Fine Soil Texture Is Conducive to Crop Productivity and Nitrogen Retention in Irrigated Cropland in a Desert-Oasis Ecotone, Northwest China

Abstract

Agricultural production in oases requires extensive irrigation and nitrogen (N) inputs, which result in a high incidence of non-point-source pollution. Information on how soil texture affects crop productivity and water and N use efficiency is needed to improve N management in oases. A two-year field study with six free-draining leaching pits was set up to quantify soil water content (SWC), drainage, yield, N uptake, soil residual N, N leaching, water-use efficiency (WUE), and N fertilizer-use efficiency (NFUE) on sandy-textured soils in a young oasis field (24 years, YOF) and loam-textured soils in an old oasis field (>60 years; OOF) within an intensive oasis agricultural zone in Northwest China. The results showed that sand content in the YOF was significantly higher than that in the OOF (p < 0.01), while both clay and silt contents in the YOF were significantly lower than those in the OOF (p < 0.01). Soil water storage (SW) in the 0–100 cm soil layer, evapotranspiration water-use efficiency (WUEET) and irrigation water-use efficiency (WUEIrrig) in the OOF were, respectively, 87.1, 23.5, and 24.1% higher than those in the YOF (p < 0.05), and deep drainage in the YOF was 6.2 times higher than that in the OOF (p < 0.01). Maize N uptake and soil N accumulation in the OOF in the 0–100 cm soil layer were, respectively, 33.4 and 10.3% higher than those in the YOF (p < 0.05), and N leaching loss in the YOF was 1.65 times higher than that in the OOF (p > 0.01). Crop yield and NFUE in the OOF were, respectively, 23.9 and 27.2% higher than those in the YOF (p < 0.05). These results indicated that sandy-textured soils with high sand content were not conducive to water and N retention, resulting in lower crop yields and water- and fertilizer-use efficiency in YOF. Therefore, it is imperative that management practices in sandy-textured land promote improvements in soil structure and maintain the long-term productivity of the young cultivated fields.

1. Introduction

China ranks first in the world in intensive agriculture, which requires substantial amounts of synthetic fertilizer to meet the demands for crop growth and yield, and this results in agricultural non-point-source pollution [1,2]. Nitrogen (N) is an essential nutrient for crop yield, and over half of the world’s population is nourished by crops grown with N fertilizers [3]. However, only about 42–47% of N added to croplands is converted to crop growth, while the rest remains in soil, often leaching into groundwater following heavy rains or irrigation [4,5,6].

Nitrate (NO₃⁻-N) concentrations increased sharply in groundwater after the 1980s in China, and 80.2% of the monitoring wells in the country contained >30 mg·L⁻¹ of NO₃⁻-N in 2016, with stable isotope analyses showing that the main source was agricultural N applications [7]. NO₃⁻-N leaching in agricultural ecosystems has become a universal challenge worldwide [8,9,10], and is also the main cause for the deterioration of groundwater quality in China.

Soil texture greatly affects soil aeration, water-holding capacity, soil fertility, and crop yield, and plays an important role in regulating N leaching [11,12]. Silt, clay, or silty clays reduce the number of soil macropores, increase water retention times, and can ensure sufficient available water and N supply for crop growth [5]; sand or sandy loams with strong saturated hydraulic conductivity and weak water holding capacity are conducive to the leaching of water and N [13,14]. Maize productivity and N-uptake efficiency in clay textures were more responsive to N fertilizer than other soil textures [15]. Therefore, soil texture was key to crop yield and the utilization efficiency of water and N fertilizer [12]. However, to date, there has been a lack of quantitative research on the impacts of soil textures on land productivity and agricultural non-point-source pollution. An increased understanding of the impacts of soil textures on water- and N-use efficiency and crop productivity is critical to efforts to improve crop N-use efficiency to achieve control of agricultural non-point-source pollution.

The Zhangye irrigation district located in the Heihe River Basin is one of the regions with the highest yield of a highest yields of crops in China. Irrigated cropland in this region covers an area of 2896 km², of which nearly half are oasis fields cultivated in the desert for the past 50 years [16,17]. The climate conditions of scarce rainfall and strong evaporation necessitate irrigation, and agriculture in the area requires 19.8 × 10⁸ m³ of irrigation water annually [18]. Approximately 75% of agricultural land is cultivated with maize using flood irrigation, with an irrigation depth of 600–1200 mm depending on soil conditions [17]. In this region, the average N fertilizer application rate is 350 kg·hm⁻²; however, N-use efficiency is only 35–49% [16]. These fertilization rates inevitably increase the possibility of NO₃⁻-N leaching, non-point-source pollution, and a decrease in land productivity.

Previous studies have focused on ways to reduce N loss by regulating hydrological processes with crop mulching [19], addition of biochar and organic fertilization [16,20], intercropping [21], and the optimization of irrigation schedule and fertilizer applications [14,22]. However, the implementation of those measures is limited in Zhangye irrigation district. Cultivation in the oasis leads to high spatial heterogeneity in soil texture because (1) long-term irrigation with silt-laden river water increases soil silt and clay contents [5], and it is the increase in clay content that dramatically enhances the development of soil aggregates and improves soil structure; and (2) inputs of manure, chemical fertilizers, and crop residues increases organic matter and nutrient content due to cultivation, which may in turn promote adsorption of clay particles and improve soil aeration and water- and fertilizer-holding capacity to some extent [5].

