International Journal of Plant Production 3 (1), January 2009

ISSN: 1735-6814 (Print), 1735-8043 (Online) REVIEW ARTICLE GUASNR

This is a refereed journal and all articles are professionally screened and reviewed. www.ijpp.info

Saffron Irrigation Regime

*
A.R. Sepaskhah, A.A. Kamgar-Haghighi
Irrigation Department, Shiraz University, Shiraz, I.R. of Iran
*Corresponding author. E-mail: [email protected]

Received 23 May 2008; Accepted after revision 05 October 2008; Published online 15 January 2009

Abstract

Saffron is grown in arid and semi-arid regions in Iran in late autumn, winter and late spring with
rainy season. It should be irrigated by supplemental basin irrigation. Since rainfall is usually delayed
in autumn, therefore, a pre-flowering irrigation of about 100 mm is needed. In areas with a seasonal
rainfall of 600 mm a post- flowering irrigation of about 50 mm is adequate for economical yield. In
areas with seasonal rainfall of 400 and 200 mm continuous supplemental irrigation is needed with
intervals of 24 and 15 days or irrigation regimes of 50% ETp and 75% ETp, respectively. In these
areas, irrigation regimes can be planned based on crop water stress index (CWSI) of 0.60 and 0.27,
respectively. Monthly values of crop coefficient (Kc) and pan coefficient (Kp) for saffron are
presented with the highest values of 1.10 and 0.84, respectively, that occurred in January. For the
three-, four-and five-year old fields with higher amount of corm intensity the optimum irrigation
water is zero for about 300 mm of rainfall. While for the six-and seven-year old fields the optimum
irrigation water is zero for about 500 mm of rainfall. Flower production decreased 49% by using
irrigation water with salinity level of 1.7 dS m-1 and no flower produced at salinity level of 4.0 dS m-1,
while leaf growth occurred in this salinity. Furthermore, 50% flower yield is produced at soil water
salinity of 3.6 dS m-1.

Keywords: Irrigation interval; Irrigation scheduling; Supplemental irrigation; Rain-fed; Crop water
stress index; Crop and pan coefficients; Irrigation salinity; Optimum irrigation water.

Introduction

Traditional agriculture in Iran is based on development of cropping systems with low

water requiring crops such as saffron (Crocus sativus L.). However, not much research has
been conducted on this crop to improve technology for its production.
Saffron belongs to Iridaceae family and it is mostly distributed in Irano-Touranian
region and west of Asia with low annual rainfall, cold winters and hot summers. At present,
saffron is cultivated in Iran and a few countries with old civilization. Iran is leading country
in saffron production with 47200 ha cultivated area and 160 ton annual production (3.4 kg
ha-1 yield) (Kafi et al., 2006). The main saffron production areas in Iran are located in
Khorasan, Fars and Kerman provinces. Its cultivation area increased by an annual rate of
2 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

about 22% in last decade (Anonymous, 2002), however, its annual production increased by
about 14%. This indicates that the saffron yield (kg ha-1) decreased about 50% that may be
due to occurrence of drought and newly cultivated fields with low yield.
The three-branch style of saffron flowers is the most important economic part of the
plant. Saffron is used as a spice and a natural food color. In traditional medication, saffron
has several properties. Further, its petals are a food color and its leaves are used as animal
feed.
There are distinct differences between eco-physiological behavior of saffron and other
crops. Flowers appear before development of other plant organs. Occurrence of flowers
coincides with cold temperature in fall. In contradiction of the economic yield of most
conventional crops, saffron yield is style/stigma that is a small part of its flower. Harvest
index of saffron is less than 0.5% compared with 30 to 60% for other crops (Ingram, 1984).
Most of the conventional crops positively respond to irrigation in summer, while summer
irrigation is avoided for saffron. Therefore, saffron irrigation is considered as supplemental
irrigation. These indicate that the eco-physiological criteria of saffron are quite different
from other crops. These differences have been reviewed by Kafi (2006). However, as he
indicated, there is a great deal of controversy related to ecological, physiological and
phonological characteristics of saffron that should be examined. Traditionally, saffron is
irrigated four times during October to May, however, to achieve high yield, appropriate
irrigation scheduling should be used (Alizadeh, 2006).

Environmental requirements

Temperature

Saffron growth in temperate and dry climate is favored. However, vegetative growth of
saffron coincides with cold air temperature with freezing conditions in winter. The mean
monthly maximum, minimum and absolute minimum air temperatures in the saffron
production areas, i.e., Khorasan and Fars provinces for cold months of growing season are
shown in Table 1. The mean monthly maximum and minimum air temperatures in October
to December in southern parts of Khorasan are 20 and 0.0˚C, and for Fars province are 15.0
and -8.9˚C, respectively (Kafi, 2006). According to this reference, absolute minimum
temperature of -22˚C occurred in Torbate-Hydarieh (saffron production area) in northern
part of Khorasan province while this value is -20.0˚C for Fars province. Monthly air and
soil temperature at different depths is shown in Figure 1. These data obtained in a soil
texture similar to that in a saffron field in Bajgah (Fars province, Iran). In summer, soil
temperature at depth of 5.0 cm (about 40.0˚C) is higher than air temperature that is
measured at a height of 2.0 m and may inhibit corm physiological activity planted at this
depth in furrow irrigation. However, soil temperature at depth of 30.0 cm (30.0˚C) is lower
than air temperature that enhances the corm physiological activity planted at this depth in
basin irrigation. In winter, soil temperature in 30.0 cm depth (5.0˚C) is higher than that in
10.0 cm (2.0˚C) depth may enhance saffron corm growth planted at this depth.
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 3

Table 1. Temperature variation in cold season for the saffron growing provinces in Iran.

