Effect of Water and Nitrogen Coupling Regulation on the Growth, Physiology, Yield, and Quality Attributes of Isatis tinctoria L. in the Oasis Irrigation Area of the Hexi Corridor

Abstract

To address the prevailing problems of high water and fertilizer input and low productivity in Isatis tinctoria L. production in the Hexi Corridor in China, the effects of different irrigation amounts and nitrogen application rates on growth characteristics, photosynthetic physiology, root yield, and quality of I. tinctoria plants were studied with the aim of obtaining the optimal irrigation level and nitrogen application rate. From 2021 to 2023, we established a two-factor split-plot experiment in the oasis irrigation area with three irrigation amounts (sufficient water, medium water, and low water are 100%, 85%, and 70% of the typical local irrigation quota) for the main zone; three nitrogen application rates (low nitrogen, 150 kg ha⁻¹, medium nitrogen, 200 kg ha⁻¹, and high nitrogen, 250 kg ha⁻¹) for the secondary zone; and three irrigation amounts without nitrogen as the control to explore the response of these different water and nitrogen management patterns for I. tinctoria in terms of growth characteristics, photosynthetic physiology, root yield, and quality. The results showed the following: (1) When the irrigation amount was increased from 75% to 100% of the local typical irrigation quota and the nitrogen application rate was increased from 150 to 250 kg ha⁻¹, while the plant’s height, leaf area index, dry matter accumulation in the stem, leaf, and root, as well as the net photosynthetic rate (Pn), the stomatal conductance (Gs), and the transpiration rate (Tr) of I. tinctoria increased gradually, and the root–shoot ratio decreased. (2) When the irrigation amount increased from 75% to 100% of the local typical irrigation quota, the yield and net proceeds of I. tinctoria increased from 43.12% to 53.43% and 55.07% to 71.61%, respectively. However, when the irrigation quota was 100% of the local typical irrigation quota, and the nitrogen application rate increased from 150 to 200 kg ha⁻¹, the yield of I. tinctoria increased from 21.58% to 23.69%, whereas the increase in nitrogen application rate from 200 to 250 kg ha⁻¹ resulted in a decrease in the yield of I. tinctoria from 10.66% to 18.92%. During the 3-year experiment, the maximum yield of I. tinctoria appeared when treated with sufficient water and medium nitrogen, reaching 9054.68, 8066.79, and 8806.15 kg ha⁻¹, respectively. (3) The effect of different water and nitrogen combination treatments on the root quality of I. tinctoria was significant. Under the same irrigation level, increasing the nitrogen application rate from 150 to 250 kg ha⁻¹ could increase the contents of indigo, indirubin, (R,S)–goitrin, total nucleoside, uridine, and adenosine in the root of I. tinctoria from 3.94% to 9.59%, 1.74% to 12.58%, 5.45% to 18.35%, 5.61% to 11.59%, 7.34% to 11.32%, and 14.98% to 54.40%, respectively, while the root quality of I. tinctoria showed a trend of first increasing and then decreasing under the same nitrogen application level. (4) AHP, the entropy weight method, and the TOPSIS method were used for a comprehensive evaluation of multiple indexes of water–nitrogen coupling planting patterns for I. tinctoria, which resulted in the optimal evaluation of the W3N2 combination. Therefore, the irrigation level was 100% of the local typical irrigation quota, the nitrogen application rate should be appropriately reduced, and controlling the nitrogen application rate at the level of 190.30–218.27 kg ha⁻¹ can improve water–nitrogen productivity yields for I. tinctoria and root quality. The results of this study can provide a theoretical basis and technical support for a more reasonable water and fertilizer management model for the I. tinctoria production industry in the Hexi Corridor in China.

1. Introduction

I. tinctoria is an anti-influenza virus with antibacterial, anti-inflammatory as well as anti-endotoxin properties. It is widely used as a conventional drug for clinical dispensation in traditional Chinese medicine [1]. The development of pharmaceutical technology and improvements in living standards have seen an increase in demand for higher quality Chinese medicine products. This has also led to higher quality control requirements for I. tinctoria production. The I. tinctoria planting industry is facing multiple pressures, such as rigid resource constraints, impure germplasm [2], continuous rises in labor costs, restricted availability of arable land for cultivation, sloppy field management [3], unscientific irrigation and fertilization methods [4], excessive use of pesticides [5], and the high energy consumption requirements of mechanization [6]. At the same time, climate change, as well as ecological and water resources crises [7,8], have severely constrained the sustainable production capacity of the I. tinctoria industry, and the development of traditional Chinese medicine eco-agriculture is still facing multifaceted challenges.

The unique natural environmental conditions and location advantages of the oasis irrigation area in the Gansu Hexi Corridor make it the best ecological area for I. tinctoria planting; the output in Minle County accounts for more than 60% of the national production, giving it the reputation of the “Hometown of Chinese I. tinctoria” [9]. However, local farmers have no quantitative concept of fertilizer application and irrigation, and there is a common problem of excessive irrigation and fertilizer application in I. tinctoria production and planting. This not only results in the loss of limited water resources through inefficient usage and increased production costs, but it also adversely affects the yield and medicinal quality of I. tinctoria and leads to environmental pollution caused by nitrate nitrogen leaching, which threatens the ecological security of the Qilian Mountains.

The China Hexi Corridor oasis irrigation area is a typical ecologically fragile area, where precipitation is low, inter-annual variability is large, and where water scarcity is still a key factor that severely limits the large-scale planting of I. tinctoria. Therefore, clarifying the crop’s water consumption characteristics, selecting efficient water-saving irrigation methods, and improving water use efficiency are the main ways to alleviate the supply and demand issues of water resources [10]. It has been confirmed through a large number of experiments that the water-saving and yield-increasing effect of drip irrigation under plastic film mulching is the best among the efficient water-saving irrigation methods, and the amount of water saved can be as much as 50% [11]. Drip irrigation under plastic film mulching can not only improve field water use efficiency, avoid deep seepage, and reduce inter-plant evaporation [12] but also perform warming and moisture keeping functions [13], increasing crop yields [14], and improving fruit quality [15], and it has been widely used in agricultural irrigation in the arid areas of northwestern China, especially in the Hexi Corridor and Xinjiang [16]. The technology has been widely popularized in the region in staple crops such as maize [17], cotton [18], and potatoes [19], and it has also been widely applied to cash crops, such as tomatoes [20] and cucumbers [21].

Nitrogen is the key nutrient element required for the growth and development of I. tinctoria, and it is also one of the main limiting factors for the yield and quality of I. tinctoria [22]. Low nitrogen application rates cannot meet the nitrogen demand for I. tinctoria production, inhibiting the growth of roots, stems, and leaves [23]. An excessive nitrogen application rate would lead to the deterioration of soil properties, nutrient imbalances, and weakened resilience to adversity [24]. Therefore, scientific and reasonable nitrogen application rates is particularly important for achieving the high quality and high yield of I. tinctoria as well as improving the nitrogen use efficiency. Previous studies have shown that for most crops, due to the lack of scientific guidance, the rates of nitrogen applied by farmers during the planting process are often greater than that required for maximum yield in order to obtain higher yield and economic value [25]. An excessive nitrogen application rate caused soil nitrogen excess, soil acidification and salinization intensification [26], affected soil permeability and aeration [27], destroyed soil granular structure, resulted in soil hardening [28], and destroyed the balance between crop nutrient growth and reproductive growth [29], reduced crop yield and nitrogen use efficiency [30], increased production costs, and declined economic benefits [31]. Therefore, it is imperative to optimize nitrogen management during the growth period of I. tinctoria, which can not only reduce the nitrogen application rates but also improve the nitrogen use efficiency and environmental sustainability.

Studies on the effects of water–nitrogen coupling on crop yield and quality and related models have been reported [32]. Jin et al. [33] showed that the irrigation amounts, nitrogen application rates, and rice yield and quality showed a synergistic effect of water and nitrogen on increasing yield and regulating quality in a certain range. Cai et al. [34] established a melon quality evaluation system based on 10 indicators from three categories of yield, quality, and efficiency, and the results showed that there was a significant interaction effect between the irrigation amounts and nitrogen application rates, which could achieve high yield while improving the quality and water–nitrogen use efficiency of melons. Ma et al. [35] constructed a water and nitrogen management model with better indicators based on a TOPSIS model and comprehensive scoring method. Using this model to comprehensively evaluate the yield, quality, and water–nitrogen use efficiency of wolfberry, it was found that the water–nitrogen interaction had a significant effect. The TOPSIS model showed the results provide a scientific basis for the optimal irrigation and management of fertilization of wolfberry in arid regions. Numerous studies have collectively confirmed that both the single factor of water and nitrogen and their interaction have significant effects on crop yield and quality but with variations due to different factors such as crop species, regional climate, water and nitrogen usage.

Although there have been relevant reports on the optimal nitrogen application rates for I. tinctoria, it is still unclear how to respond to the optimal nitrogen application rates under different irrigation amounts. Therefore, in this study, we carried out experimental studies on the effects of different water and nitrogen combinations for I. tinctoria yield and components, quality, water–nitrogen productivity through a locational experiment using a split-plot experimental design. The AHP method was used to subjectively judge and quantitatively describe the relative importance of multiple indicator factors, such as I. tinctoria yield, quality, and water–nitrogen productivity. The entropy weighting method was used to calculate the variance degree of each indicator and obtain the weighting values of each indicator. Finally, the TOPSIS method was used to compare the advantages and disadvantages of each indicator, and the optimal water and nitrogen combination model under the multi-indicator synergistic system was selected. At the same time, a binary quadratic regression simulation was carried out based on the comprehensive growth score of I. tinctoria and water and nitrogen input to derive the optimal irrigation and fertilizer interval in the agricultural production of I. tinctoria with the aim of providing a theoretical reference for the optimal utilization of water and nitrogen in the planting and production of I. tinctoria as well as for increasing yield and regulating quality.

2. Materials and Methods

2.1. Experimental Site

The experimental was conducted in Yimin Irrigation Experimental Station, Minle County, which is the middle part of the Hexi Corridor, Gansu Province (100°43′ E, 38°39′ N, 1970 m a.s.l.) (Figure 1). The climate in the experimental area is temperate continental. The average annual precipitation is 200 mm, the annual evaporation is 1900 mm, the average annual temperature is 6.0 °C, and the frost-free period is about 105 d based on meteorological data of the last 20 years. The soil is a light loam with a topsoil bulk density of 1.46 g cm⁻³ and maximum field water-holding capacity of 24%. The basal fertility status and the effective photosynthesizing radiation of the crops during the symbiosis stage are shown in

Table 1. Basic physical and chemical properties of 0–40 cm soil layer in the experimental site.

Soil Depth (cm)	OM (g kg−1)	TN (g kg−1)	TP (g kg−1)	TK (g kg−1)	NN (mg kg−1)	AN (mg kg−1)	AP (mg kg−1)	AK (mg kg−1)	pH
0–20	13.9	1.27	1.15	24.7	203.75	25.50	102.8	820	8.21
20–40	13.0	1.18	0.99	22.1	116.40	23.80	50.4	594	8.45

OM: soil organic matter content, TN: soil total nitrogen content, TP: soil total phosphorus content, TK: soil total potassium content, NN: soil nitrate nitrogen content, AN: soil ammonium nitrogen content, AP: soil available phosphorus content, AK: soil available potassium content.

, and the precipitation and the average temperature over the experiments are shown in Figure 2.

2.2. Experimental Design

The irrigation amounts were based on the evapotranspiration (ET₀) of the reference crop and precipitation, and the nitrogen application rates were based on the nitrogen uptake of I. tinctoria and the soil nitrogen contents. The experiment was a two-factor split-plot design experiment, including two factors: the drip irrigation amounts under the plastic film mulching and the nitrogen application rates, where the main factor was the irrigation amounts, and three irrigation gradients were used as follows: low (W1, 70% of the local typical irrigation quota), medium (W2, 85% of the local typical irrigation quota), and sufficient water (W3, 100% of the local typical irrigation quota). The secondary area was the nitrogen application rates, and four gradients were set up: zero (N0, 0 kg ha⁻¹), low (N1, 150 kg ha⁻¹), medium (N2, 200 kg ha⁻¹), and high nitrogen (N3, 250 kg ha⁻¹) (

Table 2. The irrigation quota and fertilizer application of different year.

Treatment	Irrigation Quota (mm)			Total Fertilizer Application (kg ha−1)			Factor Code Value
	2021	2022	2023	N	P	K	Irrigation Amount ×1	Nitrogen Application Rate ×2
W1N0	121.5	114.0	151.7	0	350	200	—	—
W1N1	121.5	114.0	151.7	150	350	200	0.0	0.0
W1N2	121.5	114.0	151.7	200	350	200	0.0	0.5
W1N3	121.5	114.0	151.7	250	350	200	0.0	1.0
W2N0	147.6	138.5	184.2	0	350	200	—	—
W2N1	147.6	138.5	184.2	150	350	200	0.5	0.0
W2N2	147.6	138.5	184.2	200	350	200	0.5	0.5
W2N3	147.6	138.5	184.2	250	350	200	0.5	1.0
W3N0	173.6	162.9	216.7	0	350	200	—	—
W3N1	173.6	162.9	216.7	150	350	200	1.0	0.0
W3N2	173.6	162.9	216.7	200	350	200	1.0	0.5
W3N3	173.6	162.9	216.7	250	350	200	1.0	1.0

Note: W1N0 indicates 70% of the local typical irrigation quota and 0 kg ha⁻¹ nitrogen application; W1N1 indicates 70% of the local typical irrigation quota and 150 kg ha⁻¹ nitrogen application; W1N2 indicates 70% of the local typical irrigation quota and 200 kg ha⁻¹ nitrogen application; W1N3 indicates 70% of the local typical irrigation quota and 250 kg ha⁻¹ nitrogen application; W2N0 indicates 85% of the local typical irrigation quota and 0 kg ha⁻¹ nitrogen application; W2N1 indicates 85% of the local typical irrigation quota and 150 kg ha⁻¹ nitrogen application; W2N2 indicates 85% of the local typical irrigation quota and 200 kg ha⁻¹ nitrogen application; W1N3 indicates 85% of the local typical irrigation quota and 250 kg ha⁻¹ nitrogen application; W3N0 indicates 100% of the local typical irrigation quota and 0 kg ha⁻¹ nitrogen application; W3N1 indicates 100% of the local typical irrigation quota and 150 kg ha⁻¹ nitrogen application; W3N2 indicates 100% of the local typical irrigation quota and 200 kg ha⁻¹ nitrogen application; W3N3 indicates 100% of the local typical irrigation quota and 250 kg ha⁻¹ nitrogen application.

). The experiment consisted of 12 treatments, each of which was repeated 3 times and arranged in randomized groups. The plots were all running east–west with an area of 60 m² (10.0 m in length and 6.0 m in width). The plots were separated by burying 1.0 m deep PVC plastic sheeting and building 20 cm high ridges to prevent water and fertilizer from affecting each other between the plots (Figure 3).