We conducted a two-year field experiment to determine the effects of soil texture on crop productivity, water- and N-use efficiency, and hydrological N losses from an intensive spring maize system in a desert-oasis ecotone of China. The specific objectives of this study were: (1) to determine how soil texture affects water and N retention; (2) to define the impacts of soil texture on leaching and accumulation of NO₃⁻-N; and (3) to establish how and to what extent soil texture affects maize productivity.

2. Materials and Methods

2.1. Study Area

A controlled field experiment was conducted at Linze Inland River Basin Research Station, Chinese Ecosystem Research Network (100°07′ E, 39°24′ N; elevation,1383 m), located in the middle reaches of the Heihe River Basin in Gansu Province, China (Figure 1a). Climate conditions in the study area are temperate, arid, and continental, with an annual precipitation of 112 mm, annual potential pan evaporation of 2400 mm, annual temperature of 7.6 °C, and annual total sunshine duration of 3051 h [22]. The depth of the water table is relatively high, and varies from 2.5 to 4.5 m due to fluctuations in the water level of the Heihe River [22]. Peripheral desert, desert-oasis ecotone, and central oasis constitute the main landscape types in this region. The sandy lands in the peripheral of oasis have been gradually cultivated for agricultural use since the 1960s; hence, the age of cultivation of irrigated fields range from 1 year to more than 60 years old [5]. The croplands and the adjacent sandy land have the same soil parent material, and there was no obvious difference in soil particle size composition and organic matter content at the initial stage of cultivation for croplands in different periods in this region [23]. The recorded meteorological information of maize growth period is shown in Figure 2. The potential evapotranspiration (ET₀) was calculated through the FAO56 Penman–Monteith equation [24].

2.2. Experimental Design and Field Management

Two irrigated fields were selected for this study. One was at the edge of the desert with a cultivation period of 24 years, considered a “young oasis field” (YOF); the other was in the central oasis, with the cultivation history of more than 60 years, designated an “old oasis field” (OOF) (Figure 1a). The soil type in the YOF was Ari psamment, and in the OOF it was irrigation-farmed fluvent. An experimental site, approximately 48 m² (6 m (width) × 8 m (length)), was established in a field where spring maize has been maintained for almost ten years. Test plots were demarcated by 1 m alleys, and contained 6 sub-plots 1 × 1 m with two treatments (the YOF and OOF), and three replicates for each treatment (Figure 1b). Soils at the two fields were sampled in triplicate from the 0–200 cm soil profile with intervals of 20 cm to analyze soil properties. Soil texture and soil properties were determined using the methods of Zhang et.al [5]. Detailed soil properties for the 0–200 cm soil profile is summarized in

Table 1. Soil physical and chemical properties in a young oasis field (YOF) and old oasis field (OOF). Averages are based on three replicates.

Items	Soil Layer (cm)	B.D (g·cm−3)	Ks mm/min	F.C (cm3·cm−3)	W.P	O.M	Total N	Total P	Total K
					(cm3·cm−3)	(g·kg−1)	(g·kg−1)	(g·kg−1)	(g·kg−1)
YOF	0–20	1.61 ± 0.14	12.65 ± 2.32	0.19 ± 0.01	0.07 ± 0.01	8.42 ± 1.01	0.62 ± 0.62	0.49 ± 0.01	17.79 ± 1.45
	20–40	1.60 ± 0.14	12.48 ± 1.27	0.18 ± 0.02	0.06 ± 0.00	7.64 ± 0.92	0.58 ± 0.07	0.38 ± 0.06	17.65 ± 1.16
	40–60	1.59 ± 0.17	15.56 ± 1.33	0.18 ± 0.01	0.04 ± 0.00	6.38 ± 0.86	0.51 ± 0.04	0.44 ± 0.03	17.46 ± 1.92
	60–100	1.59 ± 0.11	15.83 ± 1.99	0.17 ± 0.02	0.05 ± 0.00	2.18 ± 0.33	0.22 ± 0.07	0.27 ± 0.01	17.15 ± 2.04
	100–140	1.62 ± 0.17	16.45 ± 2.31	0.17 ± 0.03	0.05 ± 0.00	1.16 ± 0.21	0.17 ± 0.04	0.26 ± 0.02	17.12 ± 2.13
	140–180	1.60 ± 0.15	16.22 ± 1.12	0.17 ± 0.03	0.04 ± 0.00	1.02 ± 0.08	0.12 ± 0.02	0.20 ± 0.02	16.89 ± 1.73
OOF	0–20	1.41 ± 0.12	4.04 ± 0.56	0.36 ± 0.02	0.20 ± 0.02	11.27 ± 1.12	0.82 ± 0.15	0.95 ± 0.08	34.56 ± 3.62
	20–40	1.38 ± 0.11	4.16 ± 0.89	0.37 ± 0.04	0.17 ± 0.01	9.34 ± 0.88	0.81 ± 0.13	0.87 ± 0.06	35.14 ± 4.28
	40–60	1.34 ± 0.10	3.27 ± 0.21	0.33 ± 0.03	0.20 ± 0.03	8.568 ± 1.12	0.73 ± 0.12	0.89 ± 0.10	29.68 ± 1.54
	60–100	1.34 ± 0.08	2.87 ± 0.13	0.29 ± 0.04	0.16 ± 0.01	5.76 ± 0.69	0.45 ± 0.07	0.74 ± 0.10	28.59 ± 2.89
	100–140	1.42 ± 0.09	2.69 ± 0.14	0.27 ± 0.02	0.17 ± 0.01	5.06 ± 0.37	0.28 ± 0.03	0.69 ± 0.06	22.58 ± 2.78
	140–180	1.46 ± 0.15	2.55 ± 0.19	0.34 ± 0.02	0.14 ± 0.01	4.77 ± 0.26	0.20 ± 0.04	0.57 ± 0.05	19.58 ± 1.52

