Soil quality is a crucial component in assessing land sustainability practices in ecosystems [1
]. The concept of soil quality is rapidly developing and it involves the study of soil responses to management practices, as well as its resilience to natural forces. Doran and Zeiss [2
] defined soil quality as “the ability of soil to function as a vital living system, within ecosystems and land-use boundaries, to sustain plant and animal productivity, maintain or enhance water and air quality, and promote plant and animal health.” Soil quality integrates the inherent properties often associated with soil forming factors, as well as the dynamic properties resulting from human-driven management decisions [3
]. Soil quality indicates the ability of soil to function effectively and continuously in the present and in the future [4
Soil functions as the major source for food and fiber production, and interrelates and/or interacts with the surrounding environment. Thus, the enhancement and preservation of the soil quality in arid and semi-arid regions is critical for sustaining land productivity under different land uses and management [5
]. Seasonal variability in the soil functions and conditions are due to factors such as climate pattern, land use, cropping sequences, and farming systems [6
]. Arid and semi-arid regions are characterized by low precipitation rates of 100–250 and 250–500 mm year−1
, respectively, and the development and formation of soils in these regions is affected by water shortage, daily temperature variations, deflations by wind, and reduced microbial activities [7
Sampling for soil quality assessment is crucial to the recommendation of management practices for the overall soil improvement. The chemical soil testing that growers are familiar with are often related to the application of optimal amounts of fertilizer for crop growth and development. In New Mexico, chemical soil sampling is usually done in the spring, just before the planting of the summer annual crops. In the case of winter cash crops, soil testing is done in the fall, just before the crop establishment. As soil quality extends beyond the chemical soil assessment, other measurements belonging to the soil physical and biological attributes need to be tested for the overall assessment of soil quality or soil health. However, there is no consensus regarding the time of the year that many of the soil physical and biological measurements should be tested.
Previous studies have shown the variability that could exist, as soil quality indicators are sampled during the different months of the year within a given growing season [8
]. Moebius et al. [8
] studied the temporal variations of many soil physical quality indicators within a season, in different tillage and crop rotation trials, and found that the indicators vary significantly with the sampling dates and in the different soil types. Piao et al. [11
] studied the seasonal response of soil to microbial biomass carbon. They found an inverse relation between the soil microbial biomass carbon and the atmospheric temperature over one year of sampling. They also found that the microbial biomass carbon was the highest in winter and the lowest in the summer. Ryan et al. [10
] studied the changes in the total soil organic matter, labile carbon, and microbial biomass carbon during the season. They found that the organic matter decreased from 1.48% in February to 1.15% in August after cropping. They also found that the labile carbon followed a similar pattern.
All of these studies show the uniqueness of the variability that can exist within or between seasons for multiple soil quality indicators (SQIs), and that the seasonal variation of the different indicators measured were of different magnitudes; while some indicators remained relatively stable during the year, others fluctuated significantly. This highlights the need to study the across-season variations of the potential SQIs within a given ecosystem and climatic zone, before a reliable soil quality assessment can be developed. The within-season variation may even become complicated by soil and crop management differences.
As a successful soil quality evaluation depends on the selection of suitable and reliable soil indicator measurements, it is crucial to assess the performance of the potential indicators, as they vary from season to season during the year, in order to help with identifying the sensitivity of the SQI and to develop an appropriate sampling strategy for the SQI assessment.
There is a research gap, particularly in the selection and performance of soil quality indicators in irrigated arid farmlands. These farmlands contribute significantly to agricultural production in the United States and worldwide. This study is a contribution towards understanding the across-season variations that could exist for multiple soil measurements under the irrigated arid farming systems.
The objectives of this study were as follows:
To evaluate the variations of the selected soil measurements as a function of the sampling dates that correspond to the fall, winter, spring, and summer, in an arid agro-ecosystem; and
To evaluate how three different cropping systems affect the variability of these soil measurements.
2. Materials and Methods
2.1. Study Site and Treatments
The study was conducted at the New Mexico State University Leyendecker Plant Science Research Center, located in southern New Mexico, United states of America (32°11′5″ N, 106°44′26″ W). The crop management practices chosen for the study and the corresponding soil types are presented in Table 1
. The landform in all of the locations was flood plain, and the percent of calcium carbonate in the soil ranged from 2 to 10%.
The mean annual air temperature of the study region is 25 °C, with a mean maximum temperature of 35 °C in June, and a mean minimum temperature of −2.2 °C in January. The mean annual rainfall is about 250 mm. About 45% of the annual precipitation is concentrated within the monsoon period (July–September), while the rest of the precipitation is scattered from October to January.
Three crop production systems were sampled, including alfalfa (Medicago sativa) (multi-year field crop), upland cotton (Gossypium hirsutum) (an annual row crop), and pecan (Carya illinoinensis) (permanent tree crop). The alfalfa fields used in this study were in their third year of production. The alfalfa fields are harvested seven times in a year, and the harvested biomass is removed from the field as forage. The cotton fields used for this trial undergo yearly plowing, disking, and bed shaping before seeding. At the time of this study, the cotton fields had been growing continuous cotton for three years. The pecan fields were eight years old and in the fourth year of production at the time of this study. While the alfalfa and pecan fields were flood irrigated, the cotton field was furrow irrigated, and the irrigation frequency (around every three to four weeks) for each field was determined by the crop water demands, and was designed to minimize the moisture stress. Both the cotton and pecan fields receive synthetic fertilizer inputs of urea each year. In addition, the pecan fields also received chelated zinc applications. The cultural management practices, including tillage, irrigation, fertilization, and weed controlling, were based on the standard practices recommended by New Mexico State University.