The I. tinctoria seeds for the test were provided by the professional cooperative of Chinese Herbal Medicine Planting in Minle County, with large, full, uniform and 98.6% seed purity. A Ycbz-6 Chinese herbal medicine film-laying drip irrigation belt integrated seeder was used to carry out precision sowing with eight rows in one film and east–west rows planted at a row spacing of 0.15 m, and at a plant spacing of 0.08 m, with a planting density of 8.3 × 105 hm⁻². Field management and pest control measures were managed by the farmers in accordance with the requirements of cooperatives. Th seeds were sown on 27 April 2021, harvested on 8 October, sown on 1 May 2022, harvested on 10 October, sown on 4 May 2023, harvested on 12 October.

They were irrigated nine times during the I. tinctoria growing period, and each plot was fitted with an individual water meter to control the irrigation amounts. Nitrogen (urea, produced by Gansu Liuhua Group Co., Ltd., Linxia, China, N ≥ 46%), phosphate (calcium superphosphate, produced by Shandong Mengshi Chemical Co., Ltd., Jinan, China, P₂O₅ ≥ 12%) and potash (potassium sulfate, produced by Shandong Huali Fertiliser Co., Ltd., Liaocheng, China, K₂O ≥ 52%) were applied to the base fertilizer with the same amounts for each treatment, which were 80 kg ha⁻¹, 350 kg ha⁻¹ and 60 kg ha⁻¹, respectively. The remaining nitrogen was applied in late June, late July, and early August by combined drip irrigation at a ratio of 4:3:3, and the remaining potassium was applied in early August and September by combined drip irrigation at a ratio of 1:1. The distance between the drip irrigation belt and drip head was 30 cm, the drip flow rate was 2.5 L h⁻¹, and the normal irrigation pressure was 0.1 MPa. The plastic film was made of ordinary transparent plastic film with a width of 120 cm and a thickness of 0.08 mm. The field application rates for each factor of the experiment, as well as the coded value after normalization, are shown in Table 2, and the nitrogen application rates are the converted pure nitrogen rates.

2.3. Measurement Items and Methods

2.3.1. Meteorological Data

Meteorological data such as precipitation, solar radiation, air temperature, relative air humidity, and wind speed were obtained from a microclimate observer (MC-NQXZ type) installed in the nearby experiment station.

2.3.2. Determination of Biological Indicators

At the maturity stage of I. tinctoria, 10 standard plants were randomly selected from the non-side rows of each experimental plot. The roots were removed, the plant samples were measured with a straightedge with a range of 100 cm, the maximum distance from the bottom of the stem to the growing point of I. tinctoria was measured, and the data of three plants were recorded. Finally, the mean value was calculated to be the average height of I. tinctoria in the plot. The leaf area of all leaves were measured by YMJ-B Leaf Area Measuring Instrument (produced by Zhejiang Tuopuyunnong Technology Co., Ltd., Hangzhou, China), and finally, the average value was taken and the Leaf Area Index (LAI = Total Leaf Area/Land Area) was calculated.

The roots of I. tinctoria were cleaned, and the length of the main root was determined by using a straight ruler with a range of 100 cm, and the diameter of the main root was determined by using vernier calipers with an accuracy of 0.01 mm. After the determination, the above-ground portion and the roots were bagged and killed in a constant temperature oven at 105 °C for 30 min, dried at a constant temperature of 80 °C to a constant weight, and then weighed on an electronic balance with an accuracy of 0.01 g. The average value was taken, and the root–crown ratio was calculated (root–crown ratio = root dry matter mass/above-ground dry matter mass).

2.3.3. Determination of Photosynthetic Physiological Indicators

At the seedling, vegetative, fleshy root growth and fleshy root maturity stage of I. tinctoria, we chose several sunny and cloudless days from 9:00 a.m. to 11:00 a.m., we selected three plants with good growth condition and basically the same growth condition in each plot, and used a LI-6400 portable photosynthetic system analyzer (produced by LI-COR Company, Lincoln, NE, USA) to determine the net photosynthesis rate (Pn), the transpiration rate (Tr), and the stomatal conductance (Gs) of the third leaf from the inside to the outside, and the results showed the average values.

2.3.4. Determination of Yield

I. tinctoria was harvested separately in each plot after maturity, dried naturally, weighed and converted into yield per hectare.

2.3.5. Determination of Water–Nitrogen Productivity

• Soil bulk density

The ring knife method was used to determine the soil bulk density, the sampling depth was 100 cm, the soil was taken in layers every 20 cm and repeated three times, the soil samples were dried in the oven at 105 °C until constant weight, and the soil bulk density of each layer was calculated with a ring knife volume of 100 cm³.

$${\gamma }_{\mathrm{i}}=m_i/100$$

(1)

where γ_i is the soil bulk density of the i layer, g cm⁻³; and m_i is the drying weight of the i layer, g.

2.

Soil moisture content

Soil moisture content was determined using CPN 503DR Neutron Soil Moisture Tester (produced by Beijing Bolunjingwei Technology Development Co., Ltd., Beijing, China). The soil samples were measured every 7 days during the whole growth period of I. tinctoria with additional measurements before sowing, after harvesting, before and after irrigation and rainfall, at a sampling depth of 100 cm, with a gradient of every 20 cm. Three measurement points were taken in each plot, and the average value was taken.

3.

Water consumption of I. tinctoria

The water consumption (ET) of I. tinctoria was calculated using the water balance method. Since the irrigation method was drip irrigation under plastic film mulching, evaporation losses could be ignored; that is, the irrigation infiltration rate was 100%. For rainfall infiltration, the effect of plastic film mulching on rainfall infiltration was calculated by calculating the ratio of the increase in soil moisture before and after a single rainfall to the current rainfall, that is,

$$\lambda ={\gamma }_{\mathrm{i}}{\displaystyle \sum ({w_i}_f-{w_i}_b)}d$$

(2)

where λ is rainfall infiltration, mm; W_if is the soil volumetric water content of the i layer after rainfall, cm³ cm⁻³; W_ib is the soil volumetric water content of the i layer before rainfall, cm³ cm⁻³; and d is the thickness of the soil layer, mm.

Water consumption was calculated using the water balance formula. Since the depth of groundwater in the experimental area was greater than 10 m, and the irrigation method was drip irrigation, there was no runoff drainage during the growth period, groundwater recharge and seepage were ignored: that is,

$$ET=\mathsf{\lambda }+I+W_{\mathrm{b}}-W_{\mathrm{f}}$$

(3)

where ET is the total water consumption during the growth period of the crop, mm; λ is the rainfall infiltration, mm; I is the effective irrigation amounts, mm; W_b is soil water storage before the sowing of I. tinctoria, mm; and W_f is soil water storage after harvesting of I. tinctoria, mm.

4.

Water productivity

Water productivity (WP, kg m⁻³) is

$$WP=Y/ET$$

(4)

Irrigation water productivity (IP, kg m⁻³) is

$$IP=Y/I$$

(5)

where Y is the yield of I. tinctoria, kg ha⁻¹.

5.

Nitrogen use efficiency

Nitrogen partial productivity (PN, kg kg⁻¹) is

$$PN=Y/N$$

(6)

where N is nitrogen input, kg ha^{− 1}.

Nitrogen agronomy use efficiency (AN, kg kg⁻¹) is

AN = (Y_N − Y₀)/N

(7)

where Y_N is grain yield with nitrogen application, kg ha⁻¹; Y₀ is grain yield without nitrogen application, kg ha⁻¹.

2.3.6. Determination of Economic Benefit

(1)

Total inputs

Total inputs per unit area = irrigation water cost + fertilizer cost + seed cost + labor cost + machinery cost + other costs.

(2)

Total output value

Total output value per unit area = I. tinctoria production per unit area × I. tinctoria unit price. Through investigation, the market unit price of I. tinctoria in Minle County is 12.6 CNY kg⁻¹ in 2021, 10.2 CNY kg⁻¹ in 2022, and 8.2 CNY kg⁻¹ in 2023.

(3)

Economic benefit

Economic benefit per unit area = total output value per unit area—total input per unit area.

(4)

Output–input ratio

Output–input ratio = total output value per unit area/total input per unit area.

2.3.7. Determination of Quality Indicators

The content of the active ingredients indigo, indirubin, (R,S)–goitrin in I. tinctoria was determined by high-performance liquid chromatography (HPLC) according to the Chinese Pharmacopoeia [36]. The LC−10ATVP high-performance liquid chromatography was used with the following chromatographic conditions: SPD−10Avp (UV−VIS) detector, the column was AgilentZorbaxSB−C18 (100 mm × 4.6 mm, 3.5 μm), methanol−0.1% formic acid solution as the mobile phase; the flow rate was 1.0 mL min⁻¹, and the injection volume was 20 μL by autosampler. The detection wavelength was 280 nm, and the column temperature was 25 °C. The total nucleoside, uridine and adenosine contents of I. tinctoria root were determined by the HPLC−DAD method [37].

2.3.8. The Water–Nitrogen Coupling Model

In this experiment, we studied the regression relationships between three different levels of drip irrigation amounts under plastic film mulching, nitrogen application rates, and comprehensive growth scores of I. tinctoria. Taking the comprehensive growth score as the target, and the irrigation amounts and nitrogen application rates as the independent variables, we established the regression model of irrigation amounts–nitrogen application rates–comprehensive growth score, and expressed the model using a binary quadratic regression equation. The expression for the binary quadratic regression equation is as follows:

$$y=a_0+a_1{x}_1+a_2{x}_2+a_{12}{x}_1{x}_2+a_{11}{x}_1^2+a_{22}{x}_2^2$$

(8)

where y is the comprehensive growth score of I. tinctoria, kg ha⁻¹; a₀ is the constant term of the regression model; a₁, a₂ are the first-order coefficients of the regression model; a₁₂ is the interaction coefficient of the regression model; and a₁₁, a₂₂ are the quadratic coefficients of the regression model.

2.4. Statistical Analysis Methods

The LSD multiple comparison method in SPSS (Version 22.0, IBM, Inc., New York, NY, USA) software was used for statistical analysis and regression model building. Origin (Version 8.0, Origin Lab, Corp., Hampton, MA, USA) software was used for plotting. Yaaph (Version v12.11.8293, Meta Decision Software Technology Co., Ltd., Corp., Taiyuan, China) software was used to draw the comprehensive analytical hierarchical model of I. tinctoria and the weight analysis of each index. Microsoft Excel (Version 2010, Microsoft Corp., Raymond, WA, USA) software was used to calculate the comprehensive evaluation values according to the TOPSIS method. DPS was used to establish the mathematical model. MATLAB (Version R2023b, Math Works, Corp., Natick, MA, USA) software was used to analyze the model.

3. Results and Analysis

3.1. Effects of Water–Nitrogen Interaction on Biological Characteristics of I. tinctoria

Reasonable water–nitrogen interaction can improve the growing environment of I. tinctoria, which in turn promotes the growth of plant stems, leaves and roots and increases the dry matter accumulation. The results of water–nitrogen ANOVA showed that the irrigation amounts had a significant or extremely significant effect on the biological characteristics of I. tinctoria, while nitrogen application rates only had a significant or extremely significant effect on the plant height, taproot length and root–crown ratio, and they had no significant effect on other biological characteristics. The water–nitrogen interaction did not have a significant effect on the biological characteristics of I. tinctoria (

Table 3. Effects of different irrigation amounts and nitrogen application rates on I. tinctoria biological characteristic indicators (mean ± standard error). Different lowercase letters within columns indicate significant differences among treatments at p < 0.05.