BD, bulk density; Ks, saturated hydraulic conductivity; F.C, field capacity; W.P, wilting point; O.M, organic matter; NPK for nitrogen, phosphorus, and potassium.

.

The maize (Zea mays L.) cultivar used in this study was ‘Zhengdan 958’, which was widely planted in the area, and it was planted using a typical practice of sowing in late April and harvesting in late September. The planting pattern for maize included mulching with a polypropylene film before planting, and seed spacing of 20 cm, row spacing of 60 cm, planting depth of 5–10 cm, and planting density of 80,000 plants·hm⁻².

To determine NO₃⁻-N leaching from the two irrigated croplands, the irrigation amount and irrigation time were kept uniform, and traditional flood irrigation was carried out based on soil-moisture and crop-growth conditions. Irrigation and fertilization management measures were the same as those commonly used by local farmers. Fields were irrigated a total of 8–9 times during maize growing season, with irrigation depth of 90 mm each time. A flow meter was used to apply the same amount of water to each plot. Fertilizers were applied at rates of 330, 116, and 40 kg·ha⁻¹, respectively, for NPK. Fertilizers were uncoated urea, diammonium phosphate (DPA), and compound fertilizer. N fertilizer was applied in three parts (⅓ at jointing, ⅓ at tasseling, and ⅓ at milking stages), and phosphorus and potassium were applied at the time of tasseling and milking. All fertilizer applications were performed by broadcasting evenly in the rows, followed by irrigation. The detailed schedule of irrigation and fertilizer applications are shown in

Table 2. Irrigation and fertilization information.

Item	Management Activities
Irrigation date (2019)	4/29	6/5	6/18	6/30	7/14	7/28	-	8/9	8/25
Irrigation date (2020)	4/24	5/30	6/13	6/24	7/6	7/19	7/31	8/12	8/27
Irrigation (mm)	90	90	90	90	90	90	90	90	90
Fertilization type and rate (N-P2O5-K2O, kg·ha−1)			Urea + DPA (110-76-0)		Urea + Compoundfertilizer (110-40-40)			Urea (110-0-0)

DPA is diammonium phosphate. Value in parentheses signifies the amount of fertilizer application rate (kg·ha⁻¹).

.

2.3. The Leaching Pit

In July 2015, 6 free-draining leaching pits were installed in clusters of two at the test plot. To avoid the effects of oasis or fringe, the leaching pit was directly implemented into the test plots. The height of the leaching pit was 220 cm, and the length and width were both 100 cm, and soil (200 cm deep) was backfilled into the leaching pit with original soil bulk density. Ten soil moisture sensors (Em50, Decagon, Pullman, WA, USA) and 6 sampling devices (a suction lysimeter designed by Institute of Geographic Sciences and Natural Resources Research, which was composed of a Teflon pipe and porous ceramic tube) were installed in each leaching pit. Soil moisture sensors were installed at 10, 20, 40, 60, 80, 100, 120, 140, 160, and 180 cm depths for monitoring the soil water content, and the measurement duration was set to 30 min each time. Ceramic tubes were installed at 20, 40, 60, 100, 140, and 180 cm depths for collection of leachates. With the occurrence of flood irrigation, the soil water content increased, and leachate entered the ceramic tube and was sucked out with a vacuum pump. After irrigation, a vacuum pump at about −0.8 MPa was applied to the suction lysimeter for 12 h of equilibrium to collect soil water. To facilitate drainage and prevent clogging, the bottom of each leaching pit was designed as an inverted cone and packed with 20 cm gravel and coarse sand (Figure 1b). After irrigation, the leakage was weighed, and 250 mL of the leakage was collected with a polyethylene bottle. The water samples were collected and stored in a −20 ℃ refrigerator until the NO₃⁻-N concentration was analyzed. The collection time, leakage amount, and leakage rate during the collection process were recorded.

