The Effect of K-Fertilization and Irrigation on the Composition of Cultivated Soils: Examples from Israel

Abstract

Evaluation of soil sustainability should take into account chemical and mineralogical changes due to cultivation. Potassium (K) application is a key farming practice that may potentially affect phyllosilicate composition and soil properties. Irrigation is another agent that affects soil composition. This study aims to evaluate the imprints of cultivation on the chemical and mineralogical compositions of lowland soils under semi-arid and Mediterranean climate regimes and to assess them with the natural pedogenesis. The sites examined include two permanent plot experiments with crop rotation, and three single-plant plots. Cultivated, control, and fallow soils were analyzed for their chemical and mineralogical composition. The X-ray diffraction patterns of the clay fraction were decomposed to achieve quantitative detection of changes among the illite–smectite (IS) phases and in kaolinite amounts. In loamy and clayey soils dominated by IS phases, cultivation caused minor changes and diverse behavior of the IS phases. Yet, the K balance was negative under high fertilization levels and higher plant mass production. Fertilization enhanced the natural process of transforming IS into kaolinite and illite in the sandy soil, leading to a positive K balance. This study emphasizes the importance of IS minerals as a dynamic K pool that responds to plant needs.

1. Introduction

Fertilization and irrigation used in conjunction with intensive soil cultivation have the potential to alter soil properties over time. The stability of the soil’s constituents and the rate of pedogenic processes are mainly impacted by the soil moisture regime and chemical environment. Soil mineralogical composition is one of the critical parameters that affect cultivation, especially the phyllosilicate transformation and neoformation by leaching silicon and aluminum, and uptake and release of potassium (K), an essential plant nutrient. Accordingly, in order to achieve optimal production of sustainable soil, K management must be cautiously controlled [1,2]. If the harvested K is not replaced by K released from soil minerals or K fertilizers a negative balance prevails, depleting the soil-available K [3]. Besides optimizing plant production, the addition of K-fertilizers may also improve soil structure and water retention [4,5]. However, K application in excess may cause a reduction in soil pore size that lowers the hydraulic conductivity [6].

A simplified classification of K occurrence in soil is into four groups: water-extractable, exchangeable (adsorbed), slowly exchangeable, and structural forms [7,8]. Most of the soil K generally occurs as structural K in the primary minerals K-feldspars, biotite, and muscovite and is only available to plants through mineral weathering [9]. Structural K has been thought to be of relatively little significance for short-term plant K nutrition [10,11]; however, in the long term, it may play a substantial role in K supply [11,12,13,14]. Water-extractable K and exchangeable K are readily available to plants but typically make up only a small portion of soil K [15] (and references therein). Secondary soil minerals promote reversible K release and fixation, especially the slowly exchangeable K fixed in 2:1 mixed-layer phyllosilicate interlayer sites, namely, illite–smectite (IS) [2,8,16]. The IS phases are acknowledged as the key components for K uptake by plants and, accordingly, for agricultural production [17,18]. The chemical formula of IS is variable, reflecting its position on the smectite-to-illite continuity line. For example, K_0.28Ca_0.12(Al_1.56Fe³⁺_0.16Mg_0.31)[(Si_3.70Al_0.30)O₁₀(OH)₂] is relatively smectitic and K_0.62Ca_0.05) Al_1.78Fe³⁺_0.02Mg_0.25)[(Si_3.39Al_0.61)O₁₀(OH)₂] is relatively illitic.

Short-term changes in soil 2:1 phyllosilicate structure and composition due to K-fertilization over a number of years and even from season to season have been distinguished by X-ray diffraction (XRD) [4,19,20,21]. The reported changes are frequently minor, and the challenge of quantifying them by XRD led to the use of modeling simulations [9,18,22,23] or peak decomposition (deconvolution) techniques [11,17,24,25].

The current study aims to detect potential chemical and mineralogical changes in long-term intensively cultivated soils in Israel. The soils selected for the study are lowland IS-rich soils under semi-arid and Mediterranean climates, which have been irrigated and fertilized by K, nitrogen (N), and phosphorous (P). The lowland soils, where the majority of farming in Israel takes place, include the loess plains of the northwestern Negev, the sandy coastal plain along the Mediterranean Sea, the parallel alluvial plains between the coastal plain and the backbone mountains, and the northern valleys of the intra-mountainous range. Except for the sandy soils of the coastal plain, most cultivated lowlands soils are loamy or clayey. The current study had the opportunity to use two previous permanent plot experimental farms’ soil samples. The initial goal of the experiments was to maximize crop fertilization levels; however, soil K and mineralogy investigations have not yet been carried out. Other active farming sites that were chosen because of their high K application were also studied.

In addition to the major IS phases, kaolinite (Al₂Si₂O₅(OH)₄) is a minor (<25%) constituent in lowland soils of Israel, but in the sandy soils of the coastal plain it can, locally, become the major phyllosilicate. Illite (K_0.65Al_2.0[Al_0.65Si_3.35O₁₀](OH)₂) amounts are small but variable (<15%), chlorite (clinochlore Mg₅Al(AlSi₃O₁₀)(OH)₈) is detectable in trace amounts (<5%), and palygorskite ((Mg,Al)₂Si₄O₁₀(OH)·4H₂O) may be present in small amounts in arid soils [26]. The origin of the fine fractions of the soil in Israel is dust arriving from the Sahara and Arabian deserts [27,28]. In Israel and neighboring countries, pedogenesis promotes the transformation of phyllosilicates in response to local environmental conditions [29,30,31,32]. The major transformation takes place under the Mediterranean climate where illitic IS phases of dust gradually become more smectitic and approach smectite dominance in vertisols. Another transformation route occurs under intensive leaching, especially in porous and well-structured soils, in which IS phases may transform into both kaolinite and illite [33,34].

Hardly any attention has been accounted for by previous studies on the relations of mineralogical and chemical changes brought on by farming with the underlying long-term pedogenic processes, such as the dissolution of primary minerals and calcite, and the subsequent transformation and dissolution/precipitation processes within the soil. In the last decade, there has been an increasing number of studies that use soil archives of long-term experiments, most of them are from the northern hemisphere, mainly the USA and UK, but none are from the Mediterranean region [35]. Moreover, most studies reported changes in SOM and macronutrients [35], whereas only a few studies investigated possible changes in the mineralogy of the soils. In Israel, studying the long-term impact of fertilization on soil productivity and nutrient balances has been performed on two permanent experimental plot stations [36]. Soil chemical analyses during the experiments included available forms of fertilizers (K, N, P) and a few additional ions. No study has examined the effects of long-term farming on the mineralogy of the clay fraction and the bulk mineralogical and chemical compositions. Accordingly, the objectives of the current study were to (1) identify the effects of irrigation and fertilization on lowland soils in Israel, with a focus on the soil K balance and (2) assess the potential relations of agricultural changes with the natural pedological evolution of soils.

2. Materials and Methods

2.1. Sampling Sites

The five sampling sites are presented in Figure 1. Two sites are the experimental stations in Gilat and Bet Dagan of The Agricultural Research Center of Israel. Fertilization studies were conducted there in permanent plots with crop rotation occurring for more than thirty years. Three additional single-crop plots were a tomato plot on sandy soil (Mishmeret) in the Coastal Plain where intensive fertilization, including sheep manure, has been carried out for ten years; a banana plantation (Nir Etzion) in the Carmel Mountain’s foothills where fertilizing with higher than usual amounts of K has been carried out for fifty years; and a fifteen-year-old grapefruit orchard (Sarid) in the northern valleys, including eight years of irrigation by effluents. The basic data of the sites studied are presented in

Table 1. Basic data of the five sites studied: two sites of experimental permanent plots and three sites of single-plant plots. Each soil type includes the USDA taxonomy and the Israeli classification names.

Site Name	Geography/ Geomorphology	Coodinations		Cultivation	Years	Soil Type	pH	Texture	m.a.p.*	Elevation
		E	N						mm	m a.s.l.
Gilat	Northwestern Negev; loessial plain	31.3531	34.6649	Experimental plots	1961–1994	Calcic Haploxeralf	8.2	Loam	275	175
						(Loamy Loess)
Bet Dagan	Eastern Coastal Plain; alluvial plain	32.0124	34.8549	Experimental plots	1961–1993	Chromic Haploxerert	7.8	Silty clay	600	30
						(Vertisol)
Mishmeret	Coastal Plain; sand plain	32.2276	34.9209	Private farm; tomatoes	1965–2009	Typic Rhodoxeralf	7.5	Loamy sand	600	70
						(Hamra)
Nir Etzion	Mt. Carmel foothills; alluvial plain	32.7050	34.9704	Kib. banana plant	1958–2008	Typic Chromoxeret	8.0	Clay	625	30
						(Vertisol)
Sarid	Yizre’el Valley; margins of alluvial plain	32.6589	35.2220	Kib. grapefruit orchard	1992–2007	Typic Chromoxeret	7.9	Clay	500	80

* m.a.p. = mean annual precipitation 1961–1990; Kib. = kibbutz.

and additional information is presented below.