Temperature, ˚C Khorasan* Fars

Monthly maximum 20.0 15.0
Monthly minimum 0.0 -8.9
Absolute minimum -22.0 -20.0
* Kafi (2006)
45
40
35
Temprature, °C

30
25
20
15
10
5
0
r

r
ne

ch
y
ri l

Au y
ay

ry
be

be

be
Se g us

e
l

ar
Ju
Ap

ob

ua
Ju
M

ar
em

em

em

nu

M
ct

br
Ja
pt

ov

ec
O

fe
N

Month of year
Figure 1. Monthly air and soil temperatures at different depths in 2006 in Bajgah, Fars province: ♦, air temperature
at 2.0 m height; ■, soil temperature at 5.0 cm depth; ▲, soil temperature at 10.0 cm depth; ×, soil temperature at
30.0 cm depth

Irrigation methods
Irrigation methods and flowering

The first irrigation is usually applied at mid to end of October. The flowering is initiated
2-3 weeks after the first irrigation. Flowering period is 15-20 days and the flower harvest
period is about 10-15 days. The longer flowering period decreases the problem of labor
shortage, however, it enhances the risk of flower damage by frost. Therefore, the effect of
irrigation methods on flowering period investigated by Azizi-Zohan et al. (2005) and
results are shown in Table 2.
Table 2. Flowering period (day) as influenced by irrigation methods and intervals for two successive years.

Irrigation interval (day) Basin Furrow Mean

First year
12 16a* 16a 16A
24 17a 15a 16A
36 17a 14ab 15A
Rain-fed 11b 7c 9B
Mean 15A 13A
Second year
12 21ab 22a 21A
24 21ab 20ab 21A
36 18ab 17b 18A
Rain-fed 17b 13c 15B
Mean 19A 18A
*Means followed by the same lower case letter in each column and capital letters in each column and row are not
different significantly at 5% probability level by Duncan multiple range test.
4 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

In general, flowering period is not affected by irrigation method and irrigation interval,
however, it is significantly lower in rain-fed conditions (Table 2). Flowering initiation is
the same in different irrigation method and interval, however, it is delayed in rain-fed
conditions. Furthermore, it is indicated that for rain-fed conditions, pre-flowering irrigation
at mid to end of October results in flowering initiation similar to that occurred in irrigation
treatments. Therefore, it is concluded that for rain-fed conditions, a pre-flowering irrigation
should be applied when irrigation water is available.

Irrigation method and saffron yield

Although saffron is planted in arid and semi-arid regions in Iran and is adapted to these
conditions, however, according to the findings of Goliaris (1999) in Greece, saffron should
not be under water stress in some of the growth stages. In Greece, saffron corms grow in
March and April, and September it is the time of flower initiation, therefore, saffron should
not be under water stress in these periods. Aitoubahou and El-Otmani (1999) reported that
in Morocco, saffron is irrigated by basin irrigation. In these fields, 30-50 mm of irrigation
water is used weekly in September to November and the amount of applied irrigation water
is 35-50 mm during December to March that is applied with 2-week interval. Furthermore,
saffron is not irrigated during April to August. By this irrigation regime, saffron yield of
these fields reported to be 2-2.5 kg ha-1 and it is much lower than those reported in Italy
(10-16 kg ha-1) and Spain (10-12 kg ha-1) (Aitoubahou and El-Otmani, 1999).
Almost similar irrigation schedule is practiced in saffron plantation in Iran (Abrishami,
1997) that results in low saffron yield (3.4 kg ha-1) as reported by Kafi et al. (2006).
Therefore, it hypothesized that other irrigation methods, i.e., furrow and appropriate
irrigation interval may improve saffron yield.
Table 3. Saffron yield (kg ha-1) as influenced by different irrigation intervals and methods.

Irrigation interval Applied irrigation

Basin Furrow
(day) water (mm)
First year (437 mm rainfall)
12 610 2.59ab* 0.69c
24 386 2.71a 0.60c
36 373 1.93b 0.29c
Rain-fed 182** 0.71c 0.19c
Second year (208 mm rainfall)
12 647 5.81a 2.09d
24 364 5.19b 0.69ef
36 295 3.20c 0. 51fg
Rain-fed 0 1.09e 0.07g
*Means followed by the same letters in each year are not significantly different at 5% probability level by Duncan
multiple range test.
**It is the amount of two irrigations water as pre-flowering and post-flowering for the corm establishment.

The effects of different irrigation intervals investigated by Azizi-Zohan et al. (2006) in

basin and furrow irrigation methods with the same amount of water at each interval for
different methods. The results are depicted in Table 3. It is indicated that basin irrigation
with corm planting depth of about 25.0 cm is superior to furrow irrigation with corm
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 5

planting depth of about 10.0 cm. This may be resulted from the difference in temperature
regimes in different soil depths as shown in Figure 1. In winter, soil temperature in 30.0 cm
depth (5.0˚C) is higher than that in 10.0 cm (2.0˚C) depth may enhance saffron corm
growth planted at this depth. Further, irrigation intervals of 12 and 24-day are preferred
with seasonal rainfall of about 200 and 400 mm, respectively. Similar appropriate irrigation
interval reported by Mosaferi (2001). He indicated that saffron yield is significantly higher
with 15-day interval with up to 41% increase compared with that obtained in rain-fed
conditions in Mashhad area (northeast of Iran) an arid region with low seasonal rainfall.