Year	Treatment	Plant Height (cm)	Leaf Area Index	Dry Matter Accumulation In Stem and Leaf (g)	Taproot Length (cm)	Taproot Diameter (mm)	Dry Matter Accumulation In Root (g)	Root−Crown Ratio
2021	W1N0	19.39 ± 0.37 f	0.72 ± 0.012 g	6.84 ± 0.22 g	15.12 ± 0.61 g	8.73 ± 0.04 e	5.49 ± 0.07 g	0.80 ± 0.017 d
	W1N1	22.34 ± 0.66 de	0.94 ± 0.015 f	8.12 ± 0.31 f	17.83 ± 0.30 ef	10.64 ± 0.14 d	6.46 ± 0.15 f	0.80 ± 0.029 d
	W1N2	24.28 ± 0.30 cd	1.07 ± 0.040 e	9.85 ± 0.29 e	18.54 ± 0.38 e	12.81 ± 0.32 c	8.51 ± 0.35 e	0.86 ± 0.022 d
	W1N3	26.06 ± 1.00 bc	1.16 ± 0.045 de	10.76 ± 0.19 cd	18.87 ± 0.37 e	14.77 ± 0.42 b	9.08 ± 0.16 e	0.84 ± 0.036 d
	W2N0	21.65 ± 0.86 ef	1.13 ± 0.015 e	8.05 ± 0.24 f	16.71 ± 0.42 f	12.96 ± 0.48 c	8.84 ± 0.34 e	1.10 ± 0.023 ab
	W2N1	25.72 ± 0.97 bc	1.37 ± 0.036 c	9.94 ± 0.40 de	19.20 ± 0.43 de	14.46 ± 0.41 b	10.47 ± 0.28 d	1.05 ± 0.031 bc
	W2N2	28.08 ± 1.13 ab	1.51 ± 0.029 b	11.93 ± 0.29 b	19.44 ± 0.75 de	15.93 ± 0.30 a	11.73 ± 0.12 c	0.98 ± 0.042 c
	W2N3	29.18 ± 0.63 a	1.56 ± 0.045 b	12.66 ± 0.47 ab	20.65 ± 0.44 cd	16.25 ± 0.49 a	12.31 ± 0.18 bc	0.97 ± 0.037 c
	W3N0	24.91 ± 1.09 cd	1.25 ± 0.031 d	9.18 ± 0.26 e	18.23 ± 0.62 ef	13.70 ± 0.51 bc	10.22 ± 0.33 d	1.11 ± 0.025 ab
	W3N1	27.66 ± 0.86 ab	1.48 ± 0.036 b	10.83 ± 0.25 c	21.86 ± 0.66 bc	16.01 ± 0.53 a	12.50 ± 0.24 b	1.15 ± 0.031 a
	W3N2	30.08 ± 0.74 a	1.69 ± 0.032 a	12.95 ± 0.23 a	23.69 ± 0.81 a	17.13 ± 0.34 a	13.98 ± 0.20 a	1.08 ± 0.055 ab
	W3N3	30.16 ± 1.05 a	1.72 ± 0.061 a	13.26 ± 0.15 a	23.01 ± 0.22 ab	16.96 ± 0.34 a	13.66 ± 0.29 a	1.03 ± 0.038 bc
2022	W1N0	18.49 ± 0.25 f	0.77 ± 0.016 h	7.11 ± 0.29 g	14.58 ± 0.23 g	8.91 ± 0.20 f	5.73 ± 0.15 h	0.81 ± 0.029 de
	W1N1	20.87 ± 0.45 e	1.02 ± 0.035 g	9.73 ± 0.26 ef	16.91 ± 0.42 f	11.09 ± 0.23 e	7.18 ± 0.25 g	0.74 ± 0.026 e
	W1N2	23.28 ± 0.55 cd	1.13 ± 0.029 ef	11.01 ± 0.34 cd	18.36 ± 0.34 e	13.00 ± 0.45 d	9.02 ± 0.34 f	0.82 ± 0.020 d
	W1N3	25.89 ± 0.91 b	1.20 ± 0.020 e	11.42 ± 0.40 c	19.05 ± 0.74 de	13.94 ± 0.29 d	9.75 ± 0.10 e	0.85 ± 0.010 d
	W2N0	22.15 ± 0.43 de	1.19 ± 0.044 e	8.60 ± 0.17 f	17.02 ± 0.39 f	13.25 ± 0.36 d	8.59 ± 0.25 f	1.00 ± 0.033 ab
	W2N1	24.96 ± 0.83 bc	1.43 ± 0.053 cd	10.36 ± 0.29 cde	19.96 ± 0.47 cd	14.18 ± 0.12 cd	10.91 ± 0.18 d	1.05 ± 0.017 a
	W2N2	26.43 ± 0.87 b	1.55 ± 0.041 bc	12.95 ± 0.43 b	20.57 ± 0.21 bc	15.64 ± 0.55 ab	12.56 ± 0.25 bc	0.97 ± 0.038 b
	W2N3	27.16 ± 0.93 b	1.60 ± 0.049 b	13.80 ± 0.55 b	21.83 ± 0.13 a	15.86 ± 0.38 ab	13.23 ± 0.33 b	0.96 ± 0.025 bc
	W3N0	23.52 ± 0.95 cd	1.34 ± 0.025 d	9.93 ± 0.32 de	18.78 ± 0.38 de	14.11 ± 0.44 cd	10.63 ± 0.29 d	1.07 ± 0.020 a
	W3N1	25.88 ± 0.08 b	1.53 ± 0.036 bc	12.71 ± 0.48 b	21.25 ± 0.60 ab	15.29 ± 0.43 bc	12.08 ± 0.14 c	0.95 ± 0.051 bc
	W3N2	29.24 ± 0.93 a	1.75 ± 0.074 a	15.36 ± 0.64 a	22.36 ± 0.35 a	16.85 ± 0.36 a	14.33 ± 0.29 a	0.93 ± 0.022 bc
	W3N3	29.95 ± 0.58 a	1.79 ± 0.051 a	15.88 ± 0.56 a	21.94 ± 0.26 a	16.77 ± 0.63 a	14.02 ± 0.37 a	0.88 ± 0.043 cd
2023	W1N0	17.42 ± 0.22 f	0.68 ± 0.023 h	6.95 ± 0.14 h	13.95 ± 0.43 f	8.63 ± 0.06 g	6.11 ± 0.16 h	0.88 ± 0.012 de
	W1N1	19.53 ± 0.43 e	0.99 ± 0.026 g	9.27 ± 0.18 f	16.11 ± 0.24 e	10.85 ± 0.25 f	7.48 ± 0.29 g	0.81 ± 0.035 e
	W1N2	22.26 ± 0.60 cd	1.04 ± 0.050 fg	10.74 ± 0.41 e	17.85 ± 0.71 de	12.76 ± 0.38 e	9.29 ± 0.31 ef	0.86 ± 0.032 de
	W1N3	24.19 ± 0.31 bc	1.15 ± 0.046 ef	11.35 ± 0.13 e	18.83 ± 0.23 cd	13.71 ± 0.48 de	10.08 ± 0.20 de	0.89 ± 0.025 cde
	W2N0	21.08 ± 0.64 de	1.17 ± 0.017 e	8.08 ± 0.17 g	17.39 ± 0.70 de	12.99 ± 0.36 de	8.53 ± 0.29 fg	1.06 ± 0.046 ab
	W2N1	23.74 ± 0.93 bc	1.31 ± 0.021 cd	9.94 ± 0.22 f	20.08 ± 0.36 bc	14.04 ± 0.41 cd	11.07 ± 0.38 cd	1.11 ± 0.026 a
	W2N2	25.82 ± 0.87 b	1.49 ± 0.039 c	12.17 ± 0.13 d	21.14 ± 0.85 ab	15.17 ± 0.31 b	12.94 ± 0.28 bc	1.06 ± 0.035 ab
	W2N3	26.03 ± 0.64 b	1.63 ± 0.046 b	14.15 ± 0.37 c	22.08 ± 0.56 ab	16.02 ± 0.40 ab	13.55 ± 0.42 b	0.96 ± 0.052 cd
	W3N0	22.10 ± 0.81 cd	1.26 ± 0.035 de	9.36 ± 0.33 f	18.26 ± 0.71 cd	13.89 ± 0.46 d	10.39 ± 0.55 de	1.11 ± 0.044 a
	W3N1	24.38 ± 0.67 bc	1.40 ± 0.061 cd	12.28 ± 0.30 d	20.93 ± 0.75 ab	15.00 ± 0.08 bc	12.13 ± 0.43 bc	0.99 ± 0.031 bc
	W3N2	28.55 ± 0.98 a	1.71 ± 0.057 ab	15.14 ± 0.25 b	21.74 ± 0.64 ab	16.58 ± 0.26 a	14.08 ± 0.64 ab	0.93 ± 0.035 cd
	W3N3	29.17 ± 1.23 a	1.82 ± 0.054 a	16.26 ± 0.10 a	22.75 ± 0.86 a	16.93 ± 0.51 a	14.76 ± 0.47 a	0.91 ± 0.029 cde
Average	W1N0	18.43 ± 0.27 i	0.72 ± 0.007 h	6.97 ± 0.19 i	14.55 ± 0.28 g	8.76 ± 0.10 g	5.78 ± 0.15 h	0.83 ± 0.018 d
	W1N1	20.91 ± 0.48 gh	0.98 ± 0.022 g	9.04 ± 0.22 g	16.95 ± 0.34 f	10.86 ± 0.21 f	7.04 ± 0.19 g	0.78 ± 0.035 d
	W1N2	23.27 ± 0.36 fg	1.08 ± 0.032 fg	10.53 ± 0.28 de	18.25 ± 0.47 ef	12.86 ± 0.53 e	8.94 ± 0.22 ef	0.85 ± 0.023 d
	W1N3	25.38 ± 0.67 de	1.17 ± 0.041 f	11.18 ± 0.43 cd	18.92 ± 0.52 de	14.14 ± 0.35 c	9.64 ± 0.13 e	0.86 ± 0.025 d
	W2N0	21.63 ± 0.57 gh	1.16 ± 0.027 f	8.24 ± 0.15 h	17.04 ± 0.41 f	13.07 ± 0.62 de	8.65 ± 0.26 f	1.05 ± 0.028 a
	W2N1	24.81 ± 0.80 def	1.37 ± 0.035 de	10.08 ± 0.25 ef	19.75 ± 0.60 cd	14.23 ± 0.58 c	10.82 ± 0.18 d	1.07 ± 0.047 a
	W2N2	26.78 ± 0.92 cd	1.52 ± 0.030 bc	12.35 ± 0.18 b	20.38 ± 0.53 bc	15.58 ± 0.42 b	12.41 ± 0.30 bc	1.00 ± 0.029 ab
	W2N3	27.46 ± 0.58 bc	1.60 ± 0.044 b	13.54 ± 0.39 b	21.52 ± 0.36 ab	16.04 ± 0.39 ab	13.03 ± 0.56 b	0.96 ± 0.030 bc
	W3N0	23.51 ± 0.73 efg	1.28 ± 0.025 e	9.49 ± 0.28 fg	18.42 ± 0.74 de	13.90 ± 0.33 cd	10.41 ± 0.34 d	1.10 ± 0.053 a
	W3N1	25.97 ± 0.47 cd	1.47 ± 0.038 cd	11.94 ± 0.35 cd	21.35 ± 0.65 ab	15.43 ± 0.28 b	12.24 ± 0.41 c	1.03 ± 0.031 ab
	W3N2	29.29 ± 0.83 ab	1.72 ± 0.049 a	14.48 ± 0.51 a	22.60 ± 0.49 a	16.85 ± 0.57 a	14.13 ± 0.37 a	0.98 ± 0.028 bc
	W3N3	29.76 ± 0.69 a	1.78 ± 0.057 a	15.13 ± 0.46 a	22.57 ± 0.34 a	16.89 ± 0.45 a	14.15 ± 0.44 a	0.94 ± 0.043 c
ANOVA
Year (Y)		*	ns	ns	ns	ns	ns	***
Irrigation (W)		***	***	***	***	***	***	***
Nitrogen (N)		*	ns	ns	*	ns	ns	***
Y × W		ns	ns	ns	ns	ns	ns	**
Y × N		ns	ns	ns	ns	ns	ns	ns
W × N		ns	ns	ns	ns	ns	ns	ns
Y × W × N		ns	ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *, ** and *** indicate significant differences among different treatments at the levels of p < 0.05, p < 0.01 and p < 0.001, respectively. The ns means not significant at the level of p ≥ 0.05.

). At the same irrigation amounts, the mean values of the data showed that the plant height, leaf area index, taproot length and taproot diameter of I. tinctoria from low water to medium water showed a positive effect with significant increases of up to 13.64% to 15.91%, 40.05% to 45.08%, 8.02% to 20.90% and 25.54% to 26.70%, respectively. The increase in root dry matter accumulation was more significant than that of stem and leaf dry matter accumulation, which were 39.84% to 46.75% and 15.74% to 19.71%, respectively, and thus the root–crown ratio was increased by 21.80% to 24.24% accordingly. However, with the continuous increase in irrigation amounts, the increase in biological characters decreased significantly, which was in line with the effect of diminishing returns. The increases of plant height, leaf area index, taproot length and taproot diameter of I. tinctoria from medium water to sufficient water were 7.79% to 7.84%, 10.23% to 11.09%, 3.71% to 14.20% and 6.94% to 7.18%, respectively. There was no significant difference between dry matter accumulation in root dry matter accumulation and in stem and leaf dry matter accumulation with average increases of 15.46% and 13.40%, respectively. Although the root–crown ratio showed a decreasing trend with the increase in irrigation amounts, there was no significant change. The nitrogen application rates also significantly affected the biological characteristics of I. tinctoria. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen and from low nitrogen to medium nitrogen, the indicators of biological characteristics such as I. tinctoria plant height, leaf area index, stem and leaf dry matter accumulation, taproot diameter, and root dry matter accumulation showed a positive effect with a significant increase of up to 11.63% to 14.81%, 18.97% to 22.26%, 20.02% to 29.11%, 11.83% to 16.16%, 19.88% to 22.57% and 8.87% to 13.27%, 11.31% to 14.59%, 19.88% to 20.83%, 11.58% to 12.15%, and 16.28% to 19.03%, respectively. The taproot length significantly increased from no nitrogen to low nitrogen by 15.16% to 17.64%, but the increase from low nitrogen to medium nitrogen was not significant. From medium nitrogen to high nitrogen, although all biological characteristics indicators still showed a positive effect, the increase was not significant. The root–crown ratio showed a negative effect with the gradual increase in nitrogen application rates, but the variation was small and not a significant difference.

3.2. Effects of Water–Nitrogen Interaction on Photosynthetic Characteristics of I. tinctoria

3.2.1. Effects of Water–Nitrogen Interaction on Net Photosynthetic Rate of I. tinctoria

The net photosynthetic rate (Pn) of I. tinctoria showed a first increasing and then decreasing trend with the passage of growth period, reaching a maximum at the fleshy root growth stage (Figure 4). At the same irrigation amounts, the mean values of the data showed that from low water to medium water, Pn significantly increased by 9.25% to 13.47%, 8.32% to 18.45% and 26.88% to 29.36% at seedling, vegetative and fleshy root maturity stages, respectively, whereas the fleshy root growth stage did not show any significant difference. From medium water to sufficient water, Pn increased by 11.77% to 19.46% and 8.47% to 20.95% only at the seedling and fleshy root maturity stages, while no significant changes in Pn were observed during the vegetative and fleshy root growth stages. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen, the Pn significantly increased by 10.81 to 17.92% and 11.85 to 21.76% at the seedling and fleshy root maturity stages. In contrast, Pn increased only by 8.93 to 9.73% and 7.27 to 8.28% during the vegetative and fleshy root growth stages with no significant changes. From low nitrogen to medium nitrogen, Pn was significantly increased by 10.20% to 15.27%, 10.18% to 11.35% and 18.32% to 23.76% during the seedling, vegetative and fleshy root maturity stages, while there was no significant change in Pn during the fleshy root growth stage. From medium nitrogen to high nitrogen, there was no significant change in Pn in all the growth stages of I. tinctoria, which indicated that increasing nitrogen application rates could promote photosynthesis of I. tinctoria leaves and significantly increase Pn. But with the increasing rates of nitrogen application, Pn gradually tended to be stable. Based on the analysis of the mean values of the 3-year Pn data, it can be seen that under the water–nitrogen interaction, the W3N3 of I. tinctoria had the highest Pn at all growth stages, followed by the W3N2, and there was no significant difference between the two treatments, whose values were significantly higher than those of the other treatments. It was also found that Pn of I. tinctoria at each growth stage was different in response to the single factor and interaction of water and nitrogen, with the fleshy root maturity stage being the most sensitive, followed by the seedling and vegetative stages, while the fleshy root growth stage showed less variation.

3.2.2. Effects of Water–Nitrogen Interaction on Stomatal Conductance of I. tinctoria

The stomatal conductance (Gs) of I. tinctoria showed a continuous increasing trend with the passage of growth period, reaching a maximum at the fleshy root maturity stage (Figure 5). At the same irrigation amounts, the mean values of the data showed that from low water to medium water, and medium water to sufficient water, the Gs significantly increased by 12.47% to 28.22%, 17.50% to 28.40%, 11.91% to 13.84%, 16.32% to 21.40%, 9.65% to 13.68%, and 8.20% to 9.68% at the seedling, vegetative, and fleshy root growth stages, respectively. In contrast, the fleshy root maturity stage did not show significant differences. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen, the Gs significantly increased by 10.57% to 14.65%, 28.36% to 34.46%, 15.73% to 18.02%, and 10.05% to 12.26% at the seedling, vegetative, fleshy root growth, and fleshy root maturity stages. From low nitrogen to medium nitrogen, the Gs significantly increased at the seedling and vegetative stages by 14.36% to 16.94%, 10.63% to 12.15%, while the Gs only increased by 5.44% to 8.51%, 5.46% to 7.96% during the fleshy root growth and fleshy root maturity stages with no significant changes. From medium nitrogen to high nitrogen, there was no significant change in the Gs of I. tinctoria in all the growth stages, which indicated that increased nitrogen application rates could promote the stomatal opening of I. tinctoria leaves, which could significantly increase the Gs. But with increasing nitrogen application rates, the Gs tended to gradually stabilize. Based on the analysis of the mean values of the 3-year Gs data, it can be seen that under the water–nitrogen interaction, the W3N3 of I. tinctoria had the highest Gs at each growth stage, followed by the W3N2, and there was no significant difference between the two treatments, which was significantly higher than the other treatments. It was also found that the Gs of I. tinctoria at each growth stage was different in response to the single factor and interaction of water and nitrogen with the vegetative stage being the most sensitive, followed by the fleshy root growth and seedling stages, while the fleshy root maturity stage showed less variation.