2.1.1. Gilat Permanent Plot Experiment

The main objective of the experiment was to measure the yields of various field crop rotation systems on irrigated deep loess soil as affected by the long-term cumulative application of various combinations of N fertilizer and organic manure. Four levels of cultivation of N fertilizer were applied: a control without N fertilizer, and three dozes with constant ratios of 100%, 200%, and 300% in which the absolute value of the 100% dose was fit to the type of crop that was grown in rotation, labeled: N0, N1, N2, N3, and three organic manure treatments labeled M0 (no organic manure), M1 (dairy manure 30 t/ha), M2 (dairy manure 90 t/ha), and R (city refuse compost 30 t/ha) were applied. The samples selected for the current study are of the N and manure application and are listed in

Table 2. A list of the samples studied, the various soil types, and the levels of fertilization.

(2a) Gilat.
Lab No.	Type		Plot	Year	Fertilizers *
0–20 cm
F0–20	Fallow			2006	-
F1 0–20	Fallow			2008	-
F2 0–20	Fallow			2008	-
D	Control		24	1987	M0, N0
E	Control		24	1987	M0, N0
G	Control		21	1987	M0, N0
J	Control		38	1987	M0, N0
A			3	1987	M0, N3
C			18	1987	M2, N0
H			34	1987	M2, N0
L			43	1987	M2, N0
N			92	1987	M2, N0
B			13	1987	M2, N3
I			36	1987	M2, N3
K			42	1987	M2, N3
M			57	1987	M2, N3
20–40 cm
F 20-40	Fallow			2006	-
F1 20-40	Fallow			2008	-
F2 20-40	Fallow			2008	-
Q	Control		21	1987	M0, N0
R	Control		24	1987	M0, N0
U	Control		38	1987	M0, N0
P			18	1987	M2, N0
S			34	1987	M2, N0
W			43	1987	M2, N0
Y			92	1987	M2, N0
O			13	1987	M2, N3
T			36	1987	M2, N3
V			42	1987	M2, N3
X			57	1987	M2, N3
* N3 = 120–390 kg N ha−1 annual rate, 6450 kg N ha−1 1963–1987 (accumulated); M2 = 0–720 kg (Organic N) ha−1 annual rate, 5520 kg (Organic N), ha−1 + 7200 kg K ha−1 1963–1987 (accumulated); K was applied as KCl 0–230 kg, 2390 kg ha−1 K 1963–1987 (accumulated); P = 0–100 K ha−1 highly variable annual rates. M—manure; N—nitrogen.
(2b) Bet Dagan.
Lab No.	Type	Block	Plot	Year	Fertilizers *
0–20 cm
F1 0-20	Fallow			2006	-
F2 0-20	Fallow			2006	-
F3 0-20	Fallow			2006	-
A	Control	III	9	1965	[N,P,K]0
B	Control	III	9	1993	[N,P,K]0
G	Control	II	73	1995	[N,P,K]0
J	Control	III	9	1963	[N,P,K]0
K	Control	III	9	1996	[N,P,K]0
AE		I	41	1993	N3, P4, K0 + gM
AF		I	76	1996	N3, P4, K0 + gM
AG		V	80	1996	N3, P4, K0 + gM
C		I	85	1993	N4, P4, K2
D		V	89	1993	N4, P4, K2
H		II	86	1964	N4, P2, K2
I		II	86	1995	N4, P2, K2
L		III	20	1963	N4, P3, K2
M		III	20	1995	N4, P3, K2
N		IV	25	1963	N4, P3, K2
O		IV	25	1995	N4, P3, K2
V		V	44	1996	N4, P4, K2 + M
W		I	15	1996	N4, P4, K2 + M
Y		II	41	1996	N4, P4, K2 + M
Z		III	40	1996	N4, P4, K2 + M
20–40 cm
F1 20-40	Fallow			2006	-
F2 20-40	Fallow			2006	-
F3 20-40	Fallow			2006	-
P	Control	II	73	1995	[N,P,K]0
Q	Control	III	9	1996	[N,P,K]0
R		III	20	1995	N4, P3, K2
S		IV	25	1995	N4, P3, K2
T		II	86	1995	N4, P2, K2
AB		I	15	1996	N4, P4, K2 + M
AC		II	41	1996	N4, P4, K2 + M
AD		III	40	1996	N4, P4, K2 + M
* N0, N3 and N4 = 0, 180 and 240 kg N ha−1 annual rate, P0, P3 and P4 0, 36, 72 kg P ha−1 annual rate, K0 and K2 = 0 and 320 K ha−1 annual rate; M—manure; N—nitrogen; P—phosphorous; K—potassium; gM—green manure.
(2c) Mishmeret.
Lab no.	Type		Year	Fertilizers *
0–20 cm
YTF	Fallow		2008	-
YTF a	Fallow		2008	-
YTF b	Fallow		2008	-
YT 1	Tomatoes		2008	Fert. + M
YT 1a	Tomatoes		2008	Fert. + M
YT 1b	Tomatoes		2008	Fert. + M
5–20 cm
YTG	Grass lawn		2009	-
YTG a	Grass lawn		2009	-
YTG b	Grass lawn		2009	-
20–40 cm
YTF	Fallow		2008	-
YTF a	Fallow		2008	-
YTF b	Fallow		2008	-
YT 1	Tomatoes		2008	Fert. + M
YT 1a	Tomatoes		2008	Fert. + M
YT 1b	Tomatoes		2008	Fert. + M
YTG	Grass lawn		2009	-
YTG a	Grass lawn		2009	-
YTG b	Grass lawn		2009	-
* N = 180 kg N ha−1 annual rate as KNO3 and Ca(NO3)2; K = 464 kg K ha−1 annual rate as KNO3 + 15 m3 manure; Fert.—K and N fertilizers; M—sheep manure.
(2d) Nir Etzion.
Lab no.	Type		Year	Fertilizers *
0–20 cm
NE a	Banana 1		2008	Fert.
NE b	Banana 2		2008	Fert.
NE c	Banana 3		2008	Fert.
NE d	Olive 1		2008	-
NE e	Olive 2		2008	-
NE f	Field 1		2008	?
NE g	Field 2		2008	?
NE h	Field 3		2008	?
20–40 cm
NE i	Banana 1		2008	Fert.
NE j	Banana 2		2008	Fert.
NE k	Banana 3		2008	Fert.
NE l	Olive 1		2008	-
NE m	Olive 2		2008	-
NE n	Field 1		2008	?
NE o	Field 2		2008	?
NE p	Field 3		2008	?
* N = 500 kg ha−1 annual rate; K = 1500 kg ha−1 annual rate; Fert.—K and N fertilizers; ?—unknown fertilization.
(2e) Sarid.
Lab no.	Type		Year	Fertilizers *
0–20 cm
F1 N 0-20	Fallow		2007	-
F2 N 0-20	Fallow		2007	-
F3 N 0-20	Fallow		2007	-
A			2007	Effluent + Compost + KCl
B			2007	Effluent + Compost + KCl
C			2007	Effluent + Compost + KCl
20–40 cm
F1 20-40	Fallow		2007	-
F2 20-40	Fallow		2007	-
F3 20-40	Fallow		2007	-
D			2007	Effluent + Compost + KCl
E			2007	Effluent + Compost + KCl
G			2007	Effluent + Compost + KCl
* Estimated annual rates: N = 170 kg ha−1; K = 275 kg ha−1; P = 110 kg ha−1.

a. All plots received basic K and variable P fertilization (Table 2a); the various manure levels introduced also added variable concentrations of K and P. The following crops were rotated in the first years: sugar beets, corn, onions, potatoes, wheat, and onions, and later on other crops [36]. Fallow soil was sampled for the current study.