Saffron pre-irrigation

For flower initiation an irrigation is required that facilitates flowering. However, timing
of this irrigation is crucial. It should be scheduled at a proper time that flowers to be appear
before the vegetative growth to be started. Otherwise, flowering and leaf growth occur
simultaneously and the later may interfere with harvesting practice. Mosaferi (2001)
reported the effects of different dates and amounts of the first irrigation on saffron yield in
Mashhad (northeast of Iran). Results are presented in Table 3. It is shown that pre-
flowering irrigation at mid October resulted in the highest saffron yield. Therefore, it is
appropriate time to apply the first irrigation that is in accordance with that of indigenous
knowledge of local farmers.
Table 4. Saffron yield as influenced by time of pre-flowering irrigation.

Time of first irrigation Flower dry weight (g m-2) Saffron yield (g m-2)
Early October 8.25 0.156
Mid October 11.02 0.226
Late October 8.47 0.170
Mid November 6.74 0.114

Irrigation methods and saffron leaf growth

Effects of irrigation method and interval on leaf dry weight of saffron in two-year old
field reported by Azizi-Zohan et al. (2006) (Table 5). Differences in leaf dry weight in
different irrigation treatments are not distinct as those reported for saffron yields (Table 3).
In basin irrigation, leaf dry weight is not statistically different in different irrigation
intervals and rain-fed conditions. This is true for different irrigation intervals in furrow
irrigation (Table 5). To study the effects of irrigation on the vegetative growth, the effects
of irrigation intervals and methods on corm growth were investigated.

Table 5. Leaf dry weight of saffron (kg ha-1) as influenced by different irrigation intervals and methods at the
second growing season.

Irrigation interval(day) Basin Furrow

12 732.9a* 555.2ab
24 702.2a 502.8ab
36 815.3a 303.8bc
Rain-fed 512.5ab 259.6c
*Means followed by the same letters are not significantly different at 5% probability level by Duncan multiple
range test.
6 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

Irrigation regimes and corm growth

Effects of irrigation intervals and methods on corm numbers and weight with different
sizes are shown in Table 6 as reported by Azizi-Zohan et al. (2006). The numbers of corms
in basin irrigation are lower than those obtained in furrow irrigation, but the numbers of
corm heavier than 8.0 g are higher in basin irrigation than those in furrow irrigation. These
differences resulted in higher flowering and saffron yield in basin irrigation (Table 3).
There are no differences between the medium weight corms (4.0-8.0 g) in basin and furrow
irrigation (Table 6), however, the numbers and yield of light corms (lighter than 4.0 g) in
furrow irrigation are higher than those in basin irrigation. Therefore, it is indicated that
furrow irrigation enhances the corm propagation and inhibited its growth. The small size of
corms is the main reason for lower saffron yield in furrow irrigation (Table 3).
Table 6. Number of corms per m-2 and yield (kg ha-1) for different corm sizes as influenced by irrigation intervals
and methods at second growing season.

Irrigation Irrigation Total corm Corm yield No. of corm Corm yield
method interval per m-2 t ha-1 per m-2 t ha-1
Total and heavier than 8 g
Basin 12 206.3 15.6 72.9 11.8
24 193.9 11.8 52.9 7.7
36 187.7 11.5 64.6 8.3
Rain-fed 181.5 4.3 17.9 1.7
Furrow 12 323.5 10.7 37.1 4.3
24 330.0 7.2 16.5 1.6
36 257.8 6.2 15.9 1.7
Rain-fed 289.4 1.8 0.0 0.0
Medium (4-8 g) and lighter than 4 g
Basin 12 41.9 2.3 94.4 1.5
24 41.3 2.2 99.7 1.8
36 31.6 1.8 91.4 1.5
Rain-fed 27.5 1.5 136.1 1.1
Furrow 12 46.8 2.7 229.6 3.7
24 30.9 1.7 282.6 3.8
36 36.4 2.0 206.3 2.5
Rain-fed 6.9 0.4 282.6 1.5

In general, the effects of irrigation intervals is not significant on corm numbers,

however, lower numbers of large and medium corm obtained in rain-fed saffron (Table 6).
Corm yield in general and heavier corms yield decreased as irrigation interval increased.
However, this is not obtained for medium and light weight corms, and their yield decreased
under rain-fed conditions.
Data differences in Tables 3 and 6 indicate that in furrow irrigation the corms are
located not deep enough in soil, therefore, they are under more water stress due to faster
depletion of soil water at the soil surface layer. Further, corms experience lower soil
temperature in winter and high soil temperature in fall and spring that may show inverse
effects on the vegetative and reproductive growth of saffron in furrow irrigation.
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 7

Irrigation Scheduling
Irrigation scheduling and saffron yield

Saffron yields as influenced by different amounts of water as percentage of potential

crop evepotranspiration (100% ETp, 75% ETp, 50% ETp, and 0% ETp, i.e., rain-fed)
applied with 24-day interval are shown in Table 7 (Monfared, 2005). In general, higher
yield is produced in basin irrigation, and irrigation scheduling based on 50% ETp is
sufficient in producing acceptable yield under seasonal rainfall of about 400 mm. However,
under seasonal rainfall of about 240 mm, irrigation scheduling based on 75% ETp is
required to produce acceptable saffron yield. It is indicated that in higher seasonal rainfall
(about 400 mm) more deficit irrigation (50% ETp) is allowed, while, in lower seasonal
rainfall (about 200 mm) lower deficit irrigation (75% ETp) is allowable.
Table 7. Saffron yield (kg ha-1) as influenced by different irrigation scheduling and methods.

