2.1 Site description
This lysimeter study was conducted during the four consecutive growing seasons, 2013 to 2017 (2013–2014, 2014–2015, 2015–2016 and 2016–2017) at the Experimental Research Station of the Agricultural College, Shiraz University, Iran (29° 43' 44.0" N, 52° 35' 10.9" E, 1810 MSL) with the same experimental layout design in all the four growing seasons. The experimental site was located in a semi-arid region in southwest of Iran with a long-term average annual precipitation, relative humidity and air temperature of 387 mm, 52.2% and 13.4°C, respectively. The mean monthly climatic data for the years of experiment are presented in Tables S1 and S2. Rainfall events were mostly occurred during November to May over the years of study as 279, 233, 289 and 368 mm for the first, second, third and fourth year, respectively. Higher precipitation depths took place in November and January of 2013–2014, November and March of 2014–2015, November, December and January of 2015–2016 and February and March of 2016–2017. The physico-chemical properties of the soil of lysimeters are presented in Table S3. The whole soil profile depth (Fine, mixed, mesic, Typic Calcixerepts) was classified as clay loam. The chemical analysis for the irrigation water is illustrated in Table S4 where there were no sodium and salinity hazards in the irrigation water.
2.2 Lysimeters’ description
This experiment was conducted in 48 in-field water balance GRP lysimeters (100-cm inner diameter and 110 cm length) [Farasan Manufacturing & Industrial Company, Iran, Fars Province, Shiraz] (Fig. 1a). The bottom of each lysimeters was blocked and water sealed slopping toward a drain pipe connected to a10-litter drainage container through a flexible drain tube (Fig. 1b). A layer of 0.1 m gravel as filter was placed at the bottom of each lysimeter and 0.90 m-thick soil layer was put on top. Two 15 m×2 m guard plots were constructed along both sides of the set of lysimeters and winter wheat was planted inside to reduce the adjacent environmental influences. A number of 24 PVC micro-lysimeters (Fig. 1c) [105.6 mm inner diameter and 250-mm long] were installed into the soil of half the lysimeters and filled with surrounding soil and left exposed to environmental conditions to estimate evaporation from soil. To determine soil volumetric water content (SWC), 2-inch-diameter aluminum access tubes (1.0 m long) were installed at the center of half of the lysimeters (24 lysimeters) [Fig. 1b] and (SWC) was measured by a neutron scattering apparatus using a CPN 503DR hydroprobe (CPN Corp., Santa Barbara, CA).
2.3 Experimental design
The experiment was conducted in a \(2 \times 2 \times 4\) factorial arrangement of the treatments with three replications in a randomized complete block experimental design (RCBD) over four years comprised two systems of cropping (saffron mono-cropping and saffron intercropped with winter wheat), two different sources of nitrogen (fermented cow manure and chemical urea source of nitrogen) and four irrigation regimes [40, 60, 80, and 100% of the standard crop evapotranspiration (\({\text{E}\text{T}}_{\text{c}}\))].