2.1.2. Bet Dagan Permanent Plot Experiment

The main objective of the experiment was to measure the yields of various field crop rotation systems on irrigated deep alluvial soil as affected by long-term cumulative application of various combinations of N, P, K fertilizers, and organic manure. The following four factors were investigated: (a) five levels of N fertilizer, 0, 60, 120, 180, and 240 kg N ha⁻¹, labeled N0, N1, N2, N3, N4; (b) five levels of P fertilizer, 0, 9, 18, 36, and 72 kg P ha⁻¹, labeled P0, P1, P2, P3, P4; (c) three levels of K fertilizer, 0, 160, and 320 kg K ha⁻¹, labeled K0, K1, K2; organic manure treatments of farmyard, or dairy manure or green manure, usually pea (Table 2b). The following crops were rotated in the first years: wheat, chickpea, sugar beet, oats, cotton, and sorghum. Fallow soil was sampled for the current study. In the permanent plot stations K was determined only as the available K concentration by the NaHCO₃ 0.5 M extract method [36,37].

2.1.3. Mishmeret

The field has been cultivated for decades as part of a private farm. In the past ten years before sampling, the farmer has been growing tomatoes and continuously fertilizing them with sheep manure from a nearby pen (Table 2c). The nearby grass lawn was also sampled in order to compare an irrigation-only regime with an irrigation-and-fertilization regime. Additionally, fallow soil was sampled.

2.1.4. Nir Etzion

This kibbutz banana plantation was chosen to test the effect of intensive fertilization on a single-cropped soil for about fifty consecutive years before sampling. Because banana plants consume about 1100 kg of K per ha⁻¹, the fertilization contains a higher dose of this element (Table 2d). Since there are no fallow soils with the same or comparable qualities, samples from a nearby reaped wheat field and an olive grove were taken for comparison. The grove’s soil closely resembles fallow soil because it was not farmed for years prior to sampling.

2.1.5. Sarid

This kibbutz grapefruit orchard was planted in 1992 and after four years was experimentally irrigated by treated wastewater for ten years more before sampling. The control plots were irrigated with freshwater but were not examined in the current study. Fallow soil was sampled as well (Table 2e). The annual amounts of fertilizers were evenly estimated.

2.2. Sampling and Sample Processing

Three replicates of topsoil (0–20 cm) and subsoil (20–40 cm) samples were collected for each site and each agricultural category. Analysis was done on dried materials that had been sieved through a 2 mm sieve (bulk fraction). For the determination of the bulk mineralogical and chemical compositions, a portion each of <2 mm fraction was mechanically ground. The <1 µm fraction was selected to represent the clay fraction since it was tested to fully maintain the phyllosilicate composition while lowering the quartz amount and thus improving the preferred orientation (Figure 2).

For the clay fraction analysis, a portion of the bulk fraction was agitated in water for several hours, buffered acetic acid was added, and after the carbonates had completely dissolved, it was washed up and given a brief treatment with low-intensity ultrasound. The <1 µm fraction was separated using Stokes Law once a stable suspension was attained, and it was then saturated with SrCl₂. After washing, a thick suspension was pipetted onto glass slides and analyzed after air-drying, glycolation (at least 8 h at 60 °C and cooling overnight), and heating to 550 °C (2 h).

2.2.1. X-ray Diffraction (XRD)

A Philips X-ray diffractometer (1730/1710; Almelo, The Netherlands) was operated with 40 kV and 30 mA, CuKα radiation, 1°, 0.1 mm, 1° slits, and a sealed proportional detector. The bulk mineralogical composition was semi-quantitatively determined on side-loaded powder runs. Semi-quantitative clay composition was calculated from the peak areas of randomly ordered IS (R0), kaolinite, and illite divided by a 4:2:1 factor, respectively [26] (and references therein). The IS expandability was roughly estimated by the saddle value. The higher the saddle value, the higher the amount of interstratified illitic layers. The saddle value measures the ratio between the height from the baseline to the low next to the IS reflection, and the height from the baseline to the top of the IS reflection [26,38]. The bulk and clay fraction results are presented in 5% intervals and sum up to 100 ± 5%.

2.2.2. Quantitative Evaluation of the Clay Fraction

The air-dried X-ray patterns were decomposed into their main components using the DecompXR program [39]. Peak areas were determined as the product of peak height and width at half height. The 2:1 phyllosilicates were considered in the following groups: illite with a well (WCI) and a poorly crystallized component (PCI), illite-rich mixed layered (IS), smectite-rich mixed layered (SI), and chlorite (Ch) (Figure 3).

Illite peaks (WCI and PCI) have peak maxima near the 10 Å position, whereas mixed-layered peak positions range from 12 to 15 Å, depending on the relative proportion of illitic layers. Three indicators were calculated and used in this study: %IS, which is the ratio of the total areas of the illitic phases WCI, PCI, and IS to the total areas of all 2:1 phases. The center of gravity (cg) indicator calculates the weighed position of the (001) of all 2:1 phyllosilicates [16] to provide an estimation of the total amount of illitic layers. These indicators enabled us to compare the change in relative abundance of the potassic clays that are potentially affected by plant growth and fertilizer application. The third indicator is %K. It displays the ratio of the area of the kaolinite, on its three sub-components, to the total area of all reflections in the analyzed area, in order to track the potential formation of pedogenic kaolinite. In most cases, the kaolinite reflection can be decomposed into well-ordered kaolinite and less-ordered kaolinite. However, samples from Israel’s leached red soils may show asymmetry toward the low angle, necessitating the insertion of a third component that represents a kaolinite–smectite (KaS) mixed-layer phase. Kaolinite in all samples was decomposed into three components for the sake of uniformity, but their relative amounts were not measured.

2.2.3. Chemical Composition

Major element composition of the bulk soil fraction was determined by ICP-OES (Perkin Elmer, OPTIMA 3300, Waltham, MA, USA) after lithium metaborate (LiBO₂) fusion of ground sample (0.25 g). Each analysis run included repeated determinations of three of the international standards SO-3, BE-N, BHVO-1, SCo-1, NIM-L, and NIM-G. Errors were <2%.

3. Results

3.1. Bulk Mineralogical Composition

Quartz and phyllosilicates are the major minerals at the four fine-textured sites studied. Quartz is more than twice the amount of phyllosilicates in Gilat and about half the amount of phyllosilicates in Sarid, whereas similar amounts occur in Bet Dagan and Nir Etzion. The quartz amount in the sandy Mishmeret soil is >80% (

Table 3. Bulk mineralogical composition semi-quantitative (%).

