1. Introduction
Sustainable irrigation practices in irrigated crops are crucial crop production management decisions in Florida sandy soils. One of the most encouraging strategies for achieving this objective might be reducing water volumes during certain stages of crop development [
1]. A deficit in water or fertilizers in an active growth stage could decrease tree yield and fruit quality. The severity of deficit irrigation on trees depends on the fruit growth stage [
2]. Deficit irrigation during the first and second stages of fruit maturity has no significant effect on yield [
3]; however, during the third stage it reduces fruit size [
4,
5]. Earlier studies have reported that plant water use in the third stage was determined to be considerably greater than that in the second stage [
6], with an estimated value of around 120% of the evapotranspiration rate (ET). Thus, an accurate evaluation of crop water demands (irrigation amount and timing) during the growth period can maximize the crop yield and lower water losses and crop stress [
7].
Improving irrigation management approaches necessitates a proper understanding of crop water use and physiological responses under deficit irrigation conditions. Numerous studies have monitored the citrus physiological response to different irrigation volumes utilizing plant water stress measures including stomatal conductance [
8,
9] and the midday stem water potential [
10,
11,
12,
13]. The midday stem water potential (Ψ
stem) is affirmed as one of the most steady, dependable, and accurate plant water indicators for irrigated crops [
14,
15]. Irrigation scheduling based on soil moisture and plant water condition sensors can improve crop water productivity and enhance water conservation practices. Soil sensors can also provide information on when to restart irrigation after precipitation events by interpreting the soil water depletion rate. Recommended optimum irrigation scheduling was recommended at allowable soil water depletion in the root zone, around 20% of field capacity [
16].
Volumetric soil water content measurements have long been used for irrigation management [
17,
18,
19,
20,
21,
22]. Capacitance sensors linked with data-logger systems can be used to provide continuous measurements of volumetric water contents in real-time and facilitate accurate irrigation times and volumes to irrigate crops according to their water demands [
23,
24,
25,
26]. The sensors can inform growers when to provide irrigation and monitor the real-time soil moisture contents [
27]. Thus, the use of sensors is critical for scheduling irrigation and maximizing the water supply efficiency in irrigated plants.
Sustainable irrigation approaches in the water-scarce era aim at reducing water losses and maximizing water productivity. The irrigation demands of citrus trees differ with climatic conditions, soil type, and cultivar [
28]. Less rainfall normally results in greater irrigation demands; however, even in a remarkably wet year, noteworthy irrigation may be required due to sporadic rainfall distribution in subtropical climate regions, including Florida [
29,
30]. Supplemental irrigation is necessary to improve citrus yield [
25].
Crop evapotranspiration (ETc) and crop coefficient estimates (Kc) are the fundamental steps for improving crop water productivity [
31]. FAO-56 Penman-Monteith, a major procedure, was broadly approved in measuring crop evapotranspiration (ETc) with a single crop coefficient (Kc) [
32]. Several researchers have published a wide range of seasonal crop coefficients in citrus trees estimated between 0.30 and 1.25 [
33,
34,
35]. Besides this, annual citrus evapotranspiration is estimated to be 1143 mm and the citrus annual irrigation requirement is estimated to be between 381 and 432 mm depending on precipitation and distribution [
36]. Determining a sustainable irrigation method requires developing a sufficient time for irrigation requirements for young citrus trees. In the previous publication of the research in this series, we demonstrated that an irrigation rate of 81% ETc significantly improved citrus tree growth and root development [
37]. Investigating the decreased irrigation water application effect on young citrus tree water relations at higher densities has not been studied in Florida. Thus, a field-scale investigation in the current study was performed to determine if the current citrus irrigation practices need to be revised with different citrus planting densities.
Understanding the impact of different irrigation rates on citrus trees’ water relation parameters is crucial for their sustainable management under conditions of limited water resources. Thus, the objectives were to determine the amount of irrigation required to grow young citrus trees at different planting densities and determine the water supply influence on water relation parameters, including the stem water potential of young citrus trees and soil moisture content in addition to soil salinity.
2. Materials and Methods
2.1. Site Description and Experimental Setup
The project was conducted between November 2017 and September 2020 on ‘Valencia’ (
Citrus sinensis) trees budded on the ‘US-897’ (Cleopatra mandarin x Flying Dragon
trifoliate orange) citrus rootstock located at the University of Florida, Southwest Florida Research and Education Center (SWFREC) demonstration grove located at Immokalee, FL, USA (lat. 26.42° N, long. 81.42° W) [
37]. Citrus trees were planted in Immokalee fine sandy soil (sandy, siliceous, hyperthermic Arenic Alaquods) in November 2017 at different planting densities (
Table 1).