2.4. Sampling and Measurements

2.4.1. Soil Water Storage

Soil water storage (SW, mm) refers to the soil water content of a certain soil layer thickness. SW can reflect the migration and change in soil water in a growth period, calculated as Equation (1):

$$\mathrm{SW}=\mathrm{SWC}\times \mathrm{h}\times 10$$

(1)

where, SWC is soil water content (cm³·cm⁻³); h is soil depth (cm).

2.4.2. Soil Water Balance

The water balance in the leaching pit is shown in Equation (2), ignoring horizontal flow and capillary rise (Zheng et al., 2020):

$$P+I-S-ET-∆S=0$$

(2)

where P and I are precipitation (mm) and irrigation (mm), respectively; ET is evapotranspiration (mm); S is deep drainage (mm); and ∆S is the change in SW (mm).

2.4.3. Plant N Uptake

Four maize plants were randomly sampled from each plot to measure the aboveground biomass and plant N uptake at maturity. Plants in each plot were first separated into straw (including stems and leaves), corn, and cob, and samples were kept at 80 °C to reach a constant weight. Following this, dried samples were ground and passed through an 80-mesh sieve for analysis of TN content. Dried samples were first digested using an H₂SO₄-H₂O₂ method, and TN was then measured using an automatic flow-injection analyzer (AA3, Fia Star 5000, Foss Tecator, Zurich, Sweden). The N uptake of each plant was estimated by multiplying plant dry weight (1 combined plant) and plant N content of 3 parts per plot.

2.4.4. Soil N Residual

Prior to maize seeding and following harvest, 0~200 cm soil samples were collected at 20 cm intervals with a soil auger (3.8 cm in diameter) in the central area of each leaching pit, and holes were then backfilled with the same layers of soil. The soil NO₃⁻-N was extracted with 0.5 M K₂SO₄ and then determined with AA3. In our study area, fields enter a fallow period following maize harvest, with no new N inputs (atmospheric nitrogen deposition is negligible) [16]. Therefore, we used the measured soil NO₃⁻-N value following harvest in 2019 as the initial content for 2020. Soil residual N was calculated by multiplying soil bulk density and net accumulated NO₃⁻-N in the soil for the period after harvesting and prior to sowing.

2.4.5. NO₃⁻-N Dynamics in Soil Solution

To understand the dynamics of soil NO₃⁻-N content after irrigation, soil solutions were collected from different soil layers from treatments with and without N fertilizer separated in time in the same field. Thus, soil solution extracted on 13 June, 6 July, and 12 August was treated with N fertilizer (with N fertilizer), while that extracted on 14 June, 19 July, and 27 August did not receive prior N fertilizer (without N-fertilizer) in 2020. Normally, the maximum leaching of N fertilizer applied to soils occurs 3–5 days after irrigation due to high relative soil moisture content [14]. Therefore, soil solutions were extracted with a vacuum pump before N application (without fertilizer; day 0) and then daily following N application for 6 days (with fertilizer; days 1, 2, 3, 4, 5, 6) from 20, 40, 60, 100, 140, and 180 cm soil depth. To ensure consistency, soil solutions with and without fertilizer were collected three times during the maize growth period. Meanwhile, deep soil drainage was collected at the bottom of the leaching pits after each irrigation using a polyethylene barrel, and the volume of soil drainage was recorded. NO₃⁻-N concentrations of the soil solution and drainage were analyzed using AA3.

2.4.6. N Leaching

In the research of field crops, N leaching usually means the process of water-soluble nitrate (NO₃⁻-N) in soil migrates down to the root active layer, entering the groundwater under the influence of heavy rainfall or irrigation [25]. In addition, N leaching also refers to a difference between the amount of total N inputs and N outputs and the N change in the soil profile. Previous studies have shown that the maize roots are distributed over 0–100 cm in depth [14,26]; therefore, in our experiment, the 0–100 cm soil profile was used for calculating N balance (Equation 3):

$$L=i\mathrm{RN}\left(0-100 \mathrm{cm}\right)+AN-p\mathrm{RN}\left(0-100 \mathrm{cm}\right)-NU$$

(3)

where L is N loss through leaching; iRN and pRN are N residual in soil of pre-sowing and postharvest, respectively; AN is the amount of added N, including N input from wet deposition, irrigation, and fertilization; and NU is N uptake by maize.

Wet deposition and irrigation samples were collected from April 2019 to October 2020. Wet deposition samples were collected with an automatic wet deposition sampler (LB-8110, Lubo, Qingdao, China), which was controlled by microcomputer. The sampler was automatically opened to collect when the rain began, and the cover was closed immediately after the rain stopped. The triplicate irrigation samples (100 mL) were collected using a white polyethylene bottle after each irrigation. The N input from wet deposition and irrigation was calculated by multiplying the amount of rainwater or irrigation by their corresponding N concentrations, respectively. The rainfall data were provided by an agro-meteorological station owned by Linze Ecological Observational and Experimental Station, which is 50 m away from our experimental plots.