2.4 Agronomic practices and measurements
In all the growing seasons, in late August, the top soil of each lysimeter was deeply plowed and triple superphosphate (100 kg P ha− 1) was mixed with the soil and leveled. For the manure treatment, the top 0.15 m\(\)soil layer was supplied by 30 Mg ha− 1 fermented cow manure as a source of nitrogen. The chemical characteristics of the fermented cow manure are presented in Table S5. On August 26 of the first growing season, semi big-sized saffron corms (> 8 g) were planted at the soil depth of 15–20 cm in three 30-cm- apart rows with a density of \({15 \text{M}\text{g} \text{h}\text{a}}^{-1}\) in each lysimeter. In each growing season, on October 29 to 30, winter wheat (Shiraz cultivar) seeds [250 kg ha− 1 (Sepaskhah & Hosseini, 2008)] were planted in the intercropping treatments at 5 cm soil depth in three parallel rows each located in the middle of the bare soil space between the rows of saffron corms (Fig. S1). All procedures were conducted in accordance to the relevant guidelines and regulations. Chemical urea treatments were supplied with granular urea fertilizer (120 kg N ha− 1) half of which was applied just prior to the first irrigation immediately after the sowing of winter wheat at late October and the remaining split was applied at March (approximately 120–140 days after first irrigation). Over every November of all the growing seasons, during the flowering time, fresh flowers of all the treatments were manually picked up early every morning and the three-part stigmas and styles were separated from the stamens and petals. For each replication of any treatments, all the collected stigmas and styles were shadow-air dried in room condition for about 2–3 weeks and weight was measured precisely as saffron yield. At the end of each growing season, saffron leaves, wheat grain and straw were completely harvested from each lysimeter, oven dried, weighed and divided by its harvested area to determine saffron above-ground biomass, grain and straw yields, respectively. At the end of last (fourth) growing season, saffron corms were uprooted and corm yield was measured.
2.5 Irrigation
Prior to each irrigation event, the volumetric soil water content (θi) was measured at 0.3, 0.6 and 0.75 m of soil depths with a neutron scattering apparatus. Saffron gross water requirement [Eq. (1)] was calculated based on increasing soil water content to the field capacity (one of the applied methods in this study) considering irrigation application efficiency of 60% (common irrigation efficiency applied by local farmers) (Abbasi & Sepaskhah, 2022):
$${d}_{g}=\frac{\sum _{i=1}^{n}\left({\theta }_{FCi}-{\theta }_{i}\right)\times {\varDelta z}_{i}}{{E}_{a}}$$
1
where dg is the gross irrigation water depth [m] for irrigation regime of 100% ETc, Δzi is the soil thickness at layer i of saffron rooting depth [m], n is the number of soil layers in saffron rooting depth (RD), θFCi and θi are the volumetric soil water contents of layer i at field capacity and before irrigation [m3 m− 3], respectively and Ea is irrigation application efficiency [decimal]. For the irrigation regimes of 80, 60 and 40% ETc, 80, 60 and 40% of the amount calculated in Eq. (1) was applied, respectively. For each growing season, saffron root depth was estimated using Eq. (2) suggested by Borg & Grimes, (1986):
$$\text{R}\text{D}= {\text{R}\text{D}}_{\text{m}\text{i}\text{n}}+{\text{R}\text{D}}_{\text{m}\text{a}\text{x}} [0.5+0.5\text{sin}(3.03\frac{\text{D}\text{A}\text{F}\text{I}}{\text{D}\text{T}\text{M}}- 1.47)]$$
2
Where \(\text{R}\text{D}\) is the saffron rooting depth [m], \({\text{R}\text{D}}_{\text{m}\text{i}\text{n}}\) is the planting depth of saffron corms [m], \({\text{R}\text{D}}_{\text{m}\text{a}\text{x}}\) is maximum rooting depth of saffron [0.45 m for saffron (Sepaskhah et al., 2013)], \(\text{D}\text{A}\text{F}\text{I}\) is number of days after first irrigation which was reset for each growing season, \(\text{D}\text{T}\text{M}\) is the number of days after first irrigation that root reaches its maximum depth [173 days for saffron (Shirmohammadi & AliakbarKhani, 2002)]. However, for each growing season, to calculate saffron gross water requirement for the first irrigation event, a soil depth of 40 cm was considered instead of RD. The first irrigation of all the lysimeters was performed immediately after sowing of the winter wheat seeds in late October. Irrigation of intercropping treatments was carried out based on the saffron (main plant) irrigation amounts and interval and no extra water was applied for winter wheat crop. Over the periods of time with no adequate rain, irrigation interval of 24 days was adopted based on the saffron plant (Azizi-Zohan et al., 2009). According to this rule, 5, 6, 6 and 5 irrigation events were conducted for all the treatments in the first, second, third and fourth growing seasons, respectively. At the first growing season, in order to help a good establishment of saffron plants all the treatments were fully irrigated at the first irrigation event which was adopted in late October 2013, and afterwards the experimental irrigation regimes was carried out. Last irrigation was adopted in late April while saffron leaf senescence was initiating, and wheat plant growth continued without irrigation.