3.2.3. Effects of Water–Nitrogen Interaction on Transpiration Rate of I. tinctoria

The transpiration rate (Tr) of I. tinctoria showed a first increasing and then decreasing trend with the passage of growth period, reaching a maximum at the vegetative stage (Figure 6). At the same irrigation amounts, the mean values of the data showed that from low water to medium water, the Tr values at fleshy root growth and fleshy root maturity stages were significantly increased by 10.90% to 14.26% and 12.82% to 25.57%, respectively, while the Tr values at the seedling and vegetative stages increased by only 1.84% to 2.74% and 6.41% to 9.09%, respectively, with no significant change. From medium water to sufficient water, the Tr at the fleshy root maturity stage was significantly increased by 12.83% to 17.73%, while no significant difference was shown at other growth stages. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen, the Tr values at the seedling and fleshy root maturity stages were significantly increased by 11.20% to 12.10% and 13.77% to 17.27%, whereas the Tr values at the vegetative and fleshy root growth stages were increased by only 7.21% to 8.06% and 5.32% to 7.45%, which showed no significant change. From low nitrogen to medium nitrogen, only the Tr value of the fleshy root maturity stage was significantly increased by 9.06% to 10.30%, while the Tr values of the other growth stages did not show any significant difference. From medium nitrogen to high nitrogen, the Tr values of all the growth stages of I. tinctoria did not have any significant change, which indicated that the increased rates of nitrogen application could promote the transpiration of I. tinctoria leaves and significantly increase the Tr, but with the continuous increase in nitrogen, the Tr gradually tended to stabilize. Based on the analysis of the mean values of the 3-year Tr data, it can be seen that under the water–nitrogen interaction, the W3N3 had the highest Tr at each growth stage, followed by the W3N2, and there was no significant difference between the two treatments, which was higher than that of the other treatments. At the same time, it was found that the response degree of Tr to water and nitrogen single factor and interaction was different at each growth stage of I. tinctoria, among which the vegetative stage was the most sensitive, followed by the seedling stage, while the fleshy root growth stage and fleshy root maturity stage had smaller variations.

3.3. Effects of Water–Nitrogen Interaction on Quality of I. tinctoria

The results of the variance by irrigation amounts and nitrogen application rates showed that irrigation amounts had a highly significant effect on I. tinctoria indigo, indirubin, (R,S)–goitrin, total nucleoside, uridine and adenosin, and the rates of nitrogen application had a significant or extremely significant effect on I. tinctoria indigo, indirubin, (R,S)–goitrin, and total nucleosides. In contrast, the water–nitrogen interaction had no significant effect on the quality of I. tinctoria (

Table 4. Effects of different irrigation amounts and nitrogen application rates on quality of I. tinctoria (mean ± standard error). Different lowercase letters within columns indicate significant differences among treatments at p < 0.05.

Year	Treatment	Indigo (mg kg−1)	Indirubin (mg kg−1)	(R,S)–Goitrin (mg g−1)	Total Nucleoside (%)	Uridine (%)	Adenosine (%)
2021	W1N0	5.78 ± 0.24 c	8.04 ± 0.18 b	0.206 ± 0.008 f	0.1580 ± 0.0075 c	0.0436 ± 0.0019 f	0.0203 ± 0.0009 i
	W1N1	5.84 ± 0.17 c	8.48 ± 0.26 ab	0.218 ± 0.013 ef	0.1663 ± 0.0062 abc	0.0457 ± 0.0023 ef	0.0318 ± 0.0015 h
	W1N2	5.93 ± 0.15 c	8.83 ± 0.26 ab	0.244 ± 0.010 cd	0.1712 ± 0.0083 abc	0.0484 ± 0.0030 ef	0.0442 ± 0.0027 de
	W1N3	6.11 ± 0.23 c	8.92 ± 0.21 ab	0.253 ± 0.015 bc	0.1774 ± 0.0091 abc	0.0508 ± 0.0017 cde	0.0491 ± 0.0023 c
	W2N0	6.10 ± 0.07 c	8.76 ± 0.11 ab	0.230 ± 0.009 de	0.1670 ± 0.0055 abc	0.0502 ± 0.0026 de	0.0403 ± 0.0019 f
	W2N1	6.36 ± 0.19 bc	9.19 ± 0.10 a	0.257 ± 0.011 abc	0.1801 ± 0.0102 abc	0.0539 ± 0.0021 bcd	0.0474 ± 0.0024 cd
	W2N2	6.83 ± 0.24 ab	9.27 ± 0.28 a	0.266 ± 0.014 ab	0.1894 ± 0.0096 a	0.0581 ± 0.0033 ab	0.0512 ± 0.0030 bc
	W2N3	6.97 ± 0.05 a	9.35 ± 0.20 a	0.271 ± 0.013 a	0.1902 ± 0.0081 a	0.0597 ± 0.0024 a	0.0545 ± 0.0018 a
	W3N0	5.97 ± 0.17 c	8.69 ± 0.40 ab	0.220 ± 0.011 ef	0.1629 ± 0.0072 bc	0.0482 ± 0.0030	0.0367 ± 0.0026 g
	W3N1	6.02 ± 0.21 c	8.85 ± 0.38 ab	0.228 ± 0.008 de	0.1758 ± 0.0100 abc	0.0514 ± 0.0026 cde	0.0435 ± 0.0024 ef
	W3N2	6.25 ± 0.14 bc	9.03 ± 0.34 a	0.249 ± 0.012 bc	0.1829 ± 0.0094 ab	0.0530 ± 0.0028 bcd	0.0480 ± 0.0021 cd
	W3N3	6.36 ± 0.24 bc	9.11 ± 0.36 a	0.262 ± 0.013 ab	0.1873 ± 0.0068 ab	0.0557 ± 0.0022 abc	0.0526 ± 0.0026 ab
2022	W1N0	5.49 ± 0.14 d	7.96 ± 0.17 c	0.213 ± 0.008 g	0.1493 ± 0.0068 f	0.0440 ± 0.0018 f	0.0286 ± 0.0012 g
	W1N1	5.62 ± 0.20 d	8.27 ± 0.15 bc	0.227 ± 0.011 f	0.1629 ± 0.0084 e	0.0468 ± 0.0021 ef	0.0372 ± 0.0019 f
	W1N2	5.78 ± 0.04 cd	8.64 ± 0.22 abc	0.249 ± 0.014 cd	0.1736 ± 0.0092 cd	0.0497 ± 0.0026 de	0.0451 ± 0.0025 cd
	W1N3	6.04 ± 0.024 bcd	9.01 ± 0.25 ab	0.260 ± 0.010 bc	0.1800 ± 0.0079 bc	0.0521 ± 0.0023 c	0.0486 ± 0.0020 bc
	W2N0	6.23 ± 0.026 bc	8.85 ± 0.14 ab	0.241 ± 0.009 de	0.1724 ± 0.0083 d	0.0516 ± 0.0028 cd	0.0411 ± 0.0017 e
	W2N1	6.41 ± 0.07 ab	9.20 ± 0.36 a	0.260 ± 0.012 bc	0.1850 ± 0.0095 b	0.0541 ± 0.0027 b	0.0459 ± 0.0024 c
	W2N2	6.77 ± 0.10 a	9.32 ± 0.32 a	0.279 ± 0.014 a	0.1912 ± 0.0102 a	0.0572 ± 0.0031 a	0.0507 ± 0.0028 ab
	W2N3	6.86 ± 0.08 a	9.41 ± 0.24 a	0.286 ± 0.013 a	0.1973 ± 0.0099 a	0.0583 ± 0.0029 a	0.0538 ± 0.0022 a
	W3N0	5.75 ± 0.20 cd	8.62 ± 0.36 abc	0.226 ± 0.008 f	0.1655 ± 0.0083 e	0.0477 ± 0.0024 e	0.0383 ± 0.0016 f
	W3N1	5.94 ± 0.09 bcd	8.91 ± 0.24 ab	0.231 ± 0.010 ef	0.1729 ± 0.0067 cd	0.0508 ± 0.0015 cd	0.0421 ± 0.0024 de
	W3N2	6.21 ± 0.25 bc	9.07 ± 0.18 ab	0.252 ± 0.013 cd	0.1834 ± 0.0082 b	0.0535 ± 0.0026 bc	0.0469 ± 0.0020 c
	W3N3	6.40 ± 0.17 ab	9.23 ± 0.33 a	0.273 ± 0.016 ab	0.1885 ± 0.0112 ab	0.0560 ± 0.0028 ab	0.0510 ± 0.0018 a
2023	W1N0	5.52 ± 0.16 e	7.79 ± 0.26 d	0.209 ± 0.008 g	0.1526 ± 0.0069 f	0.0433 ± 0.0019 f	0.0246 ± 0.0009 h
	W1N1	5.69 ± 0.04 e	8.11 ± 0.24 cd	0.218 ± 0.011 fg	0.1631 ± 0.0077 e	0.0463 ± 0.0023 e	0.0335 ± 0.0017 g
	W1N2	5.85 ± 0.15 de	8.82 ± 0.19 abc	0.242 ± 0.014 e	0.1741 ± 0.0085 c	0.0486 ± 0.0027 de	0.0433 ± 0.0024 de
	W1N3	6.14 ± 0.13 cd	9.13 ± 0.26 ab	0.258 ± 0.010 de	0.1820 ± 0.0093 b	0.0511 ± 0.0026 cd	0.0468 ± 0.0029 cd
	W2N0	6.17 ± 0.03 cd	9.10 ± 0.17 ab	0.248 ± 0.013 e	0.1730 ± 0.0081 cd	0.0514 ± 0.0026 c	0.0418 ± 0.0023 e
	W2N1	6.45 ± 0.20 abc	9.31 ± 0.25 ab	0.263 ± 0.012 cd	0.1829 ± 0.0095 b	0.0545 ± 0.0030 b	0.0453 ± 0.0021 cd
	W2N2	6.68 ± 0.11 ab	9.49 ± 0.18 ab	0.275 ± 0.014 bc	0.1933 ± 0.0108 ab	0.0569 ± 0.0028 ab	0.0500 ± 0.0028 ab
	W2N3	6.79 ± 0.12 a	9.56 ± 0.22 a	0.291 ± 0.015 a	0.1951 ± 0.0088 a	0.0585 ± 0.0032 a	0.0529 ± 0.0025 a
	W3N0	5.86 ± 0.09 de	8.70 ± 0.31 bc	0.229 ± 0.011 f	0.1693 ± 0.0082 d	0.0473 ± 0.0024 de	0.0379 ± 0.0019 f
	W3N1	6.09 ± 0.08 cd	8.86 ± 0.29 ab	0.242 ± 0.012 e	0.1734 ± 0.0078 c	0.0499 ± 0.0029 d	0.0442 ± 0.0024 cd
	W3N2	6.18 ± 0.17 cd	8.95 ± 0.20 ab	0.266 ± 0.013 c	0.1818 ± 0.0090 b	0.0527 ± 0.0026 bc	0.0474 ± 0.0020 bc
	W3N3	6.33 ± 0.09 bc	9.30 ± 0.33 ab	0.280 ± 0.017 ab	0.1891 ± 0.0094 b	0.0552 ± 0.0028 ab	0.0513 ± 0.0026 a
Average	W1N0	5.60 ± 0.17 f	7.93 ± 0.23 d	0.209 ± 0.008 e	0.1533 ± 0.0069 d	0.0436 ± 0.0015 f	0.0245 ± 0.0011 h
	W1N1	5.72 ± 0.09 ef	8.29 ± 0.19 cd	0.221 ± 0.013 d	0.1641 ± 0.0091 b	0.0463 ± 0.0020 e	0.0342 ± 0.0019 g
	W1N2	5.85 ± 0.11 de	8.76 ± 0.20 abc	0.245 ± 0.016 c	0.1730 ± 0.0075 b	0.0489 ± 0.0024 de	0.0442 ± 0.0024 d
	W1N3	6.10 ± 0.15 cd	9.02 ± 0.17 abc	0.257 ± 0.010 bc	0.1798 ± 0.0084 ab	0.0513 ± 0.0025 c	0.0482 ± 0.0025 bc
	W2N0	6.17 ± 0.10 cd	8.90 ± 0.14 abc	0.240 ± 0.012 cd	0.1708 ± 0.0092 bc	0.0511 ± 0.0019 c	0.0411 ± 0.0018 e
	W2N1	6.41 ± 0.14 bc	9.23 ± 0.24 ab	0.260 ± 0.017 bc	0.1827 ± 0.0067 a	0.0542 ± 0.0027 ab	0.0462 ± 0.0023 cd
	W2N2	6.76 ± 0.16 ab	9.36 ± 0.27 ab	0.273 ± 0.014 ab	0.1913 ± 0.0099 a	0.0574 ± 0.0033 a	0.0506 ± 0.0027 ab
	W2N3	6.87 ± 0.07 a	9.44 ± 0.34 a	0.283 ± 0.014 a	0.1942 ± 0.0086 a	0.0588 ± 0.0029 a	0.0537 ± 0.0019 a
	W3N0	5.86 ± 0.15 de	8.67 ± 0.25 bc	0.225 ± 0.011 d	0.1659 ± 0.0072 b	0.0477 ± 0.0023	0.0376 ± 0.0020 f
	W3N1	6.02 ± 0.13 cde	8.87 ± 0.29 abc	0.234 ± 0.015 cd	0.1740 ± 0.0061 b	0.0507 ± 0.0026 cd	0.0433 ± 0.0024 d
	W3N2	6.21 ± 0.14 cd	9.02 ± 0.22 abc	0.256 ± 0.013 bc	0.1827 ± 0.0087 a	0.0531 ± 0.0015 bc	0.0474 ± 0.0028 bc
	W3N3	6.36 ± 0.20 bc	9.21 ± 0.33 ab	0.272 ± 0.016 ab	0.1883 ± 0.0103 a	0.0556 ± 0.0030 ab	0.0516 ± 0.0031 ab
ANOVA
Year (Y)		ns	ns	ns	ns	ns	ns
Irrigation (W)		***	***	***	***	***	***
Nitrogen (N)		***	***	**	**	ns	ns
Y × W		ns	ns	ns	ns	ns	ns
Y × N		ns	ns	ns	ns	ns	ns
W × N		ns	ns	ns	ns	ns	ns
Y × W × N		ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The ** and *** indicate significant differences among different treatments at the levels of p < 0.01 and p < 0.001, respectively. The ns means not significant at the level of p ≥ 0.05.