Irrigation Applied irrigation

Basin Furrow
schedule water (mm)
Wet year (413 mm rainfall)
100% ETp 299 9.43a* 5.95a
75% ETp 250 9.13a 5.83a
50% ETp 202 8.82a 5.79a
Rain-fed 0 5.16b 2.03b
Dry year (244 mm rainfall)
100% ETp 307 7.26a 4.67a
75% ETp 257 7.00a 4.27a
50% ETp 207 6.25b 4.02a
Rain-fed 0 3.35c 1.68b
*Means followed by the same letters in each year and each column are not significantly different at 5% probability
level by Duncan multiple range test.

Another experiment conducted by Mosaferi (2001) in Mashhad area. He used four

levels of applied water (10, 20, 40, and 80 mm) with 15-day interval during a six months
period. Saffron yields are illustrated in Table 8. It is indicated that the amount of applied
water of 20 mm at each irrigation event with 15-day interval resulted in the highest saffron
yield. This amount of water is equivalent to about 240 mm of seasonal applied water that is
in accordance to findings of Monfared (2005) with 200 mm seasonal rainfall.

Crop water stress index of saffron

Crop water stress index (CWSI) is an appropriate parameter for irrigation scheduling for
many crops (Idso, 1982). Shirmahammadi-Aliakbarkhani et al. (2006) determined this
parameter for saffron using data obtained from an experiment based on four different levels
of irrigation water application (100% ETp, 75% ETp, 50% ETp, and 0% ETp, i.e., rain-
fed). They developed the lower and upper limit lines as follows (Idso, 1982):
(Tc-Ta)ll=12.2-0.25×VPD, R2=0.95, SE=0.31, P<0.001 (lower limit) (1)
(Tc-Ta)ul=11.0 C (upper limit) (2)
where Tc and Ta are the crop canopy and air temperatures in C, respectively, VPD is the
air vapor pressure deficit in mb, R2 is the coefficient of determination, SE is the standard
error, and p is the probability level. CWSI is calculated as follows (Idso, 1982):
8 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

CWSI= {(Tc-Ta)m-[a-b(VPD)m]}/{(Tc-Ta)ul-[a-b(VPD)m]} (3)

Table 8. Saffron yield as influenced by different amounts of applied water at each irrigation event.

Applied irrigation Flower dry Saffron yield

water (mm) weight (g m-2) (g m-2)
10 25.0b* 0.150b
20 37.0a 0.190a
40 34.0ab 0.153b
80 27.1b 0.117b
*Means followed by the same letters in each column are not significantly different at 5% probability level by
Duncan multiple range test.

where a is 12.2 C, b is 0.25 C mb-1, (VPD)m is the measured vapor pressure deficit in mb,
and (Tc-Ta)m is the measured difference between crop canopy and air temperatures in C.
Crop water stress index for saffron was calculated based on Eqs. (1), (2), and (3) for
different irrigation treatments during the growing season and their mean values are 0.04,
0.27, 0.60, and 0.92 for 100% ETp, 75% ETp, 50% ETp, and 0% ETp, respectively.
Further, the relationship between CWSI and relative evapotranspiration reported as follows
(Shirmahammadi-Aliakbarkhani et al., 2006):
CWSI=1.60(1-ETa/ETp), R2=0.99, SE=0.011, P<0.001 (4)
Where ETa and ETp are the actual and potential seasonal evapotranspirations, respectively.
According to the findings of Monfared (2005) optimum required irrigation water is
determined based on 75% ETp and 50% ETp for areas with seasonal rainfall of 200 and
400 mm, respectively, that are corresponding to the CWSI of 0.27 and 0.6, respectively.
Therefore, these values of CWSI can be used in saffron scheduling.

Table 9. Saffron yield (kg ha-1) as influenced by different supplemental irrigation with a seasonal rainfall of 600 mm.

Applied irrigation Basin Furrow

water (mm)
149 6.91a* 3.96a
144 4.92b 3.21b
137 4.03c 2.18c
Rain-fed 0.84d 0.35d
*Means followed by the same letters in each column are not significantly different at 5% probability level by
Duncan multiple range test.

Supplementary irrigation

In an experiment conducted by Shirmohammadi-Aliakbarkhani (2002), the effects of

supplementary irrigation on saffron yield, under high seasonal rainfall (600 mm) was
investigated. Pre-flowering irrigation water was applied as 125 mm and after 37 days with
no rainfall, three levels of irrigation water were applied as 100% ETp, 75% ETp, 50% ETp,
and 0% ETp. Saffron yields are shown in Table 9. Again, it is indicated that basin irrigation
is preferred to furrow irrigation. Furthermore, supplemental irrigation as pre-flowering and
post-flowering irrigation in case of no occurrence of rainfall for 37 days produced
considerable yield compared with rain-fed saffron. This is due to the fact that when rainfall
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 9

is delayed at autumn, soil surface is dry and it is not in favor of flower emergence. In this
situation, most of the emerging flowers are remained in soil, therefore, saffron yield is
reduced drastically. However, supplementary pre-flowering and post-flowering irrigation
enhanced flower emergence and saffron yield. Therefore, it is proposed that for rain-fed
saffron, when rainfall is delayed, for soil wetting at pre-flowering, irrigation water should
be used for good saffron yield to be obtained.