2.6 Drainage water depth, its nitrogen concentration and leached nitrogen
The volume of the drained water collected from the bottom of each lysimeters were measured by a volumetric container after each irrigation event and divided by the area of the lysimeter (0.79 m) to get the drainage depth. After each irrigation event, the 0.1 L samples were taken from the drainage water of each lysimeter and kept at 4°C, and its nitrate (NO3−) concentration was determined spectrophotometerically in less than 24 h after sampling. For each irrigation event, leached nitrate was obtained by multiplying the drainage water volume by its NO3− concentration. For each growing season, seasonal leached nitrate was obtained from the sum of leached nitrate after each irrigation or rainfall event over that growing season. Finally, seasonal leached nitrogen was expresses as a percentage of the total applied nitrogen through manure or chemical urea fertilizer.
2.7 Evapotranspiration and its components
The plant(s) actual evapotranspiration (ETa) for the irrigation regimes was estimated through the water balance method applying Eq. (3) (Abbasi & Sepaskhah, 2022):
$${ET}_{a}=I+P-D\pm \varDelta S$$
3
where I is the irrigation depth [mm], P is the precipitation depth [mm], D is the drainage water depth [mm], and ΔS is soil water content change [mm] between two consecutive irrigation event in the root zone.
2.8 Irrigation and economic water productivities
Irrigation water productivity (IWP) was calculated as crop yield per cubic meter of total water use, including rainfall and irrigation water (Molden, 1997) [Eq. (4)].
Economic water productivity (EWP) was calculated as gross income in US$ per total water used in m3 (Molden et al., 2010) [Eq. (5)].
$$IWP=\frac{Y}{ {W}_{Irr.}+{W}_{Rain}}$$
4
$$EWP=\frac{Y*P}{ {W}_{Irr.}+{W}_{Rain}}$$
5
Where, IWP is irrigation water productivity for saffron (dry stigmas) or grain yields [kg m− 3], Y is yield (saffron dry stigmas or wheat grain) [kg ha− 1], WIrr. and WRain are irrigation and rain water use, respectively [m3 ha− 1], P is the yield (saffron stigmas or wheat grain) price [US$ kg− 1] and EWP is economic water productivity [US$ m− 3] for saffron, grain or total yields. The prices of wheat and saffron were US$ 0.38 kg− 1 and US$ 1066.7 kg− 1, respectively (Iran’s Ministry of Agriculture-Jahad, 2022). Saffron and wheat grain prices were in Iranian Rials and 1US$ = 300,000Rials (Central Bank of the Islamic Republic of Iran, 2022) was used to convert the prices from Iranian Rial to US Dollar.
2.9 Laboratory measurements
For the last (fourth) growing season, saffron corm, saffron aboveground biomass (leaves and petals) and aboveground biomass of the winter wheat plant (grain and straw) were oven dried at 70 ºC and their total nitrogen (N) and phosphorus (P) concentration were measured according to Kjeldal and ammonium-vanadate-molybdate methods, respectively (Chapman & Pratt, 1962). The protein concentration of plant organs (saffron corm, saffron aboveground biomass, wheat grain and straw biomasses) were determined through multiplying its Kjeldal nitrogen concentration by a Kjeldal-nitrogen-to-protein conversion factor of 6.25 (Magomya et al., 2014). Furthermore, the nitrate (NO3−) concentration of the drainage water was determined spectrophotometerically at 25°C using a previously calibrated scanning spectrophotometer (JENWAY 6405 UV/Vis., Dunmow, Essex, UK) set at 220 and 275 nm where the absorbance at 275 nm was taken as the background in the two wavelength determination method of nitrate. Leachate nitrate concentration was determined using a previously prepared standard nitrate curve (Baird et al., 2017).