Gilat
Lab No.	Type	Quartz	Calcite	Dolom.	K-Felds.	Plagioc.	Phyllos.	Others
0–20 cm
F 0–20	Fallow	45	15–20	<3	10	10	20	Cli.
F1 0–20	Fallow	45	15–20	<2	10	10	15–20	-
F2 0–20	Fallow	55	15–20	<2	5–10	5–10	15–20	-
D	Control	50	15	<2	10	10	15	Am.
E	Control	45–50	15	<2	10–15	10	15–20	-
A	M0, N3	50	15	<2	5	10–15	15–20	Cli.
C	M2, N0	50	10–15	<2	10	10	15	-
N	M2, N0	50	10–15	<3	5	10–15	15–20	Cli. Am.
B	M2, N3	50	15	<3	5–10	5–10	15–20	-
20–40 cm
F 20–40	Fallow	45	15	<5	10–15	10–15	15	Am.
F1 20–40	Fallow	40–45	15	<5	10	15	15–20	Cli. Am.
F2 20–40	Fallow	55	15	<2	10	10	15	-
Q	Control	45	15	5	10–15	10	15	Cli. Am.
R	Control	50	15	<3	10–15	5–10	15–20	Cli.
U	Control	40–45	10	<2	10	15	15	Am.
P	M2, N0	50	15	<2	5	5–10	15–20	Am.
S	M2, N0	45	10–15	<2	15	15	15	Cli. Am.
W	M2, N0	50–55	15	<5	5–10	15	15	Cli. Am.
Y	M2, N0	45	10–15	<3	5–10	15	20	Am.
O	M2, N3	50–55	15	<2	10	10–15	15	Cli. Am.
T	M2, N3	55	10–15	<3	10–15	5	15	Cli. Am.
V	M2, N3	55	15	<2	5	10–15	15	Am.
X	M2, N3	50–55	10	<2	5	15	20	Cli. Am.
Bet Dagan
Sample	Type	Quartz	Calcite	Dolom.	K-Felds.	Plagioc.	Phyllos.	Others
0–20 cm
F1 0–20	Fallow	40–45	10	<1	5	<5	40–45	-
F2 0–20	Fallow	40	10	<1	5	<5	45	-
F3 0–20	Fallow	40	10	<1	<3	5	45	-
A	Control	40–45	5–10	<1	5–10	5	40	-
B	Control	40–45	10	-	5	5–10	40	-
G	Control	40	10	<1	5–10	<5	40–45	-
J	Control	40	10	<1	5–10	5–10	40–45	Am.
K	Control	45	10	-	<3	5	40–45	Am.
C	N4, P4, K2	40	10	<1	<5	5	40–45	-
D	N4, P4, K2	45	5–10	<1	5	5	40	-
H	N4, P2, K2	40	10	<2	5	<5	45	-
I	N4, P2, K2	40	5–10	-	<5	<5	40–45	Am.
L	N4, P3, K2	40–45	10	-	<5	<5	40–45	-
M	N4, P3, K2	40–45	10	-	5	5–10	40	-
N	N4, P3, K2	40–45	10	<2	5–10	<5	40	-
O	N4, P3, K2	45	10	<1	<5	5	40–45	-
V	N4, P4, K2 + M	40–45	5–10	<1	5	<5	40	-
W	N4, P4, K2 + M	40	10	<1	<5	<5	40	-
Y	N4, P4, K2 + M	45	10	<2	<5	<5	40–45	-
Z	N4, P4, K2 + M	45	5–10	<1	<5	5	40–45	-
20–40 cm
F1 20–40	Fallow	40–45	10	<1	<5	5	45	Am.
F2 20–40	Fallow	40	10	<1	<5	5	45	-
F3 20–40	Fallow	40	10	<2	<5	5–10	45	-
P	Control	40–45	10	<1	10	5	45	-
Q	Control	35–40	10	<1	5–10	10	40	-
R	N4, P3, K2	40–45	5–10	<1	5–10	10	40–45	Am.
S	N4, P3, K2	35–40	10	-	5–10	10	40	Am.
T	N4, P2, K2	35–40	10	-	5–10	<5	45	-
AB	N4, P4, K2 + M	35–40	10	<1	5	5–10	45	Am.
AC	N4, P4, K2 + M	40	10	<1	5	5	45	Am.
AD	N4, P4, K2 + M	40	10	-	5	<5	45	Am.
Mishmeret
Sample	Type	Quartz	Calcite	Dolom.	K-Felds.	Plagioc.	Phyllos.	Others
0–20 cm
YTF 0–20	Fallow	85	<0.5	<0.5	5	<5	5	-
YTFa 0–20	Fallow	85–90	<0.5	<0.5	<5	<5	5–10	-
YTFb 0–20	Fallow	90	-	-	<3	<5	5	-
YT1	Tomatoes	85–90	<0.5	<0.5	<5	<5	5–10	-
YT1a	Tomatoes	85–90	<0.5	<0.5	<5	<3	5–10	-
YT1b	Tomatoes	85	<0.5	-	<3	5	10	-
5–20 cm
YTG	Grass lawn	85–90	-	<0.5	5	<3	5–10	Gyp.
YTG a	Grass lawn	80	-	-	5	5–10	5–10	Gyp.
YTG b	Grass lawn	80–85	<5	?	5	<3	10	-
20–40 cm
YTG	Grass lawn	85–90	-	-	<5	<3	5–10	Goe.
YTG a	Grass lawn	90	-	-	<5	<3	10	Gyp.
YTG b	Grass lawn	85–90	<0.5	-	<5	<3	5–10	-
Nir Etsion
Sample	Type	Quartz	Calcite	Dolom.	K-Felds.	Plagioc.	Phyllos.	Others
0–20 cm
NE a	Banana 1	40	10	<0.5	<3	<3	40	Goe.
NE b	Banana 2	45	20	<1	5	<5	30	-
NE c	Banana 3	40	5–10	<1	5	5	45	Goe.
NE d	Olive 1	45	15	-	<3	<5	30–35	Goe. Am.
NE e	Olive 2	40–45	15	<0.5	<5	<5	35–40	-
NE f	Field 1	40	15	<0.5	<5	5	35–40	Goe.
NE g	Field 2	40	15	<0.5	<5	<5	35–40	-
NE h	Field 3	45	15	-	<3	<5	35	-
20–40 cm
NE i	Banana 1	35–40	10	-	5	5	40–45	Am.
NE j	Banana 2	45	20	-	<5	<3	30	-
NE k	Banana 3	40–45	5–10	<3	<5	5	45	Am.
NE l	Olive 1	40–45	15	-	<5	5	40–45	-
NE m	Olive 2	40	15	<1	5	<3	35–40	-
NE n	Field 1	40	15	-	5	<5	35–40	Goe.
NE o	Field 2	35–40	15	-	5	5	40–45	Goe.
NE p	Field 3	40–45	25	?	<3	<3	30–35	Goe.
Sarid
Sample	Type	Quartz	Calcite	Dolom.	K-Felds.	Plagioc.	Phyllos.	Others
0–20 cm
F1 0–20	Fallow	30	20	-	?	<5	45	-
F2 0–20	Fallow	25	20–25	<3	5	5	45	Am.
F3 0–20	Fallow	25	20–25	<1	<3	<3	50	-
A	ECK	25–30	20	-	<5	<3	45–50	-
B	ECK	25–30	20–25	<1	?	?	45–50	-
C	ECK	25–30	20	<1	<3	<3	45–50	-
20–40 cm
F1 20–40	Fallow	25–30	20	-	-	<5	50	-
F2 20–40	Fallow	25–30	20–25	-	<5	5	50	-
F3 20–40	Fallow	25–30	20–25	-	-	<3	50	-
D	ECK	25	20	-	5	<5	50–55	-
E	ECK	25–30	20–25	-	<3	<3	50	-
G	ECK	25–30	20	-	<3	-	50	-

ECK = Effluent + Compost + KCl; Abbreviations: dolom. = dolomite, K-felds. = K-feldspar; Plagioc. = plagiocalase; Phyllos. = phyllosilicates; Cli. = clinoptilolite; Am. = amphibole; Gyp. = gypsum; Goe. = goethite; ? = suspected.

). Calcite is a minor mineral whose amount declines from 15%–20% in the semi-arid loess soil to about 10% in the northern sites under the Mediterranean climate, apparently due to the increase in rainfall. Exceptionally, the northernmost Sarid site has a calcite amount of 20%–25%, which is higher than the typical 5%–10% calcium amount of similar nearby soils, e.g., [4,40]. The feldspars group is of minor or trace amounts, declining significantly from south to north (Table 3).

The semi-quantitative results do not allow for distinguishing between fallow and cultivated plots, or between topsoil and subsoil, in any of the sites studied. However, the calcite amount in Gilat is exceptional, as it is on average higher in fallow than in all other cultivated samples due to leaching by irrigation.

3.2. Chemical Composition

The bulk chemical composition (

Table 4. Chemical composition (w/w %).