). At the same irrigation amounts, the mean values of the data showed that from low water to medium water, the I. tinctoria indigo, (R,S)–goitrin, total nucleoside, uridine and adenosine were significantly increased by 10.99% to 14.57%, 11.18% to 16.18%, 8.00% to 12.03%, 14.85% to 17.72%, and 20.06% to 33.01%, respectively, while there was no significant change in indirubin. From medium water to sufficient water, the quality of I. tinctoria decreased slightly, but there was no significant difference. It can be seen that increasing the irrigation amounts can significantly improve the quality of I. tinctoria, but excessive irrigation not only leads to the waste of water resources but also causes a decline of I. tinctoria quality. Among the quality indexes of I. tinctoria, only adenosine increased with increasing nitrogen application rates. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen and low nitrogen to medium nitrogen, the adenosine of I. tinctoria significantly increased by 15.93% to 26.10% and 13.98% to 16.87%, whereas there was no significant change from medium nitrogen to high nitrogen with only a 7.32% to 8.93% increase and without significant change. Although the I. tinctoria indigo, (R,S)–goitrin, total nucleoside and uridine increased with increasing nitrogen application rates, the increase was small and did not show any significant change. It can be seen that the quality of I. tinctoria was improved with increasing nitrogen application rates, but the effect did not show any significant difference. Under the water–nitrogen interaction, in terms of the analysis of the mean values of the three experimental years, the W2N3 showed the best quality, with indigo, indirubin, (R,S)–goitrin, total nucleoside, uridine and adenosine up to 6.87 mg kg⁻¹, 9.44 mg kg⁻¹, 0.283 mg g⁻¹, 0.1942%, 0.0588%, 0.0537%, respectively, which was not significantly different from the N3W3. It can be seen that choosing a reasonable combination of water and nitrogen can not only reduce the amount of water and nitrogen input but also obtain better quality.

3.4. Effects of Water–Nitrogen Interaction on Yield and Water–Nitrogen Productivity of I. tinctoria

The results of the variance through irrigation amounts and nitrogen application rates showed that irrigation amounts had a significant or extremely significant effect on yield, water productivity, nitrogen partial productivity and nitrogen agronomy use efficiency, whereas nitrogen application rates only had a significant effect on irrigation water productivity. The water–nitrogen interaction did not significantly affect I. tinctoria yield and water–nitrogen productivity (

Table 5. Effects of different irrigation amounts and nitrogen application rates on I. tinctoria yield and water–nitrogen productivity (mean ± standard error). Different lowercase letters within columns indicate significant differences among treatments at p < 0.05.

Year	Treatment	Yield (mg km−2)	Water Consumption (mm)	Water Productivity (kg m−3)	Irrigation Water Productivity (kg m−3)	Nitrogen Partial Productivity (kg kg−1)	Nitrogen Agronomy Use Efficiency (kg kg−1)
2021	W1N0	4347.32 ± 154.73 h	377.26 ± 1.72 d	1.15 ± 0.039 e	3.58 ± 0.127 fg	—	—
	W1N1	4658.46 ± 167.75 gh	385.35 ± 11.46 cd	1.21 ± 0.010 e	3.83 ± 0.140 ef	31.06 ± 1.12 e	2.07 ± 0.36 e
	W1N2	5720.64 ± 143.19 e	415.29 ± 5.97 bc	1.38 ± 0.043 d	4.71 ± 0.117 c	28.60 ± 0.72 e	6.87 ± 1.01 c
	W1N3	5061.92 ± 126.07 fg	447.64 ± 15.30 ab	1.13 ± 0.055 e	4.17 ± 0.104 de	20.25 ± 0.51 f	2.86 ± 0.83 de
	W2N0	5570.41 ± 256.75 ef	374.7 ± 16.30 d	1.49 ± 0.049 cd	3.78 ± 0.038 f	—	—
	W2N1	6414.40 ± 190.05 d	395.31 ± 15.75 cd	1.62 ± 0.058 c	4.35 ± 0.061 d	42.76 ± 0.60 bc	5.63 ± 0.22 cd
	W2N2	8176.20 ± 292.58 b	430.25 ± 14.13 ab	1.90 ± 0.047 b	5.54 ± 0.198 ab	40.88 ± 1.46 c	13.03 ± 1.30 b
	W2N3	8569.29 ± 168.13 ab	456.58 ± 7.31 a	1.88 ± 0.029 b	5.81 ± 0.047 a	34.28 ± 0.27 d	12.00 ± 0.18 b
	W3N0	5778.43 ± 279.54 e	384.71 ± 6.00 cd	1.50 ± 0.030 cd	3.33 ± 0.045 g	—	—
	W3N1	7320.22 ± 245.69 c	390.9 ± 6.98 cd	1.87 ± 0.067 b	4.22 ± 0.142 d	48.80 ± 1.64 a	10.28 ± 1.47 b
	W3N2	9054.68 ± 335.88 a	431.67 ± 13.29 ab	2.10 ± 0.043 a	5.22 ± 0.193 b	45.27 ± 1.68 b	16.38 ± 1.57 a
	W3N3	7341.35 ± 163.83 c	460.41 ± 7.67 a	1.59 ± 0.026 c	4.23 ± 0.093 d	29.37 ± 0.65 e	6.25 ± 0.37 c
2022	W1N0	4185.91 ± 66.61 g	355.63 ± 14.88 cd	1.18 ± 0.030 g	3.67 ± 0.056 ef	—	—
	W1N1	4560.98 ± 134.07 fg	363.84 ± 11.65 cd	1.25 ± 0.003 g	4.00 ± 0.117 de	30.41 ± 0.90 de	2.50 ± 0.45 e
	W1N2	5613.88 ± 175.52 de	390.8 ± 5.62 bc	1.44 ± 0.045 f	4.92 ± 0.152 b	28.07 ± 0.88 e	7.14 ± 0.54 c
	W1N3	4809.89 ± 148.89 f	421.96 ± 11.07 ab	1.14 ± 0.038 g	4.22 ± 0.132 cd	19.24 ± 0.59 f	2.50 ± 0.51 e
	W2N0	5322.75 ± 184.46 e	350.41 ± 12.91 d	1.52 ± 0.032 ef	3.84 ± 0.062 de	—	—
	W2N1	6070.93 ± 286.98 d	371.44 ± 9.90 cd	1.63 ± 0.023 cde	4.38 ± 0.098 c	40.47 ± 0.58 b	4.99 ± 0.61 d
	W2N2	7206.97 ± 271.15 b	408.33 ± 10.36 ab	1.77 ± 0.079 bc	5.20 ± 0.194 b	36.03 ± 1.36 c	9.42 ± 0.96 b
	W2N3	7978.58 ± 236.80 a	434.8 ± 8.48 a	1.84 ± 0.019 b	5.76 ± 0.170 a	31.91 ± 0.95 d	10.62 ± 0.93 b
	W3N0	5528.80 ± 172.24 e	358.09 ± 6.71 cd	1.54 ± 0.037 def	3.39 ± 0.109 f	—	—
	W3N1	6634.87 ± 133.16 c	364.47 ± 13.96 cd	1.82 ± 0.083 b	4.07 ± 0.081 cd	44.23 ± 0.89 a	7.37 ± 0.77 c
	W3N2	8066.79 ± 182.53 a	406.72 ± 13.21 ab	1.98 ± 0.022 a	4.95 ± 0.113 b	40.33 ± 0.92 b	12.69 ± 0.09 a
	W3N3	7206.98 ± 177.14 b	436.25 ± 12.75 a	1.65 ± 0.034 cd	4.42 ± 0.137 c	28.83 ± 0.71 e	6.71 ± 0.02 cd
2023	W1N0	3784.23 ± 108.16 g	317.63 ± 12.70 e	1.19 ± 0.053 f	2.49 ± 0.072 e	—	—
	W1N1	4397.14 ± 151.72 f	325.83 ± 8.99 e	1.35 ± 0.025 e	2.90 ± 0.098 d	29.31 ± 1.01 de	4.09 ± 0.73 e
	W1N2	5340.50 ± 161.30 e	357.5 ± 6.49 cd	1.49 ± 0.049 d	3.52 ± 0.108 c	26.70 ± 0.80 e	7.78 ± 0.82 cd
	W1N3	5127.17 ± 178.12 e	391.83 ± 9.81 ab	1.31 ± 0.025 e	3.38 ± 0.117 c	20.51 ± 0.71 f	5.37 ± 0.37 de
	W2N0	4753.07 ± 44.63 ef	322.39 ± 5.73 e	1.47 ± 0.032 d	2.58 ± 0.125 e	—	—
	W2N1	5958.07 ± 141.65 d	342.84 ± 2.52 de	1.74 ± 0.019 c	3.23 ± 0.122 c	39.72 ± 0.28 c	8.03 ± 0.02 cd
	W2N2	7432.37 ± 232.40 c	375.78 ± 13.19 bc	1.98 ± 0.031 b	4.04 ± 0.138 b	37.16 ± 1.16 c	13.40 ± 1.18 b
	W2N3	8159.68 ± 221.21 b	400.45 ± 5.59 ab	2.04 ± 0.042 b	4.43 ± 0.142 a	32.64 ± 0.88 d	13.63 ± 0.77 b
	W3N0	5137.90 ± 150.91 e	339.57 ± 11.70 de	1.51 ± 0.033 d	2.37 ± 0.169 e	—	—
	W3N1	7226.68 ± 317.34 c	345.28 ± 4.83 de	2.09 ± 0.067 b	3.33 ± 0.147 c	48.18 ± 2.12 a	13.93 ± 1.79 b
	W3N2	8806.15 ± 205.83 a	384.66 ± 8.12 b	2.29 ± 0.016 a	4.06 ± 0.093 b	44.03 ± 1.03 b	18.34 ± 0.66 a
	W3N3	7442.72 ± 277.98 c	412.73 ± 5.01 a	1.80 ± 0.049 c	3.43 ± 0.127 c	29.77 ± 1.11 de	9.22 ± 0.63 c
Average	W1N0	4105.82 ± 194.34 g	350.17 ± 9.49 e	1.17 ± 0.019 g	3.25 ± 0.075 fg	—	—
	W1N1	4538.86 ± 145.76 g	358.34 ± 10.66 e	1.27 ± 0.014 f	3.58 ± 0.116 e	30.26 ± 0.97 e	2.89 ± 0.46 e
	W1N2	5558.34 ± 114.67 e	387.86 ± 4.80 cd	1.44 ± 0.020 e	4.38 ± 0.085 c	27.79 ± 0.57 e	7.26 ± 0.11 d
	W1N3	4999.66 ± 141.23 f	420.48 ± 10.53 ab	1.19 ± 0.038 g	3.92 ± 0.108 d	20.00 ± 0.63 f	3.58 ± 0.49 e
	W2N0	5215.41 ± 350.49 ef	349.17 ± 11.03 e	1.49 ± 0.032 de	3.40 ± 0.036 ef	—	—
	W2N1	6147.80 ± 270.23 d	369.86 ± 9.27 de	1.66 ± 0.025 c	3.99 ± 0.049 d	40.99 ± 0.49 b	6.22 ± 0.23 d
	W2N2	7605.18 ± 195.78 b	404.79 ± 10.74 bc	1.88 ± 0.017 b	4.93 ± 0.128 b	38.03 ± 0.98 c	11.95 ± 0.73 b
	W2N3	8235.85 ± 166.35 a	430.61 ± 6.01 ab	1.92 ± 0.023 b	5.33 ± 0.107 a	32.94 ± 0.67 d	12.08 ± 0.50 b
	W3N0	5481.71 ± 129.76 e	360.79 ± 9.04 de	1.52 ± 0.013 d	3.03 ± 0.073 g	—	—
	W3N1	7060.59 ± 165.93 c	366.88 ± 8.47 de	1.93 ± 0.032 b	3.87 ± 0.090 d	47.07 ± 1.11 a	10.53 ± 0.24 c
	W3N2	8642.54 ± 210.21 a	407.68 ± 11.13 bc	2.12 ± 0.029 a	4.74 ± 0.119 b	43.21 ± 1.05 b	15.80 ± 0.42 a
	W3N3	7330.35 ± 202.35 bc	436.46 ± 7.32 a	1.68 ± 0.018 c	4.03 ± 0.134 d	29.32 ± 0.81 e	7.39 ± 0.31 d
ANOVA
Year (Y)		ns	***	*	***	ns	**
Irrigation (W)		***	ns	***	**	***	***
Nitrogen (N)		ns	ns	ns	ns	ns	ns
Y×W		ns	ns	ns	ns	ns	ns
Y×N		ns	ns	ns	ns	ns	ns
W×N		ns	ns	ns	ns	ns	ns
Y×W×N		ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *, **, and *** indicate significant differences among different treatments at the levels of p < 0.05, p < 0.01, and p < 0.001, respectively. The ns means not significant at the level of p ≥ 0.05.

). At the same irrigation amounts, the mean values of the data showed that from low water to medium water, the yield, water productivity, irrigation water productivity, nitrogen partial productivity and nitrogen agronomic use efficiency of I. tinctoria significantly increased by 38.65% to 45.19%, 34.88% to 41.42%, 14.18% to 19.57%, 39.51% to 47.57%, and 103.35% to 159.77%, respectively. From medium water to sufficient water, there was no significant change in yield, water productivity, nitrogen partial productivity, and nitrogen agronomic use efficiency, while the irrigation water productivity significantly decreased by 7.53% to 12.74%. It can be seen that increasing irrigation amounts could significantly increase I. tinctoria yield, as well as promote nitrogen uptake, and improve nitrogen partial productivity and nitrogen agronomic use efficiency. However, more irrigation amounts would result in a decrease in irrigation water productivity, but the water consumption did not increase with increasing irrigation amounts. The yield and water–nitrogen productivity of I. tinctoria were also significantly affected by nitrogen application rates. At the same nitrogen application rates, the mean values of the data showed that from no nitrogen to low nitrogen, the yield, water productivity, and irrigation water productivity significantly increased by 14.83% to 28.57%, 11.05% to 23.97%, and 14.19% to 27.16%. From low nitrogen to medium nitrogen, it was significantly increased by 20.97% to 24.78%, 10.12% to 14.27%, and 21.05% to 24.74%. From medium nitrogen to high nitrogen, the I. tinctoria yield and irrigation water productivity showed a decreasing trend, but there was no significant change; while water productivity was significantly reduced by 10.62% to 14.38%, the water consumption also showed a stable trend and did not increase with the increase in nitrogen application rates. From low nitrogen to medium nitrogen, nitrogen partial productivity decreased only by 6.41% to 9.27%, with no significant change, but it decreased significantly by 23.15% to 26.90% from medium nitrogen to high nitrogen. From low nitrogen to medium nitrogen, the nitrogen agronomic use efficiency increased significantly by 51.74% to 101.77%, but it decreased significantly from medium to high nitrogen by 28.60% to 41.82%. It can be seen that the I. tinctoria yield and water productivity were positively correlated with increasing nitrogen application rates, but they were negatively correlated after exceeding a certain threshold. In addition, 250 kg ha⁻¹ urea applications leading to excessive nitrogen also resulted in the decrease in nitrogen partial productivity and nitrogen agronomic use efficiency, which led to nitrogen losses. Under the water–nitrogen interaction, according to the analysis of the mean values of the three experimental years, the maximum of I. tinctoria yield, water productivity and nitrogen agronomic use efficiency appeared in the W3N2, which could reach 8642.54 kg ha⁻¹, 2.12 kg m⁻³, 15.80 kg kg⁻¹, respectively with significant increases of 17.90%, 26.22%, and 113.73% compared with the N3W3. The maximum irrigation water productivity and nitrogen partial productivity appeared in W1N3 and W3N1, which could reach 5.33 kg m⁻³ and 47.07 kg kg⁻¹, respectively, and they significantly increased by 32.36% and 60.53% compared with N3W3. It can be seen that sufficient water and high nitrogen did not produce high yield and high water–nitrogen productivity. Therefore, selecting a reasonable water–nitrogen combination is not only beneficial to improve the yield of I. tinctoria but also can obtain higher water productivity, irrigation water productivity, nitrogen partial productivity and nitrogen agronomic use efficiency.