2.10 Nitrogen and phosphorus indicators
2.10.1 Nutrient (nitrogen and phosphorus) harvest indices
For wheat plant, nitrogen harvest index (NHIWheat) is defined as the ratio between nitrogen uptake in grain (NGrain) and nitrogen uptake in grain plus straw (NGrain +NStraw) multiplied by hundred (Fageria, 2014) [Eq. (6)]. By a simple modification for saffron, the ratio between nitrogen uptake in corm (as the saffron plant’s main nitrogen sink) yield (NCorm) and nitrogen uptake in corm plus aboveground biomass yields (NCorm + NAbbvg.) multiplied by 100 would result in saffron plant nitrogen harvest index (NHISaffron) [Eq. (7)].
$${NHI}_{Wheat}=\frac{{N}_{Grain}}{ {N}_{Grain}+{N}_{Straw}}\text{*}100$$
6
$${NHI}_{Saffron}=\frac{{N}_{Corm}}{ {N}_{Corm}+{N}_{Abvg.}}\text{*}100$$
7
where, NHIWheat and NHISaffron are nitrogen harvest index for wheat and saffron, respectively [%], NGrain, NStraw, NCorm and NAbovg. are nitrogen uptake by wheat grain, wheat straw, saffron corm and saffron aboveground biomass yields, respectively [kg ha− 1].
By a simple modification, phosphorus harvest indices for wheat and saffron would be as Eqs. (8) and (9), respectively.
$${PHI}_{wheat}=\frac{{P}_{Grain}}{ {P}_{Grain}+{P}_{Straw}}\text{*}100$$
8
$${PHI}_{Saffron}=\frac{{P}_{Corm}}{ {P}_{Corm}+{P}_{Abvg.}}\text{*}100$$
9
where, PHIWheat and PHISaffron are phosphorus harvest index for wheat and saffron, respectively [%], PGrain, PStraw, PCorm and PAbovg. are phosphorus uptake by wheat grain, wheat straw, saffron corm and saffron aboveground biomass yields, respectively [kg ha− 1].
2.10.2 Nutrient (nitrogen and phosphorus) acquisition (uptake) efficiency
The nitrogen acquisition efficiency (NAE) is a soil-based nitrogen efficiency (Congreves et al., 2021) which addresses the nitrogen uptake by yield (grain for wheat and corm for saffron) per unit of available nitrogen in soil system (the sum of soil initial available N and fertilizers’ available N). When flowering is finished at the first growing season, the daughter corms start to develop and grow on top of the mother corms. At the end of the first growing season, the color of saffron leaves change from green to yellow and development of the daughter corms is completed (Koocheki & Seyyedi, 2015a). At the following growing seasons with the aging of saffron plant, primary mother corms gradually become smaller and smaller. Hence, in this study, mother corm nitrogen content was not considered in NAE calculation for the fourth growing season. Therefore, grain and corm NAE is calculated as Eqs. (10) and (11) (Congreves et al., 2021):
$${NAE}_{Grain}=\frac{{N}_{Grain}}{{N}_{Soil}+ {N}_{Fer.}}\text{*}100$$
10
$${NAE}_{Corm}=\frac{{ N}_{Corm}}{{N}_{Soil}+ {N}_{Fer.}}\text{*}100$$
11
where, NAEGrain and NAECorm are nitrogen acqisition efficiencies of grain and corm (%), respectively, NGrain and NCorm are the grain and corm nitrogen uptake (kg ha− 1), respectively, NSoil and NFer. are the soil and fertilizere (manure or chemical urea) available nitrogen (kg ha− 1), respectively. For PAE calculations, phosphorus values have to be replaced with nitrogen in Eqs. (10) and (11). Moreover, NAE and PAE can be calculated for the whole plant. For example NAE for saffron plant is the nitrogen uptake by saffron plant (corm and above-ground biomass) per unit of available nitrogen in soil system (soil plus applied fertilizer).