Saffron water requirement

Saffron growing season coincides with late autumn, winter and early spring with cold to
moderate temperatures. However, during the summer where water shortage is a limiting
factor for growth of many crops, saffron is in its dormant stage with no water requirement.
High yield of saffron depends on a strong vegetative growth that requires enough water
after flowering in late autumn to early spring. Therefore, for irrigated saffron or
supplementary irrigated saffron for rain-fed conditions, proper planning for irrigation
regimes is necessary.
Potential crop evapotranspiration for saffron is determined based on following equation:
ETp=Kc×ETo (5)
Where ETp is the potential crop evapotranspiration of saffron, Kc is the crop coefficient and
ETo is the reference crop evapotranspiration. According to Sepaskhah (1999), ETo can be
estimated based on Penman-FAO and Penman-Monteith equations using weather data from
the reference and non-reference weather stations, respectively.
Based on the pan evaporation data, ETp of saffron can be determined by the following
equation:
ETp=Kp×Ep (6)
Where Kp is the pan coefficient for saffron and Ep is the pan evaporation. In general, Kc and
Kp for saffron should be determined in field experiment.

Crop coefficient

Shirmohammadi-Aliakbarkhani (2002) determined ETp for a three-year old saffron field

in Bajgah (Fars province, I.R. of Iran) by using field water balance procedure as described
by Sepaskhah and Ilampour (1995). Three-year old saffron field produces almost the
economical yield for which Kc and Kp are determined. The values of saffron potential
evapotranspiration rate, reference crop evapotranspiration rate and class A pan evaporation
rate are presented in Table 10. These data were used to determine monthly values of crop
coefficient (Kc) and pan coefficient (Kp) for saffron. The highest value of Kc and Kp
obtained during January (1.1 and 0.84, respectively). These are higher than those reported
by Alizadeh (2006) for Khorasan province that might be occurred due to difference in
climate conditions. Furthermore, Azizi-Zohan et al. (2008) reported that the values of Kc
varied through the growing season from 0.22-0.24 to 0.94-1.05, and 0.68-0.78 at the
beginning, middle and end of the crop's cycle for one- and two-year old saffron fields.
Saffron vegetative cover increases due to corm propagation and canopy development in
successive years. Therefore, its water requirement increases with the age of field. The
values of Kc for a saffron field with age of one and two-year determined by Yarami (2008)
10 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

in a water balance lysimeter in Bajgah (Fars province, I.R. of Iran). The highest values of
Kc for the mid season of saffron were 0.93 and 1.05 for one and two-year old field,
respectively (Yarami, 2008). It is obvious that these values are lower than that for the three-
year old field as shown in Table 10.
Table 10. Reference crop evapotranspiration rate (ETO), class A pan evaporation rate (Ep), saffron potential
evepotranspiration rate (ETp), crop coefficient (Kc) and pan coefficient (Kp) for different months ofsaffron
growing season (3-year old field).

ETo Ep ETp
Month Kc Kp
mm d-1 mm d-1 mm d-1
October 2.55 3.00 1.15 0.45 0.38
November 2.27 2.67 1.24 0.55 0.46
December 1.65 1.98 1.54 0.93 0.78
January 1.66 2.17 1.83 1.10 0.84
February 2.25 3.15 2.10 0.93 0.68
March 3.51 4.80 2.39 0.68 0.50
April 5.61 6.84 2.80 0.50 0.41
May 7.25 8.12 3.01 0.42 0.37

According to Allen et al. (1998), Kc curve is usually plotted by dividing the plant
growing season into four stages and a linear change of Kc is assumed for each growth stage.
Shirmohammadi-Aliakbarkhani (2002) presented the four-stage Kc curve for a three-year
old saffron field as shown in Figure 2. Duration of the four growth stages for saffron in
Bajgah with semi-arid climate in Fars province and their corresponding values of Kc are
presented in Table 11. The presented values of Kc for different stages of growth are a little
different from those reported by Alizadeh (2006) that is due to difference in climatic
conditions. The reported values of Kc for initial, mid and late seasons are 0.4, 0.85 and
0.55, respectively (Alizadeh, 2006) in Khorasan province with arid climate.

1.2

0.8
Kc

0.6

0.4

0.2

0
0 50 100 150 200 250
Days after first irrigation

Figure 2. Crop coefficient for saffron at different growth stages.

A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 11

Evapotranspiration

Seasonal potential crop evapotranspiration of saffron (ETp) was determined by field soil
water balance (Azizi-Zohan et al., 2008) as 485 and 670 mm for one- and two-year old
fields, respectively in Bajgah area, Fars province, with a warm semi-arid climate. Higher
value of ETp in two-year old field is mainly due to the higher corm population compared
with that in one-year old field. Further, value of ETp for saffron was determined in a water
balance lysimeter in the same location (Yarami, et al., 2007) as 520 mm for one-year old
field in a different year. This value is in accordance to that reported by Aziziu-Zohan et al.
(2008). The value of ETp for a two-year old field in Khorasan province (cold arid climate)
measured by a water balance lysimeter as 320 mm (Mahdavi, 1999). The difference
between this value and that reported in Fars province is due to the difference in climate.

Table 11. Duration of growth stages of saffron and corresponding values of Kc in Bajgah (Fars Province).