2.4.7. Water- and Nitrogen-Use Efficiencies

Water-use efficiency (WUE) is an important indicator of the relationship between crop yield and water consumption. It mainly includes irrigation water-use efficiency (WUE_Irrig) and evapotranspiration water-use efficiency (WUE_ET). Fertilizer N-use efficiency (NFUE) is an effective index to measure nitrogen-use efficiency in agricultural production. WUE_Irrig, WUE_ET, and NFUE were calculated using Equations (4)–(6) of Ma et al. (2022).

$${\mathrm{WUE}}_{Irrig}=0.1\frac{\mathrm{Y}}{\mathrm{I}}$$

(4)

$${\mathrm{WUE}}_{ET}=0.1\frac{\mathrm{Y}}{ET}$$

(5)

$$\mathrm{NFUE}=\frac{\mathrm{Y}}{\mathrm{N}}\times 100\%$$

(6)

where Y is grain yield (kg·ha⁻¹), and N is the nitrogen application rate (kg·ha⁻¹).

2.5. Data Analysis

All data were expressed as means ± standard deviation (SD). The 2-independent-samples t-test (p < 0.05) was applied to compare soil particle size distribution, apparent water balance, and apparent N balance between the YOF and OOF. Statistical analysis was performed in SPSS 21.0 (Chicago, IL, USA). Surfer 11.0 (Golden Software, Denver, CO, USA) was used to draw the map of spatiotemporal variability in soil moisture content.

3. Results

3.1. Soil Texture

Soil particle size distribution in the 0–200 cm soil profile is shown in Figure 3. The sand content in the YOF in the 0–200 cm profile was very significantly higher than that in the OOF (p < 0.01). However, the silt and clay contents in the OOF in the whole profile were higher than those in the YOF (p < 0.01). The soil is classified as sandy loam (at 0–20 cm depth) and sandy (at 20–200 depth) in YOF, and loamy (at 0–40 cm depth) and silty loam (at 40–200 depth) in OOF, according to the USDA soil classification.

3.2. Soil Moisture

3.2.1. Dynamics of Soil Water Content

Figure 4 illustrates the spatiotemporal variability in daily soil water content (SWC) in the growing season of 2019 and 2020. Overall, SWC dynamics responded to irrigation, and distinct spikes appeared after each irrigation event. SWC in deep soil (120–180 cm) was higher than that in upper (0–60 cm) and middle soil layers (60–120 cm). The lowest value of SWC in the YOF and the OOF was 0.046 and 0.137 cm³·cm⁻³ at 40–60 cm and 10–20 cm depths, respectively, and 0.39 cm³·cm⁻³ in the YOF and 0.35 cm³·cm⁻³ in the OOF at 180 cm depth. SWC fluctuated sharply in upper and gently in deep soil. The isopleth of SWC in upper soil was dense and curved, while in the deep layer it was loose and smooth, indicating the changes in SWC in the upper layer fluctuated in deep soil. SWC also changed with stages of maize growth. The lowest SWC occurred at tasseling and milking.

3.2.2. Water Balance and WUE

The amount of irrigation was 720 and 810 mm, and that of precipitation was 119.5 and 100.5 mm in 2019 and 2020 growth seasons, respectively (

Table 3. Apparent water balance (mm) of soil–plant system in the YOF and OOF.

Items	2019		2020		Mean
	YOF	OOF	YOF	OOF	YOF	OOF
Precipitation	119.5	119.5	100.5	100.5	110	110
Irrigation	720	720	810	810	765	765
ET	576.6	709.9 **	607.6	749.4 **	592.2	729.6 **
SW (0–100 cm)	119.1	215.2 **	117.2	227.1 **	118.2	221.2 **
SW (100–200 cm)	230.2	258.6	248.4	252.8	239.3	255.7
The variation in SW (0–200 cm)	28.7	94.2 **	56.7	128.1 *	42.8	112.2 *
Deep drainage (>200 cm)	234.2 **	35.4	246.2 **	33.0	240.0 **	33.2
WUEET (kg·m−3)	1.72	1.85	1.24	1.73 *	1.45	1.79 *
WUEIrrig (kg·m−3)	1.48	1.69 *	1.19	1.6 *	1.33	1.65 *

** and * represent p < 0.01 and p < 0.05 between the young oasis field (YOF) and old oasis field (OOF) at the same depth, respectively.

). The annual in ET, SW in 0–100 cm, and the variability in SW in 0–200 cm in the OOF were higher than those in the YOF (p < 0.05). No significant difference was found between the YOF and the OOF in SW for the 100-180 cm soil layer (p > 0.05). The annual deep drainage in the YOF was 7.2 times that in the OOF (p > 0.05). ET was the largest single item of water consumption, and accounted for 67.7 and 60.4% of water input in the YOF and OOF, respectively. During the maize growing seasons, higher water-use efficiency was found in the OOF treatment (Table 3). The average annual WUE_ET and WUE_Irrig in the OOF were significantly higher than those in the YOF (p < 0.05).