A solution of 12% chelated zinc was applied yearly to the pecan orchards by foliar spray a total of four times per season, twice in April and twice in May, during the study period, at a rate of ~169 L ha−1 (~18 gallon/acre/application).
Nitrogen fertilizer applications took place in the cotton fields three times during the spring season (April, May, and June), including one time as dry (46–0–0) in late April at the rate of ~112 kg h−1 (100 pound/acre), and two times as liquid (32–0–0), with the irrigation at the rate of ~187 L ha−1 (20 gallon/acre/application) in May and June.
Intrepid Edge, an insecticide, was applied to the pecan fields in mid-May at a rate of about 0.8 L ha−1 (11 ounces/acre). Intensity, an herbicide, was applied to the cotton field in June at a rate of about 0.88 L ha−1 (12 ounces/acre).
2.2. Soil Sampling and Laboratory Analyses
The soils were collected from three locations and every location had three replicate fields of the three tested crop production systems, with the fields having soil textures of sandy loam and clay loam. The soil sampling took place in the fall (October 2015), winter (January 2016), spring (April 2016), and summer (July 2016). The April sampling in the cotton fields took place after tillage, but before planting, while the sampling in October took place just after the cotton harvest. Ten random soil samples (0–0.15 m) were collected per field using a standard soil auger to form a composite sample.
The samples collected from the field were air dried and analyzed for different laboratory measurements. The only measurement taken in the field was the soil bulk density (BD), measured by the soil core method at the surface depth (0–0.1 m), using the AMS soil core sampler (AMS Company, American Falls, ID, USA). The sampler core cup was driven into the soil by a slide hammer, and the moist weight of the soil was determined immediately, before drying the soil to a constant weight in the oven at 105 °C for 24 h. The bulk density was determined as the ratio of the dry weight of the soil to the soil volume (core cup volume) [13
The laboratory measurements included the dry aggregate size distribution comprising three parameters, which are the mean weight diameter (MWD), aggregates 2–4 mm (AGG >2 mm), and aggregates less than 0.25 mm (AGG <0.25 mm), all of which were measured using a Tyler RX-29 Rotap sieve shaker (W.S. Tyler, Mentor, OH, USA). These measurements are related to wind erosion, which commonly occurs in the study region. The higher the MWD and AGG >2 mm, the less sediments are lost to wind erosion, while the higher the AGG <0.25 mm, the more sediments are lost to wind erosion [14
]. Wet aggregate stability (WAS), which is an indicator of resistance to soil erosion by water, was measured using the Cornell Sprinkle Infiltrometer [15
]. To measure WAS, the known weight of the soil aggregates are placed on a 2-mm sieve, and a simulated storm of known energy from the Cornell Sprinkle Infiltrometer was allowed to fall on the aggregates from a height of 0.6 m. The percentage of aggregates left unbroken after the storm, on top of the 2-mm sieve, is regarded as the WAS. The available water capacity (AWC) (moisture difference between 0.03 MPa and 1.5 MPa) was measured using the pressure plate apparatus method [16
The soil samples that passed through a 2-mm sieve were used for the soil chemical analyses. The electrical conductivity (EC), pH, calcium (Ca), magnesium (Mg), and sodium (Na) were measured in the saturated paste extract [17
]. The soil organic matter (SOM) was determined using Walkley–Black method [18
], and the permanganate oxidizable carbon (POXC) was determined using the permanganate oxidation technique [19
]. The POXC is a measure of labile carbon in the soil and it has been shown to be a very useful indicator of the soil quality in agricultural systems [20
Nitrate nitrogen (NO3-N) was determined in a water extract (1:5, soil:water), using a cadmium reduction column [21
]. The Olsen bicarbonate method was used to determine the extractable phosphorus (Olsen-P) [22
], and the potassium content (K) was also determined in a water extract (1:5), using inductively coupled plasma spectroscopy [23
]. Copper (Cu), iron (Fe), manganese (Mn), and zinc (Zn) were measured by using the Diethylenetriamine Pentaacetic Acid (DPTA) extraction [21
]. The Sodium Adsorption Ratio (SAR) was calculated from the measurements of Ca, Mg, and Na in the saturated paste extract.
2.3. Statistical Analysis
The experimental design was a repeated measure within a randomized complete block design, with three replications. The statistical significance of the season (time of sampling) and crop management effects on different soil measurements were analyzed using the PROC GLM (general linear models) procedure in SAS (Statistical Analysis System) [24
], and the means were separated by Duncan’s Multiple Range test. In addition, the coefficient of variation (CV), which is the ratio of standard deviation to the mean, was calculated for the measurements collected across the different sampling periods.
The soil quality in the Southwestern United States was studied by sampling at different dates corresponding to the four seasons of the year (winter, spring, summer, and fall), and from three cropping systems. The results showed a significant effect of crop management only on a few soil quality measurements, while the effect of the sampling season was significant for 15 out of the 21 soil measurements. Most of the soil measurements that were assessed exhibited low to moderate variability, while a few had a high variability between the sampling seasons. The measurements exhibiting a high variability present a challenge for estimating the directional changes in the soil quality, as the fluctuations in their values between the sampling dates can act as a confounding factor. For most of the chemical measurements, the sampling time is often dependent on when the soil amendments are to be applied to the crops being grown in the field. However, for the physical and biological measurements, which rely more on management-induced changes, sampling should occur at approximately the same time of the year, in order to capture the long-term changes.