández, M.D., Bonachela, S., Contreras, A., González, A.M.,
2004. Evaluación de un equipo de medida continua de
humedad basado en la reflectometrı́a en el dominio de
frecuencias en un suelo enarenado. In: Actas del XXII.
Congreso Nacional de Riegos, La Rioja, Spain, pp. 61–64.
Fernández, M.D., Gallardo, M., Bonachela, S., Orgaz, F.,
Thompson, R.B., Fereres, E., 2005. Water use and production
of a greenhouse pepper crop under optimum and limited
water supply. J. Hortic. Sci. Biotechnol. 80, 87–96.
Ferré, P.A., Topp, G.C., 2002. Time domain reflectometry. In:
Dane, J.H., Topp, G.C. (Eds.), Methods of Soil Analysis. Part
4. Physical Methods. Soil Science Society of America Inc.,
Madison, WI, USA, pp. 434–446.
Gallardo, M., Thompson, R.B., Valdez, L.C., Fernández, M.D.,
2006a. Use of stem diameter variations to detect plant water
stress in tomato. Irrig. Sci. 24, 241–255.
Gallardo, M., Thompson, R.B., Valdez, L.C., Fernández, M.D.,
2006b. Response of stem diameter variations to water stress
in greenhouse-grown vegetable crops. J. Hortic. Sci.
Biotechnol. 81, 483–495.
Gardner, C.M.K., Dean, T.J., Cooper, J.D., 1998. Soil water content
measurement with a high-frequency capacitance sensor. J.
Agric. Eng. Res. 71, 395–403.
Gardner, W.R., 1993. The future of irrigated areas. In: Buxton,
D.R., Shibles, R., Forsberg, R.A., Blad, B.L., Asay, K.H.,
Paulsen, G.M., Wilson, R.F. (Eds.), International Crop
Sciences, vol. 1. Crop Sci. Soc. Am. Inc, Madison, WI, USA,
pp. 97–99.
Girona, J., Mata, M., Fereres, E., Goldhamer, D.A., Cohen, M.,
2002. Evapotranspiration and soil water dynamics of
peach trees under water deficits. Agric. Water Manage. 54,
107–122.
Goldhamer, D.A., Fereres, E., Mata, M., Girona, J., Cohen, M.,
1999. Sensitivity of continuous and discrete plant and soil
water status monitoring in peach trees subjected to deficit
irrigation. J. Am. Soc. Hort. Sci. 124, 437–444.
158
agricultural water management 88 (2007) 147–158
González, A.M., 2003. Programas de riego para cultivos
hortı́colas en invernaderos enarenados en Almerı́a, PhD
Thesis, Universidad de Almerı́a, Almerı́a, Spain.
Hanson, B.R., Orloff, S., Peters, D., 2000a. Monitoring soil
moisture helps refine irrigation management. Calif. Agric.
54 (3), 38–42.
Hanson, B.R., Peters, D., 2000b. Soil type affects accuracy of
dielectric moisture sensors. Calif. Agric. 54 (3), 43–47.
Hartz, T.K., 1993. Drip-irrigation scheduling for fresh-market
tomato production. Hortscience 28, 35–37.
Hedge, D.M., 1988. Irrigation and nitrogen requirements of bell
pepper (Capsicum annuum L.). Indian J. Agric. Sci. 58, 668–672.
Intrigliolo, D.S., Castel, J.R., 2004. Continuous measurement of
plant and soil water status for irrigation scheduling in
plum. Irrig. Sci. 23, 93–102.
ITGME, 1991. Control de la explotación del Campo de Dalı́as
(Almerı́a) 1989–1990, Ministerio de Industria y Energı́a,
Spain.
Jaimez, R.E., Rada, F., Garcı́a-Nuñez, C., 1999. The effect of
irrigation frequency on water and carbon relations in three
cultivars of sweet pepper (Capsicum chinense Jacq), in a
tropical semiarid region. Sci. Hortic. 301–308.
Kinet, J.M., Peet, M.M., 1997. Tomato. In: Wien, H.C. (Ed.), The
Physiology of Vegetable Crops. CAB International,
Wallingford, UK, pp. 207–258.
Klute, A., 1986. Water retention: laboratory methods. In: Klute,
A. (Ed.), Methods of Soil Analysis. Part 1. Physical and
Mineralogical Methods. 2nd ed. American Society of
Agronomy, Soil Science Society of America, Madison, WI,
USA, pp. 635–662.
Leib, B.G., Jabro, J.D., Matthews, G.R., 2003. Field evaluation and
performance comparison of soil moisture sensors. Soil Sci.
168, 396–408.
Mead, R.M., Ayars, J.E., Liu, J., 1995. Evaluating the influence of soil
texture, bulk density and soil water salinity on a capacitance
probe calibration, ASAE Paper No. 95-3264, American Society
of Agricultural Engineers, St. Joseph, MI, USA.
Meyer, W.S., Green, G.C., 1981. Plant indicators of wheat and
soybean crop water stress. Irrig. Sci. 2, 167–176.
Montero, J.I., Castilla, N., Gutiérrez-Rave, E., Bretones, F., 1985.
Climate under plastic in the Almeria area. Acta Hortic. 170,
227–234.