3.5. Effects of Water–Nitrogen Interaction on Economic Benefit of I. tinctoria

The results of the variance by irrigation amounts and nitrogen application rates showed that only irrigation amounts had an extremely significant effect on the economic benefit, net proceeds, and output–input ratio of I. tinctoria, whereas nitrogen application rates and water–nitrogen interaction did not have a significant effect on those (

Table 6. Effects of different irrigation amounts and nitrogen application rates on the input cost (CNY hm⁻²).

Year	Treatment	Irrigation Water Cost	Fertilizer Cost	Seed Cost	Labor Cost	Mechanical Cost	Other Cost	Total Input
2021	W1N0	316	1692	1125	2400	3150	900	9583
	W1N1	316	1908	1125	2400	3150	900	9799
	W1N2	316	1980	1125	2400	3150	900	9871
	W1N3	316	2052	1125	2400	3150	900	9943
	W2N0	384	1692	1125	2400	3150	900	9651
	W2N1	384	1908	1125	2400	3150	900	9867
	W2N2	384	1980	1125	2400	3150	900	9939
	W2N3	384	2052	1125	2400	3150	900	10,011
	W3N0	451	1692	1125	2400	3150	900	9718
	W3N1	451	1908	1125	2400	3150	900	9934
	W3N2	451	1980	1125	2400	3150	900	10,006
	W3N3	451	2052	1125	2400	3150	900	10,078
2022	W1N0	296	1750	1350	2850	3600	675	10,521
	W1N1	296	2065	1350	2850	3600	675	10,836
	W1N2	296	2170	1350	2850	3600	675	10,941
	W1N3	296	2275	1350	2850	3600	675	11,046
	W2N0	360	1750	1350	2850	3600	675	10,585
	W2N1	360	2065	1350	2850	3600	675	10,900
	W2N2	360	2170	1350	2850	3600	675	11,005
	W2N3	360	2275	1350	2850	3600	675	11,110
	W3N0	424	1750	1350	2850	3600	675	10,649
	W3N1	424	2065	1350	2850	3600	675	10,964
	W3N2	424	2170	1350	2850	3600	675	11,069
	W3N3	424	2275	1350	2850	3600	675	11,174
2023	W1N0	394	1730	1200	2250	3225	780	9579
	W1N1	394	2090	1200	2250	3225	780	9939
	W1N2	394	2210	1200	2250	3225	780	10,059
	W1N3	394	2330	1200	2250	3225	780	10,179
	W2N0	479	1730	1200	2250	3225	780	9664
	W2N1	479	2090	1200	2250	3225	780	10,024
	W2N2	479	2210	1200	2250	3225	780	10,144
	W2N3	479	2330	1200	2250	3225	780	10,264
	W3N0	563	1730	1200	2250	3225	780	9748
	W3N1	563	2090	1200	2250	3225	780	10,108
	W3N2	563	2210	1200	2250	3225	780	10,228
	W3N3	563	2330	1200	2250	3225	780	10,348

and

Table 7. Effects of different irrigation amounts and nitrogen application rates on the economic benefit (mean ± standard error). Different lowercase letters within columns indicate significant differences among treatments at p < 0.05.

Year	Treatment	Economic Benefit (CNY hm−2)	Net Proceeds (CNY hm−2)	Output–Input Ratio
2021	W1N0	54,776.23 ± 1949.60 h	45,193.28 ± 1949.64 g	5.72 ± 0.20 f
	W1N1	58,696.60 ± 2113.65 gh	48,897.64 ± 2113.69 fg	5.99 ± 0.22 f
	W1N2	72,080.06 ± 1804.19 e	62,209.11 ± 1804.17 e	7.30 ± 0.18 e
	W1N3	63,780.19 ± 1588.48 fg	53,837.24 ± 1588.51 f	6.41 ± 0.16 f
	W2N0	70,187.17 ± 3235.05 ef	60,536.51 ± 3235.16 e	7.27 ± 0.07 e
	W2N1	80,821.44 ± 2394.63 d	70,954.78 ± 2394.70 d	8.19 ± 0.12 d
	W2N2	103,020.12 ± 3686.51 b	93,081.46 ± 3686.49 b	10.37 ± 0.37 b
	W2N3	107,973.05 ± 2118.44 ab	97,962.40 ± 2118.37 ab	10.79 ± 0.08 ab
	W3N0	72,808.22 ± 3522.20 e	63,089.86 ± 3522.28 e	7.49 ± 0.10 e
	W3N1	92,234.77 ± 3095.69 c	82,300.41 ± 3095.65 c	9.28 ± 0.31 c
	W3N2	114,088.97 ± 4232.09 a	104,082.61 ± 4232.10 a	11.40 ± 0.42 a
	W3N3	92,501.01 ± 2064.26 c	82,422.65 ± 2064.30 c	9.18 ± 0.21 c
2022	W1N0	42,696.28 ± 679.42 g	32,174.80 ± 679.45 g	4.06 ± 0.06 f
	W1N1	46,522.00 ± 1367.51 fg	35,685.52 ± 1367.55 fg	4.29 ± 0.13 f
	W1N2	57,261.58 ± 1790.30 de	46,320.10 ± 1790.31 de	5.23 ± 0.16 de
	W1N3	49,060.88 ± 1518.68 f	38,014.40 ± 1518.65 f	4.44 ± 0.14 f
	W2N0	54,292.05 ± 1881.49 e	43,707.04 ± 1881.48 e	5.13 ± 0.08 e
	W2N1	61,923.49 ± 2927.20 d	51,023.48 ± 2927.24 d	5.68 ± 0.08 d
	W2N2	73,511.09 ± 2765.73 b	62,506.09 ± 2765.77 b	6.68 ± 0.25 b
	W2N3	81,381.52 ± 2415.36 a	70,271.51 ± 2415.39 a	7.33 ± 0.22 a
	W3N0	56,393.76 ± 1756.85 e	45,745.22 ± 1756.82 e	5.30 ± 0.17 de
	W3N1	67,675.67 ± 1358.23 c	56,712.13 ± 1358.22 c	6.17 ± 0.12 c
	W3N2	82,281.26 ± 1861.81 a	71,212.72 ± 1861.84 a	7.43 ± 0.17 a
	W3N3	73,511.20 ± 1806.83 b	62,337.66 ± 1806.83 b	6.58 ± 0.16 bc
2023	W1N0	31,030.69 ± 886.91 g	21,451.29 ± 886.93 g	3.24 ± 0.09 g
	W1N1	36,056.55 ± 1244.10 f	26,117.15 ± 1244.09 f	3.63 ± 0.12 fg
	W1N2	43,792.10 ± 1322.66 e	33,732.71 ± 1322.65 e	4.35 ± 0.13 e
	W1N3	42,042.79 ± 1460.58 e	31,863.40 ± 1460.55 e	4.13 ± 0.15 e
	W2N0	38,975.17 ± 365.97 ef	29,311.27 ± 365.94 ef	4.03 ± 0.04 ef
	W2N1	48,856.17 ± 1161.53 d	38,832.27 ± 1161.56 d	4.87 ± 0.03 d
	W2N2	60,945.43 ± 1905.68 c	50,801.53 ± 1905.65 c	6.01 ± 0.19 c
	W2N3	66,909.38 ± 1813.92 b	56,645.47 ± 1813.96 b	6.52 ± 0.18 b
	W3N0	42,130.78 ± 1237.46 e	32,382.36 ± 1237.49 e	4.32 ± 0.13 e
	W3N1	59,258.78 ± 2602.19 c	49,150.36 ± 2602.21 c	5.86 ± 0.26 c
	W3N2	72,210.43 ± 1687.81 a	61,982.01 ± 1687.77 a	7.06 ± 0.17 a
	W3N3	61,030.30 ± 2279.44 c	50,681.88 ± 2279.41 c	5.90 ± 0.22 c
Average	W1N0	42,834.40 ± 1033.55 g	32,939.79 ± 1033.55 g	4.34 ± 0.11 g
	W1N1	47,091.71 ± 1525.49 g	36,900.11 ± 1525.49 g	4.64 ± 0.15 fg
	W1N2	57,711.25 ± 1109.31 e	47,420.64 ± 1109.31 e	5.63 ± 0.11 e
	W1N3	51,627.95 ± 1419.59 f	41,238.35 ± 1419.59 f	5.00 ± 0.14 f
	W2N0	54,484.80 ± 541.97 ef	44,518.27 ± 541.98 ef	5.48 ± 0.05 e
	W2N1	63,867.03 ± 762.10 d	53,603.51 ± 762.11 d	6.25 ± 0.07 d
	W2N2	79,158.88 ± 2029.49 b	68,796.36 ± 2029.49 b	7.68 ± 0.19 b
	W2N3	85,421.32 ± 1612.02 a	74,959.79 ± 1612.02 a	8.21 ± 0.15 a
	W3N0	57,110.92 ± 1279.06 e	47,072.48 ± 1279.06 e	5.70 ± 0.13 e
	W3N1	73,056.41 ± 1737.77 c	62,720.97 ± 1737.77 c	7.11 ± 0.17 c
	W3N2	89,526.89 ± 2290.67 a	79,092.45 ± 2290.67 a	8.63 ± 0.22 a
	W3N3	75,680.84 ± 2012.63 c	65,147.40 ± 2012.63 c	7.22 ± 0.19 c
ANOVA
Year (Y)		***	***	***
Irrigation (W)		***	***	***
Nitrogen (N)		ns	ns	ns
Y × W		ns	ns	ns
Y × N		ns	ns	ns
W × N		ns	ns	ns
Y × W × N		ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *** indicate significant differences among different treatments at the levels of p < 0.001, respectively. The ns means not significant at the level of p ≥ 0.05.

). At the same irrigation amounts, the mean values of the data showed that from no nitrogen to low nitrogen and from low nitrogen to medium nitrogen, the economic benefit, net proceeds and output–input ratio of I. tinctoria were significantly higher by 14.83% to 28.57%, 17.92% to 37.23%, 11.49% to 23.89%, 20.97% to 24.78%, 25.53% to 28.41%, and 19.82% to 23.88%, respectively. Even though there was a decreasing trend from medium nitrogen to high nitrogen, there was no significant difference. It could be seen that increasing nitrogen application rates within a certain range could significantly improve the economic benefit, net proceeds, and output–input ratio of I. tinctoria, but beyond the threshold of nitrogen application rates, the economic benefit, net proceeds and output–input ratio of I. tinctoria slightly decreased. At the same nitrogen application rates, the mean values of the data showed that from low water to medium water, the economic benefit, net proceeds and output–input ratio of I. tinctoria were significantly increased by 38.65% to 45.19%, 49.48% to 55.16%, and 37.66% to 44.03%, respectively. While from medium water to sufficient water, the increase was less and did not show any significant change. It could be seen that with the increase in irrigation amounts, the economic benefit, net proceeds and output–input ratio of I. tinctoria could be significantly improved, but the excessive irrigation amounts did not show any significant advantage. Under the interaction of water and nitrogen, with the analysis of the mean values of the three experimental years, the maximum economic benefit, net proceeds and output–input ratio of I. tinctoria appeared in the W3N2, which were up to 89,526.89 CNY hm⁻², 79,092.45 CNY hm⁻² and 8.63, respectively. Compared with N3W3, it was significantly increased by 18.30%, 21.41% and 19.58%, respectively. It could be seen that sufficient water and high nitrogen did not obtain high yield, so choosing a reasonable combination of water and nitrogen is not only conducive to improving the economic benefit of I. tinctoria but also obtaining higher net proceeds.

3.6. Correlation Analysis of Indicators under Water–Nitrogen Interaction

The correlation analysis results of each indicator under the water–nitrogen interaction model are shown in Figure 7. Different water–nitrogen interaction models changed the original growth environment of I. tinctoria. According the correlation of the average data of the three experiment years, the net photosynthetic rate, stomatal conductance, and transpiration rate had a significant and positive correlation (p < 0.05) with the I. tinctoria yield, leaf area index, dry matter accumulation of stems and leaves, taproot length, taproot diameter, and root dry matter accumulation, which indicated that a favorable light environment could promote the I. tinctoria growth and the dry matter accumulation. The leaf area index, stem and leaf dry matter accumulation, taproot length, taproot diameter, and root dry matter accumulation were significantly and positively correlated with the yield, economic benefit, and net proceeds (p < 0.05), suggesting that good growth conditions and dry matter accumulation of I. tinctoria plants can significantly increase the yield and then enhance the economic benefit and net proceeds. There was a significant and positive correlation (p < 0.05) between I. tinctoria yield and water productivity, irrigation water productivity, and nitrogen agronomic use efficiency, with the r correlation coefficients up to 0.974, 0.744, and 0.935, which indicated that suitable water and nitrogen environments not only produce a high yield but also obtain higher water productivity, irrigation water productivity, and nitrogen agronomic use efficiency. The yield of I. tinctoria was positively correlated with indigo, indirubin, (R,S)–goitrin, total nucleosides, uridine and adenosine with r correlation coefficients up to 0.692, 0.682, 0.629, 0.737, 0.755, 0.679, which suggested that the appropriate water and nitrogen environments can improve the quality of I. tinctoria while obtaining high yield. It was further found that there was also a significant and positive correlation (p < 0.05) among the quality indicators of I. tinctoria, suggesting that indigo, indirubin, (R,S)–goitrin, total nucleosides, uridine and adenosine have a mutually promoting effect on each other.

3.7. Establishment of a Comprehensive Growth Evaluation System for Hybrid Seed Maize

3.7.1. Comprehensive Evaluation Hierarchy Model (IHM)

A hierarchical model for the comprehensive evaluation of I. tinctoria was established using Yaaph software (Version v12.11.8293). The target layer of comprehensive growth indices (C) was divided into three guideline layers: yield index (C₁), water–nitrogen use efficiency index (C₂), and quality index (C₃); the yield indices included four layers: yield (C₁₁), production value (C₁₂), net proceeds (C₁₃), and output–input ratio (C₁₄); the water–nitrogen use efficiency indices included four layers: water productivity (C₂₁), irrigation water productivity (C₂₂), nitrogen partial productivity (C₂₃), and nitrogen agronomic use efficiency (C₂₄); the quality indices included six layers: indigo (C₃₁), indirubin (C₃₂), (R,S)–glutathione (C₃₃), total nucleosides (C₃₄), uridine (C₃₅), and adenosine (C₃₆).