3.3. Apparent Nitrogen

3.3.1. Dynamic of NO₃⁻-N Concentration in Soil Solution

NO₃⁻-N concentration in soil solutions from layers of 0–180 cm varied by field, soil depth, and crop-growth stages (Figure 5 and Figure 6). In general, with irrigation, the change in NO₃⁻-N in different soil layers with fertilization was more dramatic than that without fertilization. In the former, NO₃⁻-N increased sharply, then decreased slowly in upper soil (0–60 cm), while it increased steadily in deep soil (100–180 cm). Morevoer, NO₃⁻-N at 140–180 cm depth in the YOF was significantly higher than that in the OOF after day three (Figure 4). In contrast, the concentration of NO₃⁻-N across the whole profile without fertilization exhibited a decreasing trend, then increasing, and then decreasing again (Figure 5).

With fertilization and at the jointing stage, NO₃⁻-N in the OOF reached the maximum in the upper soil (0–60 cm) on day four, NO₃⁻-N peak in the middle layers (60–120 cm) was observed on day five, while peak in deep layers (120–180 cm) was observed on day six. A NO₃⁻-N peak in the 0–100 cm layer in the YOF was observed 1–2 days earlier than that in OOF. At tasseling and milking stages in both fields, NO₃⁻-N dynamics exhibited the same trend as for the jointing stage, but peak NO₃⁻-N was observed 1–2 days earlier than at the jointing stage.

3.3.2. Residual of NO₃⁻-N in Soil Profile

Soil residual NO₃⁻-N at pre-seeding and post-harvest is shown in Figure 7. Soil NO₃⁻-N fluctuated and decreased with the increase in soil depth; the peak appeared at 20 cm in the OOF and at 40 or 60 cm in YOF. Soil NO₃⁻-N in the OOF at 20, 40, 60, 80, 120, and 180 cm was significantly different (p < 0.05) than that in the YOF at pre-seeding in 2019 and post-harvest in 2019 and 2020. There was a significant difference (p < 0.05) in soil NO₃⁻-N at 160 cm between the OOF and the YOF at pre-seeding in 2019, while no significant difference (p > 0.05) was found in this layer post-harvest in 2019 and 2020. Fertilization and irrigation resulted in NO₃⁻-N migration through the soil layers, and the maximum increase was found at 20 cm, with an average of 3.14 mg·kg⁻¹ for YOF and 9.65 mg·kg⁻¹ for OOF; the maximum decrease was observed at 60 cm for the YOF (1.85 mg·kg⁻¹) and at 140 cm for the OOF (2.0 mg·kg⁻¹).

Maximum NO₃⁻-N accumulation in the YOF was found in the upper soil (0–60 cm) with the average of 42.45 kg·hm⁻², which was significantly higher than that in the OOF (p < 0.05) (

Table 4. NO₃⁻-N accumulation (kg·hm⁻²) in different soil layers in the YOF and OOF.

Depth (cm)	2019		2020		Mean
	YOF	OOF	YOF	OOF	YOF	OOF
0–60	37.9 **	21.3	47.0 **	17.3	42.45 **	19.3
60–120	1.34	35.4 **	−4.47	37.1 **	−1.57	36.25 **
120–200	33.7	44.1 *	30.4	40.6 *	32.05	42.35 *
0–200	67.1	89.1 *	67.8	81.6 *	67.45	85.35 **

** and * represent p < 0.01 and p < 0.05 between the young oasis field (YOF) and old oasis field (OOF) at the same depth, respectively.

). NO₃⁻-N accumulation in the YOF reached the lowest level in the middle layer (60-120 cm), where it tended toward negative accumulation in 2020. Higher accumulations in the OOF were observed in the middle (36.25 kg·hm⁻²) and deep soil layers (42.35 kg·hm⁻²), and were significantly different from those in the YOF (p < 0.05).

3.3.3. Plant Biomass Production and N Uptake

Dry aboveground biomass and the amount of N uptake in the late maturity stage is shown in

Table 5. Aboveground biomass and plant N uptake in the YOF and OOF.

Year	Treatment	Aboveground Biomass (kg·ha−1)				N Uptake (kg·ha−1)
		Straw Yield	Cob Yield	Grain Yield	Total	Straw	Cob	Grain	Total
2019	YOF	7121.5	1952.3	10663.6	19679.2	54.4	12.7	121.6	188.6
	OOF	8708.6 *	1959.8	12184.7 *	23127.4 **	73.9 **	17.4 *	149.3 *	240.6 **
2020	YOF	6168.3	1806.1	9628.4	17602.8	52.4	14.8	115.2	182.4
	OOF	8718.9 **	22091 **	12963.4 **	23891.4 **	79.7 **	16.9	154.8 *	254.3 **

** and * represent p < 0.01 and p < 0.05 between the young oasis field (YOF) and old oasis field (OOF) at the same depth, respectively.

. Aboveground biomass and plant N uptake of all maize tissue exhibited the order grain > straw > cob. Biomass and total N uptake in the OOF was significantly higher than that in the YOF (p < 0.05) with the same fertilization and irrigation measures. The difference in biomass between experimental years may be attributed to the variability in climate.

3.3.4. N Leaching, N Balance, and NFUE

The apparent N balance in the soil–plant system is shown in

Table 6. Apparent N balance (kg·hm⁻²) of soil–plant system and NFUE in the YOF and OOF.