The experiment comprised of five 165 m-long blocks. Each block contained six plots; each plot was ≈27 m long and 7.4 m or 9.2 m wide for two-row or three-row blocks, respectively. Three planting densities in two-row blocks presenting 447, 598, and 745 trees per ha were replicated four times each, and three planting densities in three-row blocks presenting 512, 717, and 897 trees per ha were replicated six times. Each density row was watered to supply 62% or 100% of the crop water supplied (ETc) recommended by the Citrus Irrigation App for young citrus trees during the first 15 months (January 2018–March 2019), then adjusted according to the daily soil moisture reading from soil sensors due to the higher soil moisture contents. Trees were irrigated daily to keep the soil moisture contents above 80% of field capacity at least for the higher irrigation rates under the current study, and the irrigation schedule from the Citrus Irrigation App was determined every two weeks. Irrigation schedules (time for irrigation) were adjusted by adding or subtracting minutes from the citrus irrigation app estimates based on the daily soil moisture sensors measurements. Irrigation rates were divided into four irrigation treatments from April 2019 through the remainder of the project to provide 26.5%, 40.5%, 53%, or 81% of daily ETc (more details are presented in the first publication of this series) [
37] based on the Citrus Irrigation App using data from the Florida Automated Weather Network (FAWN,
http://fawn.ifas.ufl.edu/tools/irrigation/citrus/scheduler/) station located about 100 m away from the demonstration site at SWFREC. The irrigation was divided into three irrigation schemes, each satisfying two plots (two replications) within the block.
Trees were watered using the 360-degree micro-sprinkler (Maxijet Inc., Dundee, FL, USA), with one emitter per one tree placed ≈ 33 cm from the trunk for tree densities at 447, 512, 598, and 717 trees per ha or between two trees for 745 and 897 trees per ha to meet the proposed irrigation treatment at different flow rates. Irrigation was provided with a 172 KPa pressure pump to wet a circular area of the soil surface with 3.6 m diameter per soil surface. Irrigation was halted during precipitation events greater than 17 mm day−1 during summer seasons during the experimentation period.
2.2. Meteorological Measurements
In the current study, irrigation was programmed in keeping with calculated crop evapotranspiration, ETc, determined with a single crop coefficient procedure [
32]. Reference FAO Penman–Monteith evapotranspiration (ETo) was acquired from the FAWN station located within 100 m from the experimental grove. Thus, the estimated Kc was calculated from the water supply data (ETc) presented in this paper and FAWN from November 2017 to August 2020. Crop coefficient (Kc) estimates were performed using Equation (1):
2.3. Water Relations
2.3.1. Soil Moisture Contents
In the current study, the soil water contents were determined using ECH2O 5TE soil sensors (Meter Environment, Pullman, WA) every 30 min from April 2018 to August 2020 for each treatment at 15, 30, and 45 cm soil depth to monitor the soil water movement and water percolation below the root zone. Sixty ECH2O 5TE soil sensors were connected with 12 EM-50G data loggers (Meter Environment, Pullman, WA). All the sensors were calibrated for volumetric soil water content using soil samples collected after the installation [
38] of data loggers on 17 April 2018.
2.3.2. Stem Water Potential
The stem water potential (Ψ
stem) was examined during May, August, and November of 2018 and 2019 and September 2020, as described by Hamido et al. [
10], using a Portable Plant Water Status Console (Soil moisture Equipment Corp, Model 3115, Santa Barbara, CA). The stem water potential was determined on fully expanded sun-exposed matured leaves between 12:00 and 14:00. Three exemplary leaves per tree (two trees per sub-plot) were randomly chosen 24 h before the determinations and coated with a plastic and aluminum foil to enable the water potential of leaves and stems to be balanced. A razor-sharp cutter was utilized to crop leaf petioles nearby to the stem and set into the pressure chamber instantly to evade any changes. Then, the pressure was increased at 1 MPa/30 s using condensed nitrogen till the outflow of water was visible.
2.4. Chemical Analysis
2.4.1. Irrigation Water Quality
Duplicate water samples were collected from the irrigation well at the main pump during August 2018 and 2019. Water electrical conductivity (EC) and total dissolved solids (TDS) were obtained using an electrical conductivity meter (AB 30, Fisher Scientific, Pittsburgh, PA).
2.4.2. Soil Salinity (EC)
In October 2018 and May 2020, soil samples were collected from the demonstration grove by taking three soil samples per depth (0–15, 15–30, and 30–45 cm) from each sub-plot (60), then they were mixed and a sub-sample was chosen from each depth at different irrigation treatments. The EC of the soil extracts was found using the soil paste method recommended by Hanlon et al. [
39]. A total of 20 g of moist soil samples was placed in a 100 mL cup and 40 mL of distilled water was added. The water–soil solution was stirred, allowed to equilibrate for four hours, filtered through Whatman no. 42 filter paper, then the soil salinity (EC) was measured with an electrical conductivity meter (model AB30; Fisher Scientific, Hampton, NH) in µS cm
−1.
2.5. Experimental Design and Statistical Analysis
This experiment was carried out as a factorial of 2 (62% or 100% ETc) or 4 (26.5%, 40.5%, 53%, or 81% ETc) irrigation treatments × 6 planting densities (447, 512, 598, 717, 745, or 897 trees per ha) in a complete randomized block design. Treatments were replicated at least four times. Data were analyzed using the appropriate Statistical Analysis System (SAS for Windows, Ver. 9.4, SAS Institute Inc., Cary, NC). All the measures collected at the three depths including soil moisture contents were not independent and were analyzed as a repeated measure. The analysis of variance (ANOVA) was employed to determine treatment effects on the measured stem water potential and yearly soil salinity. Statistical differences among means were plotted using the general linear model procedure (PROC GLM). The least significant difference test (LSD) was used to segregate the mean of the main effect at p < 0.05.