Moreno, M.F., 2003. Contribuciones a las medidas del contenido
de agua del suelo y a la evapotranspiración de los cultivos,
PhD Universidad de Córdoba, Córdoba, Spain.
Morgan, K.T., Parson, L.R., Wheaton, T.A., Pitts, D.J., Obreza,
T.A., 1999. Field calibration of a capacitance water content
probe in fine sand soils. Soil Sci. Soc. Am. J. 63, 987–989.
Orgaz, F., Fernández, M.D., Bonachela, S., Gallardo, M., Fereres,
E., 2005. Evapotranspiration of horticultural crops in
an unheated plastic greenhouse. Agric. Water Manage. 72,
81–96.
Paltineanu, I.C., Starr, J.L., 1997. Real-time soil water dynamics
using multisensor capacitance probes: laboratory
calibration. Soil Sci. Soc. Am. J. 61, 1576–1585.
Pellitero, M., Pardo, A., Simón, A., Suso, M.L., Cerrolaza, A., 1993.
Effect of irrigation regimes on yield and fruit composition of
processing pepper (Capsicum annuum L.). Acta Hortic. 335,
257–263.
Pérez-Parra, J.J., Céspedes, A., 2001. Análisis de la demanda de
inputs para la producción en el sector de cultivos protegidos
de Almerı́a. In: Cuadrado, I.M. (Ed.), Estudio de la Demanda
de Inputs Auxiliares: Producción y Manipulación en el
Sistema Productivo Hortı́cola Almeriense. FIAPA, Almerı́a,
Spain, pp. 1–102.
Pew, W.D., Gardner, B.R., 1983. Effects of irrigation practices on
vine growth, yield, and quality of muskmelon. J. Am. Soc.
Hort. Sci. 108, 134–137.
Pulido-Bosch, A., 2005. Recarga en la Sierra de Gádor e
Hidrogeoquı́mica de los aquı́feros del Campo de Dalı́as.
Escobar Impresores S.L. El Ejido, Almerı́a, Spain.
Pulido-Bosch, A., Navarrete, F., Martı́nez, J.F., Molina, L.,
Sánchez, F., Vallejos, A., Martı́n, W., 1997. La contaminación
en los acuı́feros del Campo de Dalı́as y Delta del Andarax
(Almerı́a). In: Recursos Naturales y Medio Ambiente en el
Sureste Peninsular, Instituto de Estudios Almerienses,
Almerı́a, Spain, pp. 363–381.
Pulido-Bosch, A., Bensi, S., Molina, L., Vallejos, A., Calaforra,
J.M., Pulido-Leboeuf, P., 2000. Nitrates as indicator of aquifer
inter-connection Application to the Campo de Dalı́as (SESpain). Environ. Geol. 39, 791–799.
Sadras, V.O., Milroy, S.P., 1996. Soil-water thresholds for the
response of leaf expansion and gas exchange: a review.
Field Crops Res. 47, 253–266.
Sinclair, T.R., Hammond, L.C., Harrison, J., 1998. Extractable soil
water and transpiration rate of soybean on sandy soils.
Agron. J. 90, 363–368.
Smajstrla, A.G., Locascio, S.J., 1996. Automated drip irrigation
scheduling of tomato using tensiometers. In: Camp, C.R.,
Sadler, E.J., Yoder, R.E. (Eds.), Evapotranspiration and
Irrigation Scheduling. Proceedings of the International
Conference on American Society of Agricultural Engineers,
San Antonio, TX, USA, pp. 845–850.
Smittle, D.A., Dickens, W.L., Stansell, J.R., 1994. Irrigation
regimes affects yield and water use by bell pepper. J. Am.
Soc. Hort. Sci. 119, 936–939.
SOWACS, 2002. http://www.sowacs.com, Soil water content
sensors and measurement, confirmed June 21, 2006.
Starr, J.L., Paltineanu, I.C., 2002. Capacitance devices. In: Dane,
J.H., Topp, G.C. (Eds.), Methods of Soil Analysis. Part 4.
Physical Methods. Soil Science Society of America, Inc.,
Madison, WI, USA, pp. 463–474.
Taylor, S.A., 1965. Managing irrigation water on farm. Trans.
Am. Soc. Agri. Eng. 8, 433–436.
Tedeschi, P., Zerbi, G., 1984. Flowering and fruiting courses and
yield of sweet peppers (Capsicum annuum L.) plants grown in
lysimeters with relation to different water regimes. Rivista
di Ortoflorofrutticoltura Italiana 68, 323–329.
Thompson, R.B., Gallardo, M., Fernández, M.D., 2004. Irrigation
scheduling of drip-irrigated vegetable crops grown in
greenhouses using continuous soil moisture monitoring.
Acta Hortic. 664, 653–660.
Thompson, R.B., Gallardo, M., Agüera, T., Valdez, L.C.,
Fernández, M.D., 2006. Evaluation of the watermark sensor
for use with drip irrigated vegetable crops. Irrig. Sci. 24, 185–
202.
Tomer, M.D., Anderson, J.L., 1995. Field evaluation of a soil
water-capacitance probe in a fine sand. Soil Sci. 159, 90–97.
Wang, Q., Klassen, W., Li, Y., Codallo, M., Abdul-Baki, A.A., 2005.
Influence of cover crops and irrigation rates on tomato
yields and quality in a subtropical region. Hortscience 40,
2125–2131.