Items	2019		2020		Mean
	YOF	OOF	YOF	OOF	YOF	OOF
(1) N input from Fertilizer	330	330	330	330	330	330
(2) N input from irrigation	67.4	67.4	79.4	79.4	73.4	73.4
(3) N input from wet deposition	6.3	6.3	6.9	6.9	6.6	6.6
(4) Maize N uptake	188.6	240.6 **	182.4	254.3 **	185.5	247.5 **
(5) Soil N accumulation (0–100 cm)	39.3	56.7 **	54.3 *	46.4	46.8	51.6 *
(6) Soil N accumulation (100–200 cm)	32.43	27.81	20.27	27.3	26.4	27.5
(7) N leaching loss (>100 cm)	131.8 **	48.2	122.9 **	48.1	127.4 **	48.1
(8) N leaching loss (>200 cm)	99.4 **	20.4	102.6 **	20.8	101.0 **	20.6
(9) Apparent N loss (= 1 + 2 + 3 − 4 − 5 − 6 − 8)	44.0	58.2 *	56.7	67.5 *	50.3	62.8 *
(10) NFUE (kg·kg−1)	32.31	38.92 *	29.18	39.28 *	30.75	39.1 *

** and * represent p < 0.01 and p < 0.05 between the young oasis field (YOF) and old oasis field (OOF) at the same depth, respectively.

. Maize N uptake and NFUE in the OOF were significantly higher than those in the YOF (p < 0.05). No significant difference was found between the YOF and the OOF in soil N accumulation at the 100–200 cm layer (p > 0.05). There were significant differences between the YOF and the OOF in N-leaching losses below 100 and 200 cm. Maize N uptake was the largest single item in N consumption, which accounted for 45.2 and 60.4% of N input in the YOF and OOF, respectively.

4. Discussion

4.1. Impact of Soil Texture on Water Balance and WUE

Variability in SWC results from the balance of irrigation, precipitation, evaporation, root water absorption, and deep drainage [27,28]. Deep drainage from a soil profile depended predominantly on the soil permeability coefficients of individual soil layers and on irrigation intensity [22]. The soil permeability coefficient is mainly determined by soil texture under a certain irrigation intensity. In general, sandy soils with strong soil water repellency lead to high soil permeability coefficients, with preferential flow [4,22] which increases water sequestration and the possibility of solute movement via macropore flow and funnel flow, resulting in a lower WUE [29,30]. Fine soil texture with high organic matter and clay contents enhances crop WUE by decreasing deep percolation [15]. Loamy and clayey textures result from high silt and clay, which act as binders to promote the development of soil aggregates and water retention [5,31]. In our study, SW 0-100 cm deep in the OOF was significantly higher than that in the YOF (p < 0.05); however, deep drainage in the YOF was 6.22 times higher than in the OOF (Table 3), resulting in significantly higher WUE_ET and WUE_Irrig in the OOF than in the YOF (p < 0.05) (Table 6). In the study area, a 60% increase in clay and silt content in the YOF could increase SW within the rooting zone by 87.2%, reduce deep drainage 6.22 times, and increase WUE_ET by 20.9% and WUE_Irrig by 24.1% following irrigation strategies implemented by local farmers. Bruand et al. [32] also observed that sandy soils may reduce percolation by 65–236% and increase water productivity by 30–50% when clay and silt content increases from 15% to 60%. Therefore, field management measures should be implemented in the YOF that are conducive to improvement in soil water retention.

Soil moisture has a strong response to irrigation and is more sensitive in upper soil in arid regions [28,33]. The contour lines of SWC in this study at 0–40 cm were relatively intensive and had columnar distribution, indicating that SWC fluctuated sharply after irrigation; this phenomenon was more noticeable in the YOF than in the OOF (Figure 4), suggesting that sandy soils are more sensitive to changes in irrigation because sandy soils have smaller porosity and relatively uniform particle-size distribution compared to loam soils [34]. The annual average ET of the OOF was greater than that of the YOF (Table 3). This discrepancy was likely due to sandy soil with low field capacity, where the surface becomes dry 1–2 days after irrigation, effectively reducing surface evaporation [33]. SW below the rooting zone of the YOF and the OOF was 462.3 and 241.4 mm (Table 3), accounting for 60.4% and 31.6% of irrigation, respectively, indicating that with current irrigation strategies, irrigation exceeds crop water demand, especially in YOF. Therefore, soil texture must be taken into consideration when deciding on irrigation strategies.

4.2. Impact of Soil Texture on N Balance and NFUE

To understand the use of N fertilizer, a detailed calculation of N accumulation in soil profile, N deep leaching, plant N uptake, and other potential losses is required [27,35,36]. Yield, a part of N uptake, is an important criterion for evaluating NFUE and land productivity. In this study, yield in the OOF was significantly higher than that in the YOF (p < 0.05) (Table 5), likely due to the differences in soil [37]. Soil properties, such as soil texture, control both water and N transport, and therefore the partitioning of N between plant uptake, retention in soil, leaching, and losses by denitrification, which eventually affect crop yield and NFUE [6,12,38].