Growth stage Duration (day) Crop coefficient (Kc)

Initial 30 0.46
Development 45 -
Mid-season 70 0.95
Late season 55 0.33

Irrigation water salinity

Saffron (Crocus sativus L.) is the most important export crop in Iran. It is mostly
produced in Khorasan, and Fars provinces with arid and semi-arid climates under irrigated
condition (Abrishami, 1987). However, renewable water resources with high quality in
these areas are limited. The main water source for saffron irrigation is groundwater that is
overdrawn with increasing salinity. Furthermore, because of restricted water resources,
wastewater and saline water are used in irrigation. Therefore, effects of irrigation water
salinity on saffron yield should be examined.
An experiment was conducted by Sepaskhah and Yarami (2008) to determine the effect
of water salinity on yield and yield components of saffron in pot experiment in open space
under transparent shelter in Bajgah (Fars province, I.R. of Iran). The results are illustrated
in Table 12. Response of different growth traits to salinity levels of irrigation water is not
similar. Flower fresh weight and corm and root growth decreased drastically by increasing
salinity level. Flower production decreased 49% by using salinity level of 1.7 dS m-1 and no
flower produced at salinity level of 4.0 dS m-1, while leaf growth occurred in this salinity.
Flower yield reduction is in accordance with very low corm and root growth (about 18 and
25% of control salinity for corm and root, respectively). However, by increase in salinity
level, lower reduction in leaf growth and seasonal evapotranspiration occurred. In
contradiction to flower yield, it is indicated that vegetative growth is not sensitive to water
salinity and showed a low vegetative growth reduction (18%) even at salinity level of 4.0
dS m-1. This is similar to reduction in seasonal evapotranspiration at salinity level of 4.0 dS
m-1 (21%).
12 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

Yield-salinity functions

Relationship between relative yield and seasonal mean drainage water salinity reported
by Sepaskhah and Yarami (2008) as follows:
For flower yield:
yf/yfm=1-0.142(ECss-0.08), R2=0.83 (7)
For leaf dry weight:
yl/ylm=1-0.029(ECss-0.29), R2=0.40 (8)
For corm growth:
yc/ycm=1-0.108(ECss-0.29), R2=0.54 (9)
For root dry weight:
yr/yrm=1-0.091(ECss-0.67), R2=0.64 (10)

Table 12. Effects of salinity levels of irrigation water on saffron growth and seasonal evepotranspiration in pot
experiment.

Flower fresh Leaf dry Root dry Seasonal

Salinity level Corm growth
weight Weight weight evapotranspirati
(dS m-1) -1
(g pot )
(g pot-1) (g pot-1) (g pot-1) on (mm)
0.5 2.86a* 9.46a 11.24a 9.84a 624a
1.7 1.46b 8.79b 8.65b 6.38b 543b
2.9 0.43c 8.78b 5.11c 4.34c 535b
4.0 0.00d 7.75c 2.06d 2.46d 492b
*Means followed by the same letters in each column are not significantly different at 5% probability level by
Duncan multiple range test.

where y/ym is the relative yield for different plant components, i.e., f for flower, l for leaf, c
for corm and r for root. The slopes of these equations are the yield reduction (fraction) per
unit of increase in drainage water salinity. Further, the values in parentheses are the
threshold salinity of drainage water for different yield reduction initiation. These values
show that the flower is the most sensitive part of plant with the highest and lowest values of
yield reduction (%) per unit salinity and salinity threshold, respectively. According to these
values in Eqs (7) to (10), the leaf is the least sensitive part of the plant. Further, the
sensitivity of corm and root to salinity lay in between of leaf and flower. Based on Eq. (7),
50% flower yield is produced at soil water salinity of 3.6 dS m-1.

Irrigation water productivity

Saffron yield ratio to the amount of applied irrigation water considered as water
productivity (WP). This parameter determined by using data reported by Azizi-Zohan et al.
(2006), Shirmahammadi-Aliakbarkhani (2002) and Monfared (2005) for different irrigation
methods, intervals and regimes in successive growing seasons in Bajgah (Fars province).
Results are depicted in Table 13. In general, WP is lower in furrow irrigation compared
with basin irrigation. In older fields, the reduction in WP for furrow irrigation is decreased
from about 80% to 30%. This indicates that corm growth in older furrow irrigation fields is
adapted to new conditions of plantation and resulted in improved saffron yield in older
furrow irrigation fields.
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 13

In one-and two-year old fields irrigated with different irrigation intervals, WP is the
highest with 24-day interval in basin irrigation. This indicates that 24-day irrigation interval
is appropriate for the semi-arid region in Fars province. In 4-year old field with
supplementary irrigation, in a season with higher seasonal rainfall, WP is the highest for the
irrigation regime based on 100% ETp. Furthermore, the maximum WP(about 3.0-4.6 g m-3).
Table 13. Water productivity of saffron (g m-3) as influenced by different irrigation scheduling and methods in
different years with different seasonal rainfall.

Seasonal rainfall Irrigation schedule/

Field age (year) Basin Furrow
(mm) interval (day)
1 437 12 0.43 0.11
24 0.70 0.16
36 0.52 0.08
2 208 12 0.90 0.32
24 1.43 0.19
36 1.09 0.17
4 620 100% ETp 4.64 2.66
75% ETp 3.42 2.23
50% ETp 2.94 1.59
6 413 100% ETp 3.15 1.99
75% ETp 3.65 2.33
50% ETp 4.37 2.87
7 244 100% ETp 2.37 1.52
75% ETp 2.72 1.66
50% ETp 3.02 1.94

is obtained in 4- to 6-year old saffron fields with basin irrigation. Similarly, WP in furrow
irrigation is the highest in fields with the same age. For irrigation regimes based on
different percentages of ETp, and different seasonal rainfall in 6-year and 7-year old saffron
fields, the most appropriate WP is obtained for 50% ETp and 75% ETp with seasonal
rainfall of about 400 and 200 mm, respectively.