(4a) Gilat
0–20
	F 0-20	F1 0-20	F2 0-20	D	G	J	A	C
SiO2	59.4	60.7	61.9	65.2	64.6	64.6	62.5	65.0
Al2O3	6.9	6.4	6.3	6.7	6.9	7.0	6.9	6.3
Fe2O3	3.3	3.0	3.4	2.8	3.2	2.9	3.1	3.0
TiO2	0.8	0.8	0.7	0.8	0.8	0.8	0.8	0.8
CaO	13.2	12.0	11.9	10.1	10.1	10.2	11.3	10.0
MgO	2.0	1.9	1.8	1.7	1.7	1.8	1.8	1.6
MnO	0.05	0.1	0.1	0.05	0.06	0.06	0.05	0.05
Na2O	0.7	0.8	0.8	0.7	0.9	0.9	0.8	0.7
K2O	1.2	1.3	1.2	1.2	1.2	1.3	1.1	1.1
P2O5	0.1	0.1	<0.1	0.1	0.1	0.2	0.1	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	12.8	12.7	12.5	10.6	10.1	10.3	11.9	10.9
Total	100.5	99.7	100.5	100.0	99.7	100.0	100.4	99.7
	H	L	N	B	I	K	M
SiO2	64.5	63.5	62.2	63.8	63.8	65.0	64.1
Al2O3	6.9	6.9	7.1	6.6	6.8	6.9	7.0
Fe2O3	2.9	2.9	2.9	2.7	3.4	3.3	3.3
TiO2	0.8	0.8	0.8	0.8	0.8	0.8	0.8
CaO	10.3	10.4	11.5	10.5	10.2	9.9	9.7
MgO	1.7	1.8	1.9	1.7	1.7	1.7	1.7
MnO	0.06	0.06	0.06	0.05	0.06	0.06	0.06
Na2O	0.9	0.9	0.8	0.8	0.9	0.9	0.9
K2O	1.3	1.3	1.3	1.2	1.2	1.2	1.3
P2O5	0.2	0.2	0.2	0.2	0.2	0.2	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	10.5	10.6	11.5	11.1	10.8	10.0	10.3
Total	100.0	99.3	100.3	99.5	99.9	100.0	99.4
20–40
	F 20-40	F1 20-40	F2 20-40	Q	R	U	P
SiO2	59.2	62.0	60.8	59.2	63.9	64.0	61.4
Al2O3	6.8	6.6	6.6	6.9	6.7	7.0	6.6
Fe2O3	2.8	2.8	3.2	2.9	2.9	3.1	2.5
TiO2	0.8	0.8	0.8	0.8	0.8	0.9	0.8
CaO	13.4	11.5	12.2	13.3	11.1	10.2	12.4
MgO	1.8	1.7	1.9	1.8	1.7	1.7	1.7
MnO	0.06	0.1	0.1	0.06	0.05	0.06	0.05
Na2O	0.8	0.8	0.8	0.6	0.6	0.6	0.6
K2O	1.2	1.2	1.2	1.2	1.3	1.3	1.2
P2O5	0.1	0.1	0.1	0.1	0.1	0.2	0.1
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	13.3	11.9	12.5	12.9	11.0	10.8	12.1
Total	100.3	99.5	100.1	99.8	100.2	99.9	99.5
	S	W	Y	O	T	V	X
SiO2	63.0	63.5	61.0	63.3	63.0	64.0	64.0
Al2O3	7.0	7.0	7.1	6.9	6.8	7.0	7.3
Fe2O3	3.0	3.1	3.0	3.1	2.7	2.9	3.2
TiO2	0.9	0.9	0.8	0.8	0.8	0.9	0.9
CaO	10.6	10.5	11.9	10.7	10.8	10.4	10.0
MgO	1.8	1.7	1.8	1.7	1.7	1.7	1.7
MnO	0.06	0.06	0.06	0.06	0.06	0.06	0.06
Na2O	0.6	0.7	0.6	0.7	0.6	0.6	0.6
K2O	1.3	1.3	1.3	1.3	1.3	1.3	1.4
P2O5	0.2	0.2	0.2	0.2	0.2	0.2	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	11.7	10.5	12.0	10.9	11.5	10.7	10.4
Total	100.2	99.5	99.8	99.7	99.5	99.8	99.8
(4b) Bet Dagan
0–20
	F1	F2	F3	A	B	G	J	K	C	D
SiO2	62.6	62.0	61.9	63.1	62.6	62.9	63.4	63.6	61.6	64.3
Al2O3	11.1	11.0	11.0	10.5	10.4	11.0	10.9	10.8	10.2	10.5
Fe2O3	5.6	5.5	5.4	5.4	5.2	5.7	5.4	5.5	5.6	5.2
TiO2	1.3	1.2	1.3	1.2	1.2	1.3	1.3	1.4	1.2	1.3
CaO	5.8	6.5	6.1	5.8	5.8	5.9	6.0	5.8	5.8	5.2
MgO	1.6	1.7	1.7	1.5	1.5	1.7	1.7	1.7	1.5	1.4
MnO	0.12	0.10	0.12	0.11	0.10	0.12	0.12	0.12	0.10	0.11
Na2O	0.5	0.4	0.5	0.5	0.4	0.5	0.5	0.5	0.4	0.4
K2O	1.2	1.1	1.2	1.1	1.1	1.1	1.1	1.2	1.0	1.1
P2O5	0.2	0.1	0.2	0.1	0.1	0.1	0.1	0.2	0.2	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	9.8	10.2	10.0	10.8	11.2	9.4	8.9	9.1	11.8	9.7
Total	99.8	99.8	99.4	100.1	99.6	99.7	99.4	99.9	99.4	99.4
	H	I	L	M	N	O	V	W	Y	Z
SiO2	62.1	62.0	63.4	64.2	63.7	64.9	63.6	61.5	64.4	64.4
Al2O3	11.8	11.3	10.8	10.9	10.7	10.8	10.7	11.1	10.7	10.7
Fe2O3	5.9	5.7	5.3	5.2	5.2	5.4	5.2	5.9	5.2	5.2
TiO2	1.4	1.4	1.4	1.4	1.3	1.4	1.3	1.3	1.3	1.3
CaO	6.0	5.5	6.0	5.4	5.8	5.5	6.0	6.4	5.3	5.3
MgO	1.7	1.7	1.6	1.6	1.6	1.5	1.5	1.6	1.5	1.5
MnO	0.13	0.13	0.12	0.12	0.12	0.12	0.12	0.13	0.12	0.12
Na2O	0.4	0.4	0.5	0.5	0.6	0.5	0.5	0.5	0.5	0.5
K2O	1.2	1.2	1.2	1.2	1.1	1.2	1.1	1.2	1.2	1.2
P2O5	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	9.4	9.9	9.0	8.8	9.1	8.7	9.9	10.0	9.3	9.3
Total	100.2	99.4	99.5	99.5	99.4	100.2	100.1	99.8	99.7	99.7
20–40
	F1 20-40	F2 20-40	F3 20-40	P	Q	R	S	T	AB	AC	AD
SiO2	60.7	60.4	60.5	62.5	64.0	64.6	64.3	62.0	61.5	64.1	60.6
Al2O3	11.5	11.6	11.6	11.4	10.8	10.7	10.7	11.5	11.5	10.7	11.4
Fe2O3	6.5	5.9	5.8	5.8	5.4	5.7	5.3	6.3	6.0	5.9	6.1
TiO2	1.3	1.3	1.3	1.3	1.3	1.3	1.3	1.3	1.3	1.3	1.3
CaO	6.3	6.5	6.5	6.1	5.9	5.4	5.4	5.6	6.5	5.7	6.2
MgO	1.7	1.7	1.7	1.7	1.6	1.5	1.5	1.6	1.7	1.5	1.7
MnO	0.13	0.12	0.13	0.12	0.12	0.12	0.12	0.12	0.13	0.12	0.13
Na2O	0.3	0.3	0.4	0.3	0.3	0.3	0.3	0.3	0.6	0.6	0.5
K2O	1.2	1.2	1.2	1.1	1.2	1.2	1.2	1.2	1.1	1.1	1.2
P2O5	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2	0.2
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
LOI	10.2	10.3	10.2	9.5	9.1	9.1	9.2	9.6	9.8	9.0	10.7
Total	100.0	99.5	99.5	100.0	99.9	100.1	99.5	99.7	100.3	100.2	100.0
(4c) Mishmeret.
0–20							5–20
	YTF	YT a	YTFb	YT-1	YT-1a	YT-1b	YTG	YTGa	YTGb
SiO2	90.5	91.1	92.0	88.8	89.7	88.2	89.0	88.5	89.8
Al2O3	3.4	3.6	3.4	4.2	4.0	4.7	3.8	3.7	3.7
Fe2O3	1.5	1.7	1.4	2.2	1.8	2.3	1.3	1.4	1.4
TiO2	0.3	0.3	0.2	0.5	0.5	0.5	0.3	0.5	0.3
CaO	0.7	0.7	0.5	0.7	0.6	0.6	0.7	0.6	0.8
MgO	0.3	0.3	0.2	0.3	0.3	0.3	0.2	0.2	0.2
MnO	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Na2O	0.3	0.2	0.3	0.3	0.2	0.2	0.2	0.2	0.2
K2O	0.7	0.6	0.7	0.7	0.7	0.7	0.5	0.6	0.5
P2O5	0.1	< 0.1	≤ 0.1	0.2	0.2	0.2	<0.1	<0.1	<0.1
L.O.I.	2.1	1.8	1.7	2.6	2.4	2.6	3.7	3.7	3.0
Total	99.8	100.4	100.4	100.4	100.4	100.3	99.7	99.4	99.9
20–40
	YTF 20-40	YTFa 20-40	YTFb 20-40	YT-1	YT-1a	YT-1b	YTG	YTGa	YTGb
SiO2	89.2	89.7	90.8	89.0	90.0	88.7	89.4	88.3	90.5
Al2O3	4.3	4.2	3.7	4.0	3.8	4.3	4.4	5.0	4.2
Fe2O3	1.9	1.9	1.7	2.3	2.0	2.0	1.8	1.9	1.8
TiO2	0.4	0.4	0.3	0.4	0.5	0.5	0.4	0.5	0.3
CaO	0.6	0.5	0.5	0.6	0.5	0.7	0.5	0.5	0.4
MgO	0.3	0.3	0.2	0.3	0.2	0.3	0.2	0.2	0.2
MnO	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Na2O	0.4	0.3	0.3	0.3	0.3	0.3	0.2	0.2	0.1
K2O	0.7	0.6	0.6	0.7	0.7	0.7	0.6	0.6	0.5
P2O5	0.2	0.2	0.2	0.2	0.2	0.2	<0.1	<0.1	<0.1
L.O.I.	2.3	1.9	1.7	2.6	2.3	2.6	2.3	2.4	2.1
Total	100.3	100.0	100.0	100.4	100.5	100.3	99.8	99.6	100.0
(4d) Nir Etsion.
0–20
	NEa	NEb	NEc	NEd	NEe	NEf	NEg	NEh
SiO2	54.5	56.2	57.5	50.9	52.6	51.4	51.9	50.5
Al2O3	13.1	8.5	13.7	11.8	11.1	12.6	12.4	12.5
Fe2O3	7.0	4.6	8.0	6.0	6.4	6.9	6.5	7.3
TiO2	1.4	0.8	1.4	1.2	1.2	1.3	1.2	1.3
CaO	6.4	12.6	4.6	10.5	9.3	8.6	9.0	9.6
MgO	1.8	1.3	1.8	1.4	1.4	1.8	1.8	1.8
MnO	0.2	0.1	0.2	0.1	0.1	0.1	0.1	0.1
Na2O	0.4	0.4	0.4	0.3	0.3	0.3	0.3	0.3
K2O	1.4	0.9	1.4	1.1	1.1	1.4	1.5	1.6
P2O5	0.3	0.1	0.2	0.1	0.1	0.3	0.3	0.3
LOI	13.7	14.8	11.1	16.7	16.8	15.0	15.3	15.0
Total	100.1	100.4	100.2	100.1	100.4	99.6	100.4	100.2
20–40
	NEi	NEj	NEk	NEl	NEm	NEn	NEo	NEp
SiO2	55.5	56.5	57.0	54.4	52.5	50.8	52.0	45.3
Al2O3	13.2	9.2	14.0	13.0	12.5	12.8	13.3	11.7
Fe2O3	7.7	4.6	8.1	6.8	6.8	6.9	8.0	6.2
TiO2	1.4	0.9	1.4	1.3	1.3	1.3	1.3	1.1
CaO	6.0	12.2	5.4	8.9	10.1	10.3	8.3	15.2
MgO	1.7	1.2	1.7	1.5	1.4	1.7	1.8	1.6
MnO	0.2	0.1	0.2	0.1	0.1	0.1	0.1	0.1
Na2O	0.3	0.4	0.4	0.3	0.3	0.3	0.3	0.3
K2O	1.3	0.9	1.4	1.0	1.0	1.2	1.1	1.0
P2O5	0.2	0.1	0.2	0.1	0.1	0.3	0.2	0.2
LOI	13.0	14.5	10.7	13.1	14.4	14.8	14.0	17.8
Total	100.4	100.5	100.4	100.5	100.5	100.4	100.4	100.5
(4e) Sarid.
0–20
	F1 0–20	F2 0–20	F3 0–20	A	B	C
SiO2	42.4	42.3	42.3	44.2	42.0	42.0
Al2O3	10.5	10.9	10.7	11.3	10.6	10.8
Fe2O3	5.4	6.2	5.5	6.6	5.6	6.3
TiO2	1.0	1.1	1.1	1.1	1.0	1.0
CaO	16.2	15.2	16.1	13.9	14.9	15.0
MgO	1.9	2.2	2.0	2.2	2.1	2.3
MnO	0.11	0.11	0.11	0.12	0.13	0.12
Na2O	0.3	0.3	0.3	0.3	0.3	0.4
K2O	1.0	1.0	1.1	1.2	1.2	1.4
P2O5	0.2	0.3	0.2	0.6	1.1	0.7
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	0.1
L.O.I	20.7	20.2	20.0	18.5	20.7	20.2
Total	99.7	99.8	99.4	100.0	99.6	100.3
20–40
	F1 20-40	F2 20-40	F3 20-40	D	E	G
SiO2	47.3	46.1	45.3	47.1	46.0	46.4
Al2O3	11.9	11.6	11.4	11.9	11.4	11.7
Fe2O3	7.0	6.9	6.9	7.0	6.2	7.0
TiO2	1.1	1.1	1.1	1.1	1.1	1.1
CaO	11.8	12.0	13.5	11.6	12.8	11.4
MgO	2.2	2.2	2.1	2.3	2.3	2.3
MnO	0.1	0.1	0.1	0.1	0.1	0.1
Na2O	0.3	0.3	0.3	0.3	0.3	0.3
K2O	1.0	1.1	1.0	1.3	1.3	1.3
P2O5	<0.1	0.1	0.1	0.4	0.6	0.4
SO3	<0.1	<0.1	<0.1	<0.1	<0.1	<0.1
L.O.I	17.3	18.5	18.7	17.3	18.1	18.2
Total	100.0	100.0	100.4	100.4	100.0	100.3