NO₃⁻-N leaching loss occurs when NO₃⁻-N accumulation and water percolation exist within a soil profile [27]. NO₃⁻-N concentrations 120–180 cm deep in the YOF were higher than those in the OOF after day three, and NO₃⁻-N peaked 1–2 days earlier in the 0–100 cm soil layers in the YOF (Figure 5). NO₃⁻-N leaching loss in the YOF was significantly higher than that in the OOF (p < 0.05); however, NO₃⁻-N accumulation within the root zone in the OOF was significantly higher than that in the YOF (p < 0.05) (Table 5). Sand content was strongly negatively correlated with N uptake (p < 0.01), and it was positively correlated with N leaching (p < 0.01). These results indicated that a fine soil texture is conducive to N retention and crop growth, resulting in significantly higher N uptake and NFUE of the OOF (p < 0.05) (Table 5 and Table 6). A sandy texture will lead to excessive water infiltration and NO₃⁻-N leaching, which reduces N-use efficiency [4]. Maize grown in the loamy texture was more responsive to N uptake than that grown in the sandy texture [15]. In our study, N-leaching loss in the YOF accounted for 38.6% of applied N (Table 6), which was much higher than that in other areas of China (2.3–2.8% in Guanzhong Plain [36]; 8.6–16.8% in Loess Plateau [19]; 9.6–14% in North China Plain [27]; 18% in Hetao Irrigation District [39]. Therefore, the management of soil texture and improvements in soil structure and nutrient leaching are needed for sustainable agricultural and environmental management in sandy soils.

The amount of applied N fertilizer (330 kg·hm⁻²), which exceeded crop N uptake (by 185.5 kg·hm⁻² in the YOF and by 247.5 kg·hm⁻² in OOF), resulted in 127.4 kg·hm⁻² (YOF) and 48.1 kg·hm⁻² (OOF) of applied N leaching below the rooting zone (Table 6) and becoming a potential source of nitrate pollution in groundwater [30]. N fertilizer with site-specific management is an effective strategy to improve N-use efficiency and reduce the risk of leaching [40], but soil texture should be considered. Nitrate leaching decreased with an increased soil clay content. Crop yields in a silty loam were more responsive to increases in N [41], while crops grown in soils with low water- and nutrient-holding capacity may exhibit a lack of response to the application of N fertilizer [40]. However, excessive application of N fertilizer is a common practice for high yields, with some farmers applying 450–500 kg·hm⁻²·yr⁻¹ to achieve high yields in young reclaimed oasis fields, greatly increasing the risk of groundwater contamination [17]. These results make a strong case for incorporating soil textural information into an optimal fertilization strategy.

4.3. Regional Soil and Water Management Practices

An increasing demand for food production fueled intensive land conversion from the native desert to irrigated croplands in arid regions of northwestern China. The cultivated area in Zhangye irrigation district in 2010 was 2896 km². Nearly half of this was young oasis fields cultivated in the past 50 years in sandy lands in the periphery of the oasis, resulting in sandy fields playing an important role in agricultural production [16,17]. However, we found in previous studies that sandy soils still maintained high sand content after 40 years of cropping, and soil aggregate, SOC, and nutrients were insufficient to support high crop production. We propose that YOFs will contribute to water and fertilizer waste for a long time. Therefore, it is urgent to mitigate this negative impact through informed field management.

Best management practices (BMPs) can reduce fertilizer and irrigation inputs, improve crop yield and productivity, and reduce the potential risks of groundwater contamination in sandy soils [30]. A number of management practices have been proposed to improve soil structure and reduce infiltration in sandy soils, including enhanced aggregation [42]; the use of soil amendments such as biochar, clay [43], and organic fertilization in fertilized soils; conservation tillage; crop and grass rotation [13]; and mulching [19]. Water-saving irrigation, including micro-irrigation [44], drip irrigation [14], and border irrigation [39], may be another way to improve soil productivity by reducing water loss and deep drainage to improve N retention.

5. Conclusions

Two-year field experiments were conducted in the Heihe irrigation district in arid northwestern China to investigate the effects of soil texture on crop yield, NO₃⁻-N leaching, and water and nitrogen productivity. Under the current irrigation and fertilization regimes, crops grown in the OOF had a higher crop yield, total N uptake, residual N in 0–100 cm soil layer, WUE_ET, WUE_Irrig, and NFUE and a lower deep drainage and NO₃⁻-N leaching than the YOF (Table 3, Table 4, Table 5 and Table 6). Therefore, sustainable agricultural production requires appropriate field management for YOFs to increase land productivity and reduce the loss of water and nitrogen resources.

Neglecting capillary rise returning nitrogen to the root zone, residual N and N leaching below the root zone will eventually enter groundwater with irrigation due to the relatively high water table in the study area. However, there were differences in N loss due to the differences in soil texture across different cultivation spans. Therefore, how to quantify the contribution rate of N loss to the N load of groundwater under different soil textures is an important topic for future agricultural non-point-source pollution research in the study area. More research is needed to explore the effects of different soil textures on maize productivity and nitrogen retention.