Optimum deficit irrigation for saffron

In analyzing the water-scarcity issues, the contribution of green water (rainfall through
evapotranspiration use) is usually not included in terms of fresh water to cover crop water
requirements (Rockstrom et al., 2003). By considering the green water in this analysis
water productivity (WP) is increased. To increase WP of blue-water, deficit irrigation is
used. The optimum water use in deficit irrigation is obtained by an economic analysis using
water production and cost functions as described by English (1990) and applied for saffron
(Sepaskhah et al., 2008).
Saffron in Iran is planted as a perennial crop and its yield varies in successive years
(Sepaskhah et al., 2008). The crop production function for water is dependent on the year of
production. This is due to the fact that the amount of corm per unit area (corm intensity)
varies in successive years. Therefore, the crop production function of saffron for water is
dependent not only on the amount of applied water but also it depends on the amount of
corm per unit area. Sepaskhah et al. (2008) derived equations for determination of required
water at variable seasonal rainfall and different prices of water and corm leading to
14 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

maximum crop yield or profit with limited water conditions for saffron. The equations for
optimum irrigation water, ww, are obtained as follows:
ww=(R2-1.24R+8.52×10-4O2-0.046O+0.798)0.5 (11)
Where ww is the optimum irrigation water at water limiting condition in m, R is the seasonal
rainfall in m and O is the corm planting intensity t ha-1. Using different values of R and O
for different years of experiment, the values of ww are calculated and the results are shown
in Figure 3. The least amount of water is required for the three-year old field with the
highest amount of corm planting intensity (Figure 3). For the three-, four- and five-year old
fields with higher amount of corm planting intensity the optimum irrigation water is zero
for about 300 mm of rainfall. While for the six- , and seven-year old fields the optimum
irrigation water is zero for about 500 mm of rainfall. However, for one-, and two-year old
fields the lowest optimum irrigation water is obtained for 650 mm of rainfall (Figure 3). In
this Figure, it is shown that the optimum amounts of irrigation water are 0.25 and 0.16 m
for the one-, and two-year old fields, respectively.

Summary

Saffron is grown in arid and semi-arid regions in Iran. Its flower appears in mid autumn
and coincides with cold temperature. It shows positive response to irrigation in autumn, and
winter with rainy season, but summer irrigation should be avoided. Therefore, saffron
irrigation is supplemental irrigation especially in semi-arid region. In contradiction to most
other crops, basin irrigation should be used for saffron with a pre-flowering irrigation for
economical production. Even for rain-fed saffron with seasonal rainfall of about 600 mm,
pre-flowering irrigation is required in area that rainfall delayed until late in autumn. The
most appropriate time for pre-flowering irrigation is mid October in saffron plantation areas
in Iran. In rain-fed saffron with occasional seasonal rainfall of about 600 mm, pre-flowering
irrigation of about 150 mm is needed, while continuous supplemental irrigation is needed
for regions with seasonal rainfall of about 400 mm or lesser.
0.7
Optimum water at limited water,

0.6

0.5

0.4
Ww , m

0.3

0.2

0.1

0.0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
(a) Seasonal rainfall, m

Figure 3. Relationships between optimum water and seasonal rainfall under water limiting conditions in different
years after planting with different corm intensity: ♦, 1 year (9.5 t ha-1); ■, 2 years (11.8 t ha-1); ▲, 3 years (16.7 t
ha-1); ×, 4 years (15.8 t ha-1); ○, 5 years (16.4 t ha-1); ●, 6 years (10.4 t ha-1); ◊, 7 years (12.0 t ha-1)
A.R. Sepaskhah& A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16 15

For scheduling supplemental irrigation of saffron by irrigation intervals, the appropriate

irrigation interval is 24 and 15 days in semi-arid and arid regions, respectively.
Furthermore, irrigation regime should be planned based on 75% and 50% ETp for seasonal
rainfall of about 200 mm (arid region) and about 400 mm (semi-arid region), respectively.
This irrigation scheduling resulted in higher water productivity. Saffron irrigation
scheduling can be conducted based on crop water stress index (CWSI) of 0.27 and 0.60 in
areas with seasonal rainfall of about 200 and 400 mm, respectively.
For 3-year old saffron fields, duration of initial, development, mid season and late
season stages is 30, 45, 70, and 55 days, respectively and estimated values of crop
coefficients (Kc) for these successive stages are 0.46, 0.95 and 0.33. Furthermore, monthly
values of Kc and pan coefficient (Kp) for saffron are presented with the highest values of
1.10 and 0.84, respectively, that occurred in January.
It is shown that the optimum amounts of irrigation water are 0.25 and 0.16 m for the
one- and two-year old fields, respectively. For the three-, four-and five-year old fields with
higher amount of corm intensity the optimum irrigation water is zero for about 300 mm of
rainfall. While for the six-and seven-year old fields the optimum irrigation water is zero for
about 500 mm of rainfall.
Flower production decreased 49% by using irrigation water with salinity level of 1.7 dS
m-1 and no flower produced at salinity level of 4.0 dS m-1, while leaf growth occurred in this
salinity. Furthermore, 50% flower yield is produced at soil water salinity of 3.6 dS m-1.

Acknowledgements

This research was supported, in part, by Grant No. 87-GR-AGR-42 of Shiraz University
Research Council and the Center of Excellence for On-Farm Water Management.