) essentially reflects the mineralogical composition: the higher the quartz, the higher the SiO₂, and the higher the phyllosilicates, the higher the Al₂O₃. Thus, the sandy Mishmeret soil has the highest average SiO₂ concentration of 90% and the lowest Al₂O₃ concentration of 4.0%. The clayey Sarid soil has the lowest average SiO₂ concentration of 43% and the highest Al₂O₃ concentration of 11%. This relationship is not straightforward, as phyllosilicates also comprise SiO₂, which is higher in smectitic IS than in illitic IS or kaolinite. Thus, Gilat and Bet Dagan soils have the same average SiO₂ concentration of 63% but different quartz amounts due to the higher phyllosilicate amount and the smectitic nature of the IS in Bet Dagan. CaO directly reflects the amount of calcite, as its concentration in phyllosilicates and plagioclase is rather marginal, and the amount of dolomite is negligible. Fe₂O₃ is strongly associated with phyllosilicates, as indicated by its correlation with Al₂O₃ (Figure 4). The K₂O average concentration K₂O is 1.1%–1.2% in four sites, indicating K is rather stable in the clayey soils studied, whereas in the Mishmeret sandy soil, it is 0.6%. The K₂O correlation with Al₂O₃ (Figure 5) exhibits a complex behavior that is discussed below.

Differences in the bulk chemical composition between fallow and cultivated plots are quite few. CaO decreases in cultivated soils due to irrigation. It is lower by 13% in Gilat, 8% in Bet Dagan, Nir Etzion, and Sarid’s topsoils. Sarid is the only site where a decrease in CaO is distinguished in the subsoil.

In general, average K₂O concentrations are comparable in all sites between topsoil and subsoil and between fallow and cultivated soils. In Gilat, there is a minor rise of 4% in average K₂O concentration and K₂O/Al₂O₃ ratio, from topsoil to subsoil, which may be related to some clay eluviation. In Gilat, Bet Dagan, and Mishmeret, average K₂O concentrations for fallow and cultivated soils are comparable, but the lawn samples in Mishmeret show a remarkable decline of 24%. The K₂O concentrations in the banana plots in Nir Etzion are variable but on average they are higher than those in the olive grove in both depths (but lower than those in the wheat field topsoil). Though fertilization data on the wheat field are missing, the highest K₂O and P₂O₅ concentrations in its topsoil and a 27% K₂O depletion in the subsoil are evidence for fertilization during the growing season. Sarid is the only site where the cultivated soil is enriched by K₂O, with an average concentration higher than fallow by 24%. This is accompanied by an average P₂O₅ concentration, which is four times greater than that in the fallow soil due to fertilization and irrigation by the wastewater. Another noteworthy change in chemical composition is observed in Mishmeret, where Al₂O₃ is 10% higher in fertilized plots compared to fallow and lawn soils, suggesting a rise in phyllosilicate amount in both depths.

3.3. Clay Fraction Mineralogical Composition

The dominant phyllosilicate mineral is IS, and kaolinite is the secondary one in the four fine-textured soils (

Table 5. Mineralogical composition of the clay fraction (<1 µm); semi-quantitative (%).