3.7.2. Indicator Weights

(1) Determination of weights based on the AHP method

After the establishment of the hierarchical model, a scale of 1 to 10 was used to establish the judgement matrix and test the consistency of the matrix; the judgement matrices for the comprehensive growth index, yield index, water–nitrogen use efficiency index, and quality index are as follows:

$$\begin{array}{l}C=\left[\begin{array}{ccc}1.0000& 3.0000& 3.5000\\ 0.3333& 1.0000& 1.5000\\ 0.2857& 0.6667& 1.0000\end{array}\right]C_1=\left[\begin{array}{cccc}1.0000& 1.0000& 1.0000& 1.2000\\ 1.0000& 1.0000& 1.0000& 1.2000\\ 1.0000& 1.0000& 1.0000& 1.0000\\ 0.8333& 0.8333& 1.0000& 1.0000\end{array}\right]C_2=\left[\begin{array}{cccc}1.0000& 2.0000& 3.0000& 0.9000\\ 0.5000& 1.0000& 2.5000& 0.5000\\ 0.3333& 0.4000& 1.0000& 0.3333\\ 1.1111& 2.0000& 3.0000& 1.0000\end{array}\right]\\ C_3=\left[\begin{array}{cccccc}1.0000& 0.5000& 1.2000& 0.5000& 0.2500& 0.3333\\ 2.0000& 1.0000& 1.5000& 0.5000& 0.3333& 0.5000\\ 0.8333& 0.6667& 1.0000& 0.5000& 0.2500& 0.3333\\ 2.0000& 2.0000& 2.0000& 1.0000& 0.5000& 1.5000\\ 4.0000& 3.0000& 4.0000& 2.0000& 1.0000& 1.8000\\ 3.0000& 2.0000& 3.0000& 0.6667& 0.5556& 1.0000\end{array}\right]\end{array}$$

The consistency test coefficients C_R of the comprehensive growth index, yield index, water–nitrogen use efficiency index, and quality index were <0.10, the consistency test results were better, and the established judgement matrix was reliable and reasonable (

Table 8. Results of AHP hierarchical analysis for calculating weights.

Hierarchical Structure	Local Weight	Final Weight	Consistency Test Parameter
Target Layer C	0.6153	0.6153	CR = 0.0068 < 0.1 λmax = 3.0070
	0.2230	0.2230
	0.1617	0.1617
Criterion layer C1	0.2611	0.1607	CR = 0.0016 < 0.1 λmax = 4.0042
	0.2611	0.1607
	0.2497	0.1536
	0.2280	0.1403
Criterion layer C2	0.3396	0.0757	CR = 0.0128 < 0.1 λmax = 4.0342
	0.1997	0.0445
	0.1029	0.0229
	0.3578	0.0798
Criterion layer C3	0.0781	0.0126	CR = 0.0146 < 0.1 λmax = 6.0917
	0.1140	0.0184
	0.0765	0.0124
	0.1969	0.0318
	0.3361	0.0543
	0.1983	0.0321

, where λ_max is the maximum eigenvalue). The results showed that the weights of each index, in descending order, were yield, net proceeds, economic benefit, output–input ratio, nitrogen agronomic use efficiency, water productivity, uridine, irrigation water productivity, adenosine, total nucleoside, nitrogen partial productivity, indirubin, indigo, and (R,S)–goitrin.

(2) Determination of weights based on entropy weighting

The entropy weighting method was used to assign weights to the single index of I. tinctoria, and the weights of each index were calculated (

Table 9. Weights of single index of I. tinctoria determined by the entropy weighting method.

Index	C11	C12	C13	C14	C21	C22	C23
Weight	0.0789	0.0794	0.0799	0.0812	0.0778	0.0928	0.0617
Index	C24	C31	C32	C33	C34	C35	C36
Weight	0.0909	0.0833	0.0469	0.0597	0.0579	0.0649	0.0447

). From the table, it can be seen that the weights of each index of yield determined by the entropy weighting method, in descending order, were irrigation water productivity, nitrogen agronomic use efficiency, indigo, output–input ratio, net proceeds, economic benefit, yield, water productivity, uridine, nitrogen partial productivity, (R,S)–goitrin, total nucleosides, indirubin and adenosine.

(3) Combinatorial Empowerment Based on Game Theory

In order to improve the reliability of the weight assignment values and avoid the influence of subjective factors on the evaluation, a basic weight set was constructed based on the two assignment values obtained with the AHP and the entropy weight method $w={\displaystyle \sum _{\mathrm{k}=1}^l{\alpha }_k}\times w_k^T({\alpha }_k>0)$, where a_k is derived from the AHP method and w_k is derived from the entropy weight method.

Based on the weight set model of game theory, the game model was derived $M\mathrm{in}\Vert {\displaystyle \sum _{j=1}^i{a}_j\times u_j^T-u_i^T}\Vert$ i = 1, 2. The optimal combination coefficients of the above equation could be obtained using Matlab: a₁ = 1.1252, a₂ = −0.1252. This yields a vector of combined weights: $w^{*}={\displaystyle \sum _{k=1}^2{a}_k^{*}}\times u_k^T$, and the final results are presented in

Table 10. Single index weights for I. tinctoria assigned based on game theory.

Index	C11	C12	C13	C14	C21	C22	C23
Weight	0.1709	0.1709	0.1628	0.1477	0.0754	0.0384	0.0180
Index	C24	C31	C32	C33	C34	C35	C36
Weight	0.0784	0.0037	0.0148	0.0065	0.0285	0.0530	0.0305

. The weights of the indices of I. tinctoria decreased in the order of yield, net proceeds, economic benefit, output–input ratio, nitrogen agronomic use efficiency, water productivity, uridine, irrigation water productivity, adenosine, total nucleosides, nitrogen partial productivity, indirubin, (R,S)–goitrin, and indigo.

3.7.3. Comprehensive Evaluation of I. tinctoria Based on the TOPSIS Method

Based on the evaluation of the TOPSIS comprehensive model with combined assignment, the decision matrix was normalized, the weighting matrix was established, and then the ideal solution and the fit C_i of the evaluation index were calculated (

Table 11. Comprehensive indices of I. tinctoria based on TOPSIS method and their ranking.

Treatment	C11	C12	C13	C14	C21	C22	C23	C24	C31
W1N1	0.2219	0.2222	0.2032	0.2262	0.2485	0.2748	0.2854	0.1011	0.3043
W1N2	0.2718	0.2722	0.2612	0.2744	0.2818	0.3362	0.2621	0.2541	0.3112
W1N3	0.2445	0.2436	0.2271	0.2437	0.2328	0.3009	0.1886	0.1253	0.3245
W2N1	0.3006	0.3013	0.2952	0.3047	0.3248	0.3063	0.3866	0.2177	0.3410
W2N2	0.3719	0.3734	0.3789	0.3744	0.3678	0.3785	0.3587	0.4182	0.3596
W2N3	0.4027	0.4030	0.4128	0.4002	0.3757	0.4092	0.3107	0.4228	0.3655
W3N1	0.3452	0.3446	0.3454	0.3466	0.3776	0.2971	0.4439	0.3685	0.3202
W3N2	0.4226	0.4223	0.4356	0.4207	0.4148	0.3639	0.4075	0.5530	0.3303
W3N3	0.3584	0.3570	0.3588	0.3520	0.3287	0.3094	0.2765	0.2586	0.3383
$S^{+}$	0.4226	0.4223	0.4356	0.4207	0.4148	0.4092	0.4439	0.5530	0.3655
$S^{-}$	0.2219	0.2222	0.2032	0.2262	0.2328	0.2748	0.1886	0.1011	0.3043
Treatment	C32	C33	C34	C35	C36	$D^{+}$	$D^{-}$	Ci	Sorted
W1N1	0.3061	0.2874	0.3016	0.2909	0.2430	0.2201	0.0137	0.0586	9
W1N2	0.3234	0.3186	0.3180	0.3072	0.3140	0.1594	0.0647	0.2887	7
W1N3	0.3330	0.3342	0.3305	0.3223	0.3425	0.2031	0.0279	0.1209	8
W2N1	0.3408	0.3381	0.3358	0.3405	0.3283	0.1424	0.0854	0.3748	6
W2N2	0.3456	0.3550	0.3516	0.3606	0.3595	0.0589	0.1644	0.7363	3
W2N3	0.3485	0.3680	0.3570	0.3694	0.3815	0.0452	0.1857	0.8043	2
W3N1	0.3275	0.3043	0.3198	0.3185	0.3076	0.0885	0.1384	0.6101	4
W3N2	0.3330	0.3329	0.3358	0.3336	0.3368	0.0161	0.2196	0.9315	1
W3N3	0.3400	0.3537	0.3461	0.3493	0.3666	0.1066	0.1271	0.5438	5
$S^{+}$	0.3485	0.3680	0.3570	0.3694	0.3815	—	—	—	—
$S^{-}$	0.3061	0.2874	0.3016	0.2909	0.2430	—	—	—	—

Note: S⁺ is the ideal solution, S⁻ is the inverse ideal solution; D⁺ is the distance of each treatment from the ideal solution, and D⁻ is the distance of each treatment from the inverse ideal solution.

). As can be seen from the table, the W3N2 (sufficient water and high nitrogen) had the largest fit degree of comprehensive indices, and the comprehensive indices was optimal, followed by W2N3 (medium water and high nitrogen) and W2N2 (medium water and medium nitrogen), while W1N1 (low water and low nitrogen) had the smallest fit degree, which indicated that the comprehensive performance of I. tinctoria was the worst under the conditions of low water and low nitrogen.

3.8. Coupled Water–Nitrogen Response Modeling for Comprehensive Growth of I. tinctoria

Based on the comprehensive growth score of I. tinctoria and the amounts of water and nitrogen input, the regression model between the comprehensive growth score (y) and the coded values of irrigation amounts (x₁) and nitrogen application rates (x₂) was obtained as follows:

$$y=-0.025+1.455x_1+1.140x_2-0.851x_1^2-0.934x_2^2-0.192x_1{x}_2$$

(9)

The regression equation was tested for significance, and the correlation coefficient R² = 0.947, with a high fit degree, p = 0.001 < 0.01, indicating that the regression relationship reached a highly significant level, which proves that the regression model is reliable.

3.8.1. Single Factor Effect of Water–Nitrogen

In order to further investigate the effect of single factors on the comprehensive growth of I. tinctoria, the single-factor equations of irrigation amounts (y_w) and nitrogen application rates (y_n) were obtained by dimensionality reduction in the binary quadratic regression model.

$$y_w=-0.025+1.455x_1-0.851x_1^2$$

(10)

$$y_{\mathrm{n}}=-0.025+1.140x_1-0.934x_1^2$$

(11)

As can be seen from Figure 8, the comprehensive score of I. tinctoria increased with the increasing irrigation amounts within the design range of irrigation levels, and the effect of the nitrogen application rates on the comprehensive score of I. tinctoria was a downward parabola. Overall, the comprehensive score of I. tinctoria showed an increasing and then decreasing trend with increasing irrigation amounts or nitrogen application rates, which is consistent with the diminishing reward effect. That is, the comprehensive score showed a decreasing trend when the irrigation amounts and nitrogen application rates exceeded a certain range. The comprehensive growth scores changed gently with the increasing irrigation amounts but changed sharply with the increasing nitrogen application rates, indicating that the comprehensive growth scores were more sensitive to the change in nitrogen application rates.

3.8.2. Analysis of the Water–Nitrogen Interaction

As can be seen from Figure 9, we divided the closed zone of water–nitrogen coupling based on 90% of the maximum comprehensive score, and this closed zone appeared at the medium-sufficient irrigation level and medium fertilization application level. It can be concluded that the optimal irrigation amounts for agricultural production were 129.39 to 171.05 mm (in 2021), 145.74 to 160.50 mm (in 2022), and 193.89 to 213.52 mm (2023), respectively, and the optimal nitrogen application rates were between 190.30 and 218.27 kg ha⁻¹; that water–nitrogen combination was the most favorable for achieving the high yield and high quality of I. tinctoria.

4. Discussion

Water and nitrogen are the two major factors limiting crop growth, and different water and nitrogen supply conditions can significantly affect crop growth and development as well as yield formation [38]. The results of this study showed that low water supply caused severe drought stress, which significantly inhibited leaf area expansion as well as the stem and leaf dry matter mass accumulation of I. tinctoria; in addition, it also caused irreversible damage to stem and leaf development as well as limited root growth in the later stages. In contrast, the insufficient nitrogen supply caused by no or low nitrogen application significantly inhibited the growth of I. tinctoria plants and the accumulation of stem and leaf dry matter mass. Meanwhile, different levels of water and nitrogen supply in this study did not show significant interaction effects on I. tinctoria stem and root growth. With the aggravation of drought stress, the promotional effects of low nitrogen and no nitrogen on stems, leaves and roots dry matter mass accumulation, and roots growth of I. tinctoria gradually weakened, while there was no significant difference between the effects of medium nitrogen and high nitrogen on the growth of I. tinctoria. This indicates that improving water conditions can enhance the advantages of appropriate nitrogen supply and exert the water–fertilizer coupling effect as well as alleviate the negative effects of excessive nitrogen application and reduce soil nitrogen excess. Therefore, appropriate water and nitrogen supply can create a good growing environment for I. tinctoria. The results of this experiment showed that when the irrigation amount was 100% of the local typical irrigation quota, the nitrogen application rates of 200 to 250 kg ha⁻¹, it could promote the growth of I. tinctoria leaf area and the accumulation of stem and leaf dry matter mass as well as effectively promote the taproot growth and improve the root–crown ratio, which is consistent with the conclusions of He et al. [39], both of which showed that reasonable nitrogen management measures can help exert the effects of water–nitrogen coupling, promote water absorption by the root of I. tinctoria, enhance the growth and development of stems and leaves, and jointly achieve a high yield and high efficiency of water and fertilizer, which further proved that reasonable water and nitrogen management measures are the key ways to promote the root growth and increase the dry matter accumulation of I. tinctoria. However, the optimal water and nitrogen application rates varied slightly in this study results, which may be due to the differences in the experimental program, the period of water and nitrogen supply, and the climatic conditions of the experimental area. The results of Chun et al. [40] showed that reducing irrigation amounts and giving moderate drought stress during the crop growth period could change the root morphology, and the crops would allocate more photosynthetic products to the root system to promote root growth, form more fine roots, increase the root length and root expansion area, and form a deeper root distribution [41]. Thus, the root system ability to obtain water is enhanced to maintain plant growth, and the physiological responses ability of plant osmosis regulation, carbon and nitrogen metabolism, and enzymatic defense is enhanced, which will play a compensatory and stimulating effect on plant growth after rehydration, thus promoting crop growth and improving dry matter accumulation [42]. The results of this experiment did not well support this conclusion, which may be mainly related to the experimental program, different test crops, and other factors. It can be seen that crop root development and morphological plasticity have complex response characteristics to water and nitrogen coupling, and more studies are needed to be explored and verified in the future.