References

Abrishami, M.H., 1987. Saffron Investigation. 1st Edition. Tus Publications, Mashhad, Iran. 320 pp. (In Persian).
Abrishami, M.H., 1997. Iranian Saffron: Historic, Cultural and Agronomic Prospects. Astan Ghods Razavi
Publishing Company, Mashhad, Iran. (In Persian).
Anonymous, 2002. Saffron Production in Iran. Report No. 12. Statistics and Information Center of Ministry of
Jehade Keshavarzi. (In Persian).
Ait-Oubahou, A., and El-Otmani, M., 1999. Saffron cultivation in Morocco. In: Saffron. (Neghbi, M., ed.).
Hardwood Academic Publications, Amsterdam. 154 pp.
Alizadeh, A., 2006. Irrigation. pp: 80-90. In: Kafi, M., Koocheki, A., Rashed, M.H., Nassiri, M. (eds.). Saffron
(Crocus sativus) Production and Processing. Science Publishers, Enfield.
Azizi-Zohan, A.A., Kamgar-Haghighi, A.A., Sepaskhah, A.R., 2005. Effect of irrigation method and interval on
saffron flowering duration and yield (Crocus sativus L.). 9th Conference of Iran Soil Sciences. pp: 143-144.
(In Persian).
Azizi-Zohan, A.A., Kamgar-Haghighi, A.A., Sepaskhah, A.R., 2006. Effect of irrigation method and frequency on
corm and saffron production (Crocus sativus L.). J. Sci. Tech. Agric. Natur. Resour. 10(1), 45-54. (In
Persian).
Azizi-Zohan, A., Kamgar-Haghighi, A.A., Sepaskhah, A.R., 2008. Crop and pan coefficients for saffron in a semi-
arid region of Iran. J. Arid Enviorn. 72, 270-278.
English, M.J., 1990. Deficit irrigation. I: Analytical framework. J. Irrig. Drain. Eng. 116, 399-412.
Goliaris, A.H. (1999). Saffron cultivation in Greece. In: Saffron. (Neghbi, M., ed.). Hardwood Academic
Publications, Amsterdam. 154 pp.
Idso, S.B., 1982. Non-water-stressed baselines: A key to measuring and interpreting plant water stress. Agric.
Meteorol. 27, 59-70.
16 A.R. Sepaskhah & A.A. Kamgar-Haghighi / International Journal of Plant Production (2009) 3(1): 1-16

Ingram, J.S., 1984. Saffron (Crocus sativus L.). Trop. Sci. 11, 1771-1774.
Kafi, M., 2006. Saffron ecophysioligy. pp: 39-57. In: Kafi, M., Koocheki, A., Rashed, M.H., Nassiri, M. (eds.).
Saffron (Crocus sativus) Production and Processing. Science Publishers, Enfield.
Kafi, M., Hemmati-Kakhki, A., Karbasi, A., 2006. Historical background, economy, acreage, production, yield
and uses. pp: 1-11. In: Kafi, M., Koocheki, A., Rashed, M.H., Nassiri, M. (eds.). Saffron (Crocus sativus)
Production and Processing. Science Publishers, Enfield.
Mahdavi, M., 1999. Crop coefficient and evapotranspiration of saffron under standard conditions. MSc. Thesis,
Ferdowsi University of Mashhad, Iran. (In Persian).
Monfared, M., 2005. Effect of water stress on physiological parameters of saffron (Crocus sativus L.). MSc.
Thesis. Irrigation Department, Shiraz University, Iran. (In Persian).
Mosaferi, Z.H., 2001. Effects of different irrigation regimes on saffron yield. M.Sc. Thesis, Ferdowsi University of
Mashhad, Iran. (In Persian).
Rockstorm, J., Barron, J., Fox, P., 2003. Water productivity in rain-fed agriculture: Challenges and opportunities
for smallholder farmers in drought-prone tropical agro ecosystems, In: Water Productivity in Agriculture:
Limits and Opportunities for Improvement, (Kinjie J.W. et al. Eds), International Water Management
Institute, Colombo, Sri Lanka, 2003, pp: 145-162.
Sepaskhah, A.R., Ilampour, S., 1995. Effects of soil moisture stress on evapotranspiration partitioning. Agric.
Water Manage. 28, 311-323.
Sepaskhah, A.R., 1999. A review on methods for calculating crop evepotranspiration . In: Proceeding of the 7th
National Conference on Irrigation and Evapotranspiration. Shahid Bahonar University, Kerman, Islamic
Republic of Iran. pp: 1-10. (In Persian).
Sepaskhah, A.R., Yarami, N., 2008. Interaction effects of deficit irrigation and water salinity on yield and yield
components of saffron. J. Hort. Sci. Biotech. (In Persian).
Sepaskhah, A.R., Dehbozorgi, F., Kamgar-Haghighi, A.A., 2008. Optimal irrigation water and saffron corm
planting intensity under two cultivation practices in a semi arid region. Biosys. Eng. 101:452-462.
Shirmahammadi-Aliakbarkhani, Z., 2002. Investigating the effects of irrigation methods and water deficit on leaf
area index, canopy temperature and yield of saffron (Crocus sativus L.). MSc. Thesis. Irrigation Department,
Shiraz University, Iran. (In Persian).
Shirmahammadi-Aliakbarkhani, Z., Kamgar-Haghighi, A.A., Sepaskhah, A.R., 2006. Use of crop water stress
index (CWSI) for evaluation of water status and irrigation scheduling of saffron. J. Hort. Sci. Tech. 7(1), 23-
32. (In Persian).
Yarami, N., Kamgar-Haghighi, A.A., Sepaskhah, A.R., Zand-Parsa, Sh., 2007. Determination of ETp and ETo of
saffron by water balance lysimeter. 9th Seminar on Irrigation and Evaporation Reduction. Martyr Bahonar
University of Kerman, Iran. pp: 48.
Yarami, N., 2008. Determination of saffron potential evapotranspiration and crop coefficient using water balance
lysimeter. MSc. Thesis. Irrigation Department, Shiraz University, Iran. (In Persian).