Gilat
Sample	IS	Saddle	Kaolinite	Illite *	Chlorite	Palygorskite	Quartz
0–20 cm
F 0-20	50–55	0.83	15–20	10	<3	15	?
F1 0-20	60–65	1	15–20	5–10	<3	10	-
F2 0-20	60–65	0.86	15–20	5–10	<3	10	-
D	50–55	0.98	25	5–10	<3	10	?
E	50–55	0.98	25	10	<3	10	?
G	55	0.95	20–25	5–10	<3	5–10	<1
J	50–55	0.92	25	10	<3	5–10	<1
A	55	0.93	20	10	<3	10	?
B	50	0.97	20–25	10	<3	10	?
C	50–55	0.93	20–25	10	<3	10	?
H	55	0.97	20–25	10	<3	5–10	?
I	50–55	0.94	25	5–10	<3	10	<1
K	55	0.92	20	10	<3	10	<1
L	50	1	20–25	10–15	<3	10	?
M	55	0.94	25	10	<3	5–10	-
N	50–55	0.99	20	5–10	<3	10–15	?
20–40 cm
F 20-40	55	0.86	20	5–10	<3	10	?
F1 20-40	60-65	0.95	15–20	5–10	<3	10	?
F2 20-40	55	0.75	20	5–10	<3	10	?
Q	50	0.83	20–25	10–15	<3	10–15	-
R	50	0.86	20	10–15	<3	10	-
U	60	0.92	20	5	<3	10	-
O	50–55	0.9	20–25	10	<3	10	-
P	50	0.82	20	10	<5	10	<2
S	55	0.92	20	5–10	<3	10–15	?
T	50–55	0.88	20	10	<3	15	?
V	55-60	0.91	20	5	<3	10	?
W	45–50	0.87	20–25	10–15	<3	10–15	<2
Y	50–55	0.91	20	10	<3	10–15	?
Bet Dagan
Sample	IS	Saddle	Kaolinite	Illite	Chlorite	Quartz	Goethite
0–20 cm
F1 0-20	80–85	0.39	15	<3	<3	?	-
F2 0–20	80–85	0.53	10–15	<5	<3	?	?
F3 0–20	80–85	0.46	10	<5	<3	?	?
A	80	0.51	10–15	<5	<3	?	?
B	80	0.56	10–15	<5	<3	?	?
G	80	0.41	15	<5	<3	-	?
J	85	0.37	10–15	?	<3	-	?
K	80–85	0.44	10–15	?	<3	?	?
C	80–85	0.50	10–15	<5	<3	?	?
D	80	0.55	10–15	<3	<3	?	?
E	80–85	0.54	10	<5	<3	?	?
H	80–85	0.35	10–15	?	<3	?	-
I	85	0.45	10	<3	<3	?	?
L	80	0.38	10–15	<3	<3	<1	<3
M	80–85	0.41	10–15	<3	<3	?	?
N	80–85	0.39	10–15	<3	<3	?	?
O	80	0.38	15	<5	<3	?	?
V	80–85	0.46	10–15	<3	<3	-	?
W	80	0.44	15	<3	<3	-	-
Y	80–85	0.49	10–15	?	<3	-	?
Z	80–85	0.50	10–15	<3	<3	-	-
20–40 cm
F1 20-40	80–85	0.50	10–15	<3	<3	<1	<3
F2 20-40	80–86	0.48	15	?	<3	?	?
F3 20-40	80–86	0.41	10–15	<3	<3	?	-
P	85	0.42	15	?	<3	-	-
Q	80	0.31	15	<3	<3	?	-
R	80–85	0.45	10–15	<3	<3	<1	<3
S	80–85	0.39	10–15	<3	<3	-	?
T	85	0.42	10–15	<3	<3	?	?
AB	85	0.47	10	?	<3	?	?
AC	85	0.48	10	?	<3	-	-
AD	85	0.44	10	?	<3	?	-
AE	80–85	0.51	10	<3	<3	<1	<3
AF	80	0.50	10–15	<3	<3	<1	<3
AG	80	0.45	10–15	<3	<3	<1	<3
Mishmeret
Sample	IS	Saddle	Kaolinite	Illite *	KaS	Quartz	Goethite
0–20 cm
YTF1 0-20	40	>1	35	15	~5	<1	<5
YTF1a 0-20	40	>1	40	15	~5	<1	<5
YTF1b 0-20	40	>1	35	15	~5	<1	<3
YT1	25	>1	45	20	~5	<3	~5
YT1a	25	>1	45	20	~5	<3	~5
YT1b	30	>1	40	20	~5	<3	~5
5–20 cm
YTG 5-20	55	>1	35	5–10	<5	<1	<3
YTG a 5-20	55	>1	35	5–10	<5	?	<3
YTG b 5-20	55–60	0.98	30–35	5–10	<5	?	<3
20–40 cm
YTF1 20-40	45	>1	40	10	~5	<1	<3
YTF1a 20-40	50	>1	35	10	~5	<1	<3
YTF1b 20-40	40	>1	40	15	~5	<1	<3
YT1	30	>1	45	15	~5	<3	~5
YT1a	25	>1	50	15	~5	<3	<5
YT1b	35	>1	40	15	~5	<3	~5
YTG	60	0.95	30–35	5–10	<5	<1	<3
YTG a	60	>1	30	5	<5	<1	<5
YTG b	60–65	>1	30	<5	<5	?	<5
Nir Etsion
Sample	IS	Saddle	Kaolinite	Illite	Chlorite	Quartz	Goethite
NE a	75	0.72	15	5	<3	?	?
NE b	75–80	0.68	10–15	5	<3	-	?
NE c	80	0.67	10–15	<5	<3	<1	?
NE d	75–80	0.72	10–15	<5	<5	<1	-
NE e	80	0.72	10–15	<5	?	<1	-
NE f	75–80	0.62	10–15	<5	<5	?	?
NE g	75–80	0.72	10–15	<5	<3	-	-
NE h	75	0.61	15	<5	<5	<1	<3
20–40 cm
NE i	75	0.68	15	5–10	<5	<1	?
NE j	75–80	0.65	15	5	<5	<1	?
NE k	75	0.70	15	5	<3	?	-
NE l	75–80	0.67	10–15	<3	5	<1	?
NE m	75	0.57	15	5	<3	<1	?
NE n	75	0.61	15	<5	<5	?	?
NE o	80	0.56	15	<5	?	-	-
NE p	75	0.63	10–15	5	<5	?	-
Sarid
Sample	IS	Saddle	Kaolinite	Illite	Chlorite	Quartz	Goethite
0–20 cm
F1 0-20	85	0.66	10	<3	<3	?	?
F2 0–20	85	0.57	10–15	<5	<3	?	?
F3 0–20	85	0.65	10–15	<3	<3	-	-
A	80–85	0.64	10–15	<3	<3	-	?
B	85	0.67	10–15	<3	<3	-	-
C	85–90	0.76	10–15	<3	<3	-	-
20–40 cm
F1 20-40	80	0.55	15	<5	<3	-	-
F2 20-40	80–85	0.50	15	<3	<3	-	-
F3 20-40	80–85	0.49	10–15	<3	<3	-	-
D	80–85	0.51	15	5	<3	?	-
E	80–85	0.50	10	<5	<3	?	-
G	80–85	0.59	10–15	<3	<3	-	-

* Illite +Illitic IS; ? = suspected.

), as in most soils in Israel (examples in [41], where it is termed montmorillonite; [26]). In many of the sandy Mishmeret soil samples, however, kaolinite is the major mineral, like in some red sandy soils along the coastal plain [42]. The saddle value, which estimates the illite-like layers in the IS phases, is rather variable (Figure 2). The highest value, >1, indicating the highest amount of illitic layers, is in Mishmeret. It is 0.82–1 in Gilat, 0.49–0.76 in Sarid, 0.56–0.72 in Nir Etzion, and 0.31–0.55 in Bet Dagan. The amount of illite decreases from south to north: 5%–15% in Gilat, ~5% in Nir Etzion, and <5% in Bet Dagan and Sarid, excluding Mishmeret, where it is 10%–20%. Palygorskite is present only in Gilat (Figure 2) in amounts of 5%–15%. Chlorite is present in trace amounts (<5%) in the fine-textured soils, whereas in the sandy soil (Mishmeret), it is undetectable. The latter site is distinguished also by the presence of the rare mixed-layered mineral kaolinite–smectite (KaS), which has already been formerly recorded in small amounts in some well-leached red soils in Israel [32,43].

The semi-quantitative composition of the clay fraction does not allow for the detection of changes between fallow and cultivated soil or between topsoil and subsoil in the Gilat and Bet Dagan experimental permanent plots. A slightly higher average illite amount in the banana plots than that of the olive grove and the wheat field is observed in Nir Etzion. In contrast, prominent changes are observed in Mishmeret, where the fallow soil has higher illite and kaolinite amounts relative to the other sites studied. The highest illite and kaolinite amounts occur in the heavily fertilized tomato plots, indicating that cultivation enhances the natural pedogenic processes. The grass lawn, however, has similar kaolinite but lower illite amounts than the fallow soil. In the three fertilization levels, illite is higher in the topsoil than in the subsoil.