The dry matter yield characteristics of I. tinctoria roots were the result of the accumulation of the plant photosynthetic products and their distribution among different organs. Water supply and nitrogen application rates, which were closely related to root growth, would affect the significant changes of Pn, Gs and Tr indexes of I. tinctoria, which in turn affect the accumulation of dry matter. In this study, under the same nitrogen application rates, the Pn, Gs and Tr indexes of I. tinctoria reached the peak value when the irrigation amounts reached 100% of the local typical irrigation quota, which indicates that sufficient irrigation can significantly improve I. tinctoria photosynthesis characteristics, and whether a higher irrigation rate still has positive significance needs to be further verified by relevant experiments. In addition, compared with low irrigation amount, the Pn, Gs, and Tr of I. tinctoria improved with increasing irrigation amounts, which may be due to the fact that sufficient irrigation promoted the expansion of leaf area, leading to an increase in the number of stomata and a stronger transpiration pull. Reasonable nitrogen application rates can effectively increase the green holding time of crop leaves, which is positive for the improvement of photosynthesis [43]. Generally, the fertility in the soil is limited, but long-term excessive inputs of fertilizer may lead to high residual fertility in the soil. Therefore, in this study, under the same irrigation amount conditions, the differences in photosynthetic indexes of I. tinctoria between no nitrogen and low nitrogen in 2021 were small. However, with the increase in experimental years, the residual nitrogen in the experimental area of no nitrogen was exhausted, and the Pn, Gs, and Tr of I. tinctoria showed a trend of increasing and significant differences when the nitrogen application rate was increased to 200 kg ha⁻¹, which was due to the fact that topdressing nitrogen application could increase the chlorophyll content in I. tinctoria leaves, which in turn increased Pn. It was also found that no significant difference was observed when the nitrogen application rate was increased from 200 to 250 kg ha⁻¹, although all photosynthetic physiological indexes of I. tinctoria still tended to increase, which undoubtedly suggests that higher nitrogen application rates did not have a positive significance on the photosynthetic characteristics of I. tinctoria. Generally, the increase in leaf area has a positive significance for the increase in Pn [44]. The leaf area was largest and Pn was highest when the irrigation amount was 100% of the local typical irrigation quota and the nitrogen application rate was 200–250 kg ha⁻¹, and the correlation analysis also showed the same results. Lv et al. [45] showed that under medium water and sufficient water conditions, topdressing nitrogen application rates could significantly increase the Pn, maximum fluorescence, maximum photochemical efficiency and chlorophyll SPAD values of liquorice, but it no longer increased significantly after nitrogen application rates exceeded 140 kg ha⁻¹. Liu et al. [46] found that under the same irrigation amounts, the photosynthetic characteristics of red bean showed an increasing trend with the increase in nitrogen application rates and reached the maximum at the medium nitrogen rate of 80 kg ha⁻¹, but the photosynthetic characteristics showed a decreasing trend when the nitrogen application rates exceeded the middle nitrogen rate. Similarly, Qi et al. [47] also found a similar rule, that is, the coupling of sufficient water and medium nitrogen can effectively enhance maize Pn and Tr, and obtain the optimal photosynthetic characteristics, which could better support the conclusion of this experiment. Xia [48] and Meng [49] et al. showed that under the same irrigation amounts, high nitrogen treatments significantly increased peanut and cotton Pn and Tr. This was inconsistent with the study conclusions of this experiment and the former study. The reason for the difference may be related to the experimental program, crop type, regional climatic conditions and other factors.

Water and nitrogen are key factors affecting the growth and development of I. tinctoria, and the two factors can produce a coupling effect to increase yield and economic benefit and improve water–nitrogen productivity. Related studies have shown that a moderate water deficit can increase I. tinctoria yield, but under a low irrigation amount, I. tinctoria yield was significantly reduced [50]. Nitrogen has a significant effect on the yield of I. tinctoria as well as yield components. Increasing nitrogen application rates can improve the yield and economic benefit of I. tinctoria. However, an excessive application of nitrogen can lead to the loss of soil nitrate nitrogen, damage the soil environment, affect the stability of the agricultural ecosystem, and meanwhile reduce the yield of I. tinctoria [51]. Jiang et al. [52] found that with the increase in nitrogen application rates, the fresh mass and dry mass of I. tinctoria leaves increased gradually, while the fresh mass, dry mass and root–crown ratio of roots showed a trend of first increasing and then decreasing with increasing nitrogen application rates. Through the fitting analysis of nitrogen application rates and the quadratic function of I. tinctoria yield, it was found that the highest yield could be obtained with the nitrogen application rate of 220 kg ha⁻¹. The results of Cao et al. [53] showed that nitrogen application rates were positively correlated with the taproot length and taproot diameter of I. tinctoria. With increasing nitrogen application rates, both the fresh weight and dry weight of I. tinctoria from Shanxi Province showed an increasing trend, and the maximum yield was obtained at the nitrogen application rate of 675 kg ha⁻¹. Meanwhile, I. tinctoria from Gansu Province showed a trend of first increasing and then slightly decreasing, and the maximum yield was obtained at the nitrogen application rate of 338 kg ha⁻¹. The results of this study showed that the single factor of water and nitrogen and their interaction significantly affected the yield and economic benefit of I. tinctoria. Under the same irrigation amounts, I. tinctoria yield increased with the increase in nitrogen application rates, which was due to the fact that increasing nitrogen application rates could regulate the soil moisture, promote the root development, enhance the root water absorption capacity, increase the leaf area and chlorophyll content, enhance the photosynthesis, and increase dry matter accumulation and distribution to the root, thus promoting the improvement of root dry matter. Under the same rates of nitrogen application, the yield of I. tinctoria increased with increasing irrigation amounts. This was due to the fact that the increase in irrigation amounts promoted the soil nutrient transport and nitrogen conversion and enhanced the nitrogen metabolism ability of the plant. Especially under the low nitrogen environment, increasing irrigation amounts had a promotion effect on the yield of I. tinctoria, realizing the regulation effect of fertilizer regulating the water and promoting the fertilizer with the water. The two-factor interaction effect analysis of water and nitrogen showed that under the condition of planting I. tinctoria with drip irrigation under plastic film in the irrigation area of the Hexi Corridor, the appropriate irrigation amount was 100% of the local typical irrigation quota, and the nitrogen application rate was 200 kg ha⁻¹, which was consistent with the overall results of He et al. on the water and nitrogen regulation of I. tinctoria [39], but the threshold ranges of the irrigation amount and nitrogen application rate differed slightly, which may be related to the experimental program, the I. tinctoria variety, and the timing of nitrogen application.

Crop water and nitrogen productivity is an important indicator reflecting the relationship between crop water–nitrogen uptake and utilization and yield [54]. Reasonable water and nitrogen supply can greatly improve crop yield and water–nitrogen productivity, because reasonable irrigation amounts reduce ineffective soil water depletion, enhance crop transpiration intensity, and effectively coordinate the reduction in consumption in the farm-crop system, increasing water productivity [55]. It has been shown that appropriately increasing the irrigation amounts and nitrogen application rates within a certain range will increase crop water–nitrogen productivity, but excessive irrigation and nitrogen application will result in a waste of resources and serious reduction in water–nitrogen productivity [56,57]. In this study, when the irrigation quota increased from 70% to 85% of the typical irrigation quota, the water productivity, irrigation water productivity, nitrogen partial productivity, and nitrogen agronomic use efficiency showed a significant increasing trend, while the water productivity, nitrogen partial productivity, and nitrogen agronomic use efficiency showed an increasing trend, but the increase was not significant, while the irrigation water productivity showed a significant decreasing trend, which was due to the fact that on the one hand, the effect of irrigation on the yield of I. tinctoria was in accordance with the law of diminishing returns; on the other hand, I. tinctoria was subjected to drought stress at low irrigation amounts, which suppressed physiological activities and negatively affected root growth and development, and then the yield of I. tinctoria was lower at low irrigation amounts, which resulted in a low water–nitrogen productivity. Meanwhile, the results showed that the water productivity, irrigation water productivity, and nitrogen agronomic use efficiency increased significantly when nitrogen application rates increased from 150 to 200 kg ha⁻¹, while nitrogen partial productivity showed a decreasing trend. When nitrogen application rates increased from 200 to 250 kg ha⁻¹, nitrogen partial productivity and nitrogen agronomic use efficiency showed a decreasing trend, while water productivity and irrigation water productivity also showed a decreasing trend, but the increase was not significant, which was similar to the results of previous studies [58,59]. Therefore, supplemental irrigation is needed to promote water–nitrogen coupling to improve the water–nitrogen productivity of I. tinctoria along with additional nitrogen application rates. Tang et al. [60] pointed out that the water–nitrogen interaction has a significant role in regulating water and nitrogen utilization in crops; in this study, we also found that irrigation combined with nitrogen was more effective in promoting the growth of I. tinctoria plants than irrigation alone, and that moderately increasing the amount of nitrogen under a certain irrigation level can improve water productivity and irrigation water productivity. This may be due to the fact that both irrigation and nitrogen application contributed to the vigorous growth of I. tinctoria, which increased the water and nutrient demand of the plant and thus enhanced the water and nutrient uptake of the root. Meanwhile, irrigation could promote nitrogen uptake and utilization by crops, but excessive irrigation and nitrogen application may cause the negative effect of downward leaching of soil nitrogen with irrigation [61]. In this study, it was found that increasing the irrigation amounts was beneficial for enhancing the nitrogen partial productivity and nitrogen agronomic use efficiency, but the increase in irrigation amounts led to a decrease in irrigation water productivity; in contrast, water productivity and irrigation water productivity were maintained at a high level when the nitrogen application rate was at 200 kg ha⁻¹. Therefore, in order to balance water and nitrogen productivity with I. tinctoria yield, and to facilitate production practices, it is recommended to apply 200 kg ha⁻¹ of nitrogen at 100% of the local typical irrigation quota.

At present, a large number of scholars have conducted irrigation amounts optimization experiments on tomato, cucumber, alfalfa, potato and other crops, and it has been found that reasonable irrigation amounts improved crop quality and were conducive to the formation and accumulation of crop quality [62,63,64]. It has also been shown that timely and appropriate irrigation was the basis for I. tinctoria to obtain higher indigo, indirubin and (R,S)–goitrin, while lower and excessive irrigation levels are not conducive to the accumulation of indirubin and (R,S)–goitrin and the formation of integrated quality. In this study, it was found that under the same nitrogen application rates, when the irrigation amounts increased from 70% to 85% of the local typical irrigation quota, indigo, indirubin, (R,S)–goitrin, total nucleosides, uridine and adenosine tended to increase significantly, whereas all the quality factors tended to decrease when irrigation amounts increased from 85% to 100% of the local typical irrigation quota, but the decreasing rate was not significant. This is due to the fact that on the one hand, excessive irrigation changes the microbial activity trajectory, reduces the effectiveness of nutrients in the soil, and reduces the amount of nutrient uptake by root; on the other hand, excessive irrigation caused a decline in the oxygen content of the soil, inhibiting root respiration, which led to a decline in the vigor of the root, and it also weakened the absorption of water and fertilizer capacity. Deng et al. [65] concluded that a reasonable reduction in irrigation level was effective in promoting the accumulation of I. tinctoria indigo, indirubin, and (R,S)–goitrin compared to conventional sufficient irrigation, but excessive water deficit instead resulted in a significant reduction in the content of effective ingredients, which was similar to the findings of this study. Meanwhile, the results of this study showed that under the same irrigation amounts, indigo, indirubin, (R,S)–goitrin, total nucleosides, uridine, and adenosine showed an increase in the nitrogen application rates from 150 to 250 kg ha⁻¹, where the maximum increment could be obtained when the irrigation amount was 85% of the local typical irrigation quota, which was similar to the results obtained by Deng et al. [65]. Therefore, reasonable irrigation is needed to promote water–nitrogen coupling and improve the quality of I. tinctoria while increasing nitrogen application. Ma et al. [66] pointed out that water–nitrogen interaction can significantly regulate the crop quality. In this study, we also found that irrigation combined with nitrogen application could improve the content of effective ingredients of I. tinctoria more than irrigation alone; that is, increasing the nitrogen application rates under certain irrigation amounts could improve the quality of I. tinctoria, which may be due to the fact that a favorable water–nitrogen environment is conducive to the activation of the secondary metabolism in the I. tinctoria root and thus increased its content of active components [61]. In conclusion, 85% of the local typical irrigation quota coupled with a 250 kg ha⁻¹ nitrogen application rate can improve the content of effective ingredients in the root of I. tinctoria, which had the highest nutritional value.

In this study, a hierarchical model for the comprehensive evaluation of I. tinctoria was established using Yaaph software, the weight hierarchy was determined based on the AHP method, and the entropy weighting method was used to assign weights to the single index of I. tinctoria. Finally, the comprehensive evaluation was carried out based on the TOPSIS comprehensive model with combined weighting, and W3N2 (sufficient water and medium nitrogen) was determined to be a more suitable water–nitrogen combination, which indicated that the water–nitrogen system matching the sufficient water and medium nitrogen under the conditions of drip irrigation under plastic film can not only ensure the yield of I. tinctoria but also improve the water–nitrogen productivity and the contents of indigo, indirubin, (R,S)–goitrin, total nucleosides, uridine, and adenosine in root. It can be seen that W3N2 is beneficial for achieving the purpose of increasing the production and high quality of I. tinctoria planting in the experimental area. However, the influence of irrigation levels on yield, water–nitrogen productivity, and quality in this study lacked the upper limit of the dosage. It is necessary to continue to increase the gradient of irrigation amounts and conduct a more comprehensive and integrated evaluation on more indexes in further studies. The most appropriate water–nitrogen regime needs to be further verified.

5. Conclusions

A comprehensive evaluation of multiple indexes of the I. tinctoria water–nitrogen coupling planting model using the AHP method, entropy weight method, and TOPSIS method concluded that 100% of the local typical irrigation quota and 200 kg ha⁻¹ nitrogen application was evaluated optimally, which could improve the soil water and nitrogen environment in the I. tinctoria planting area and increase the dry matter mass accumulation of I. tinctoria while maintaining the photosynthetic capacity of leaves, facilitating the distribution of dry matter to the root, and laying the foundation for high yield in the later stage. It can also obtain better quality of medicinal materials as well as increase the water productivity, irrigation water productivity, nitrogen partial productivity, and nitrogen agronomic use efficiency, leading to greater net proceeds. Therefore, in the actual production of I. tinctoria field in the oasis irrigation area of the Hexi Corridor in China, in order to obtain medicinal materials with stable yield and better quality, it is recommended to control the nitrogen application rates at 190.30 to 218.27 kg ha⁻¹ by appropriately adopting reduced nitrogen application rates under the level of maintaining the local typical irrigation quota.