2.10.3 Nutrient (nitrogen and phosphorus) utilization efficiency
The nutrient (N and P) utilization efficiency (NUtE and PUtE, respectively) addresses the yield produced per unit of N and P, respectively, acquired (uptake) by the plant [Eqs. (12) and (13)] (Moll et al., 1982).
$${NUtE}_{Grain}=\frac{{Y}_{Grain}}{{N}_{Grain}+ {N}_{Straw}}\text{*}100$$
12
$${NUtE}_{Corm}={\frac{{Y}_{Corm}}{{N}_{Corm}+ N}}_{Abvg.}\text{*}100$$
13
where, NUtEGrain and NUtECorm are grain and corm nitrogen utilization efficiencies [kg kg− 1], respectively, YGrain and YCorm are the grain and corm yields [kg ha− 1], respectively, NGrain, NStraw, NCorm and NAbvg. are nitrogen uptake by whrat grain, wheat straw, saffron corm, saffron aboveground biomasses (kg ha− 1), respectively. For PUtE calculations, phosphorus values have to be replaced with nitrogen in Eqs. (12) and (13).
2.10.4 Nutrient (nitrogen and phosphorus) use efficiency
Nitrogen use efficiency is a soil-based nitrogen efficiency and is defined as yield per unit of soil-system available nitrogen (the sum of soil initial available N and fertilizer available N) (Congreves et al., 2021; Moll et al., 1982) as Eqs. (14) and (15):
$${NUE}_{Grain}=\frac{{Y}_{Grain}}{{N}_{Soil}+ {N}_{Fer.}}$$
14
$${NUE}_{Corm}=\frac{{ Y}_{Corm}}{{N}_{Soil}+ {N}_{Fer.}}$$
15
where, NUEGrain and NUECorm are nitrogen use efficiencies of grain and corm (kg kg− 1), respectively, YGrain and YCorm are the grain and corm yields (kg ha− 1), respectively, NSoil and NFer. are the soil and fertilizere (manure or chemical urea) available nitrogen (kg ha− 1), respectively. This definition can be simply modified for phosphorus use efficiency as yield per unit of P available in soil system (the sum of soil initial available P and fertilizer available P). For PUE calculations, phosphorus values have to be replaced with nitrogen in Eqs. (14) and (15).
2.10.5 System N balance index (SNBI)
The nitrogen balance index of a system (SNBI) is calculated as Eq. (16) (Sainju, 2017):
$$SNBI={N}_{Input}-{N}_{Output}-\varDelta \text{S}\text{o}\text{i}\text{l} \text{t}\text{o}\text{t}\text{a}\text{l} \text{N}$$
16
where SNBI is the system N balance index, NInput is the system measured nitrogen inputs including nitrogen supply in fertilizer (chemical urea or organic manure), nitrogen from irrigation water and rain N depositions, NOutput is the measured system nitrogen outputs including crop N removal (saffron corm, saffron aboveground biomass, wheat grain and straw biomass), N losses through N leaching and ∆ Soil total N, is nitrogen change in soil, all in kg ha− 1. In this study, SNBI shows N loses through NH4 volatilization, denitrification, gas emissions [NOx] and plant senescence which could not be determined directly in this study. Since this is a lysimetric study and the lysimeters are closed and water-sealed around systems, there would be no nitrogen loss through surface runoff and soil erosion to be taken into account in nitrogen loss calculations.
2.11 Statistical analysis
Minitab 16.2.4 statistical software was applied to determine interaction effects of irrigation regimes, sources of nitrogen and cropping systems. Analysis of variance (ANOVA) was carried out according to Tukey test to determine statistically significant differences between the means at 5% probability level.
2.12 Data availability statement
All data generated or analyzed during this study are included in this manuscript and its supplementary attachment.