3.4. Relative Quantitative Analysis of the Clay Fraction

The values of the three indicators, %IS, cg, and % Ka, described above in Material and Methods, were calculated for all samples and are presented in

Table 6. Relative quantitative indicator of the clay fraction that were calculated after decomposition by the DecomXR program.

Gilat
Sample	Type	%Ka	cg °2θ	%IS
0–20 cm
F 0-20	Fallow	26.5	6.65	63.0
F1 0-20	Fallow	15.5	6.76	43.4
F2 0-20	Fallow	12.4	6.63	42.9
D	Control	16.9	6.64	49.8
E	Control	14.4	6.83	60.9
G	Control	15.7	6.35	40.0
J	Control	16.8	6.49	46.6
A	M0, N3	14.5	6.85	58.2
C	M2, N0	15.0	6.85	62.3
H	M2, N0	14.4	6.40	45.5
L	M2, N0	15.4	6.44	43.6
N	M2, N0	14.6	6.68	51.7
B	M2, N3	17.0	6.83	66.3
I	M2, N3	14.9	6.43	46.3
K	M2, N3	14.3	6.57	53.1
M	M2, N3	15.5	6.50	43.6
20–40 cm
F 20-40	Fallow	13.1	6.64	47.4
F1 20-40	Fallow	12.2	6.58	40.1
F2 20-40	Fallow	12.8	6.48	45.8
Q	Control	14.4	6.52	52.4
R	Control	13.1	6.67	66.0
U	Control	15.4	6.65	73.4
P	M2, N0	16.0	6.62	56.8
S	M2, N0	14.5	6.69	68.7
W	M2, N0	13.7	6.64	59.1
Y	M2, N0	14.3	6.78	72.6
O	M2, N3	15.2	6.47	51.2
T	M2, N3	13.8	6.56	62.3
V	M2, N3	14.9	6.69	63.1
X	M2, N3	14.8	6.75	58.1
Bet Dagan
Sample	Type	%Ka	cg °2θ	%IS
0–20 cm
F1 0–20	Fallow	6.7	6.04	22.1
F2 0–20	Fallow	6.5	6.10	22.0
F3 0–20	Fallow	6.1	6.18	28.3
A	Control	6.3	6.01	21.7
B	Control	5.6	5.93	28.8
G	Control	6.2	5.74	24.4
J	Control	6.2	5.90	17.6
K	Control	6.6	6.15	28.0
C	N4, P4, K2	6.5	6.02	20.3
D	N4, P4, K2	6.4	6.03	20.0
H	N4, P2, K2	6.0	5.69	23.4
I	N4, P2, K2	6.0	5.64	19.2
L	N4, P3, K2	5.6	5.99	23.9
M	N4, P3, K2	6.8	5.79	20.2
N	N4, P3, K2	6.6	5.80	27.8
O	N4, P3, K2	6.3	5.79	28.5
V	N4, P4, K2 + M	6.1	5.65	15.7
W	N4, P4, K2 + M	7.2	5.83	21.5
Y	N4, P4, K2 + M	6.8	5.77	18.4
Z	N4, P4, K2 + M	6.2	5.70	14.0
20–40 cm
F1 20-40	Fallow	6.4	6.22	29.0
F2 20-40	Fallow	7.2	6.24	37.8
F3 20-40	Fallow	7.6	6.25	39.7
P	Control	5.9	6.09	38.6
Q	Control	7.0	6.19	34.5
R	N4, P3, K2	6.8	6.22	39.6
S	N4, P3, K2	6.4	6.22	29.0
T	N4, P2, K2	8.1	6.15	28.0
AB	N4, P4, K2 + M	7.3	6.31	29.6
AC	N4, P4, K2 + M	6.5	6.27	33.6
AD	N4, P4, K2 + M	7.0	6.17	28.6
Mishmeret
Sample	Type	%Ka	cg °2θ	%IS
0–20 cm
YTF1 0-20	Fallow	35.8	7.23	65.0
YTF1a 0-20	Fallow	28.2	7.19	68.1
YTF1b 0-20	Fallow	37.3	6.99	59.3
YT1	Tomatoes	43.1	7.91	87.9
YT1a	Tomatoes	51.1	7.88	80.6
YT1b	Tomatoes	36.9	7.68	89.2
5–20 cm
YTG 5-20	Grass lawn	24.1	6.27	57.8
YTG a 5-20	Grass lawn	25.6	6.03	55.7
YTG b 5-20	Grass lawn	26.9	5.99	52.7
20–40 cm
YTF1 20-40	Fallow	37.0	6.28	56.5
YTF1a 20-40	Fallow	34.3	6.35	57.0
YTF1b 20-40	Fallow	45.8	6.63	52.5
YT1	Tomatoes	44.4	6.56	57.0
YT1a	Tomatoes	52.9	6.64	62.2
YT1b	Tomatoes	50.3	6.49	51.7
YTG	Grass lawn	28.6	6.05	53.9
YTG a	Grass lawn	30.9	6.13	52.4
YTG b	Grass lawn	25.7	6.27	64.8
Nir Etsion
Sample	Type	%Ka	cg °2θ	%IS
0–20 cm
NE a	Banana 1	7.7	5.75	15.0
NE b	Banana 2	6.3	5.81	12.4
NE c	Banana 3	6.8	5.91	13.6
NE d	Olive 1	6.1	5.86	17.0
NE e	Olive 2	5.8	5.86	19.7
NE f	Field 1	7.1	5.95	23.6
NE g	Field 2	7.1	5.95	23.6
NE h	Field 3	6.5	6.05	22.0
20–40 cm
NE i	Banana 1	8.5	6.70	32.5
NE j	Banana 2	8.1	6.84	25.7
NE k	Banana 3	8.4	6.83	26.9
NE l	Olive 1	7.1	6.23	28.8
NE m	Olive 2	7.2	6.30	27.7
NE n	Field 1	7.1	6.24	28.8
NE o	Field 2	7.1	6.23	24.6
NE p	Field 3	7.3	6.51	42.3
Sarid
Sample	Type	%Ka	cg °2θ	%IS
0–20 cm
F1 0–20	Fallow	5.6	5.79	29.7
F2 0–20	Fallow	5.1	5.80	26.5
F3 0–20	Fallow	5.7	6.06	28.8
A	ECK.	5.6	5.65	21.9
B	ECK.	4.7	5.67	24.0
C	ECK.	5.6	5.76	22.6
20–40 cm
F1 20-40	Fallow	5.7	6.02	31.6
F2 20-40	Fallow	5.8	6.17	34.2
F3 20-40	Fallow	5.8	6.20	34.4
D	ECK.	5.4	5.98	31.9
E	ECK.	6.3	6.21	26.5
G	ECK.	6.1	6.20	32.6

. The values of the three indicators in Gilat topsoil, where the clay fraction is dominated by illitic IS, are rather scattered (Figure 6). It might be, in part, an analytical issue, a consequence of the overlap of the main reflection of palygorskite with that of PCI. However, in the subsoil, the three indicators display a similar trend as follows: an increase from fallow to control and medium-level fertilized plots, and then a decrease in the cg and %IS values in the highly fertilized plots, suggesting that where plant growth was maximal, the K balance was negative.

In Bet Dagan, the clay fraction is dominated by smectitic IS, and neoformation of smectite is evidenced by the cg indicator, whose values gradually decrease along with the level of fertilization in the topsoil (Figure 6). However, in the subsoil, it decreases from fallow to control plots but gradually increases with the level of fertilization. This increase is probably due to higher clay amount and lower permeability, which promote K trapping and absorption by the smectitic IS. As most K activity here occurs in the smectitic realm, the %IS values are rather low and uniform. However, in the highly fertilized plots, a decrease in the %IS values occurs in both depths due to the disintegration of the illitic IS into both smectite and kaolinite. The %Ka values decrease from fallow to control plots in both depths and gradually increase with the level of fertilization, mainly in the topsoil. This decrease suggests that irrigation here enhances the natural neoformation of smectitic IS at the expense of kaolinite and other minerals, but that fertilization promotes kaolinite formation.

The three indicators in the one-growth plots are first examined in the clayey soils of Nir Etzion and Sarid, which share similar properties with Bet Dagan (e.g., the same range of the cg values). The cg and %IS values in the topsoil of the Nir Etzion banana plots are similar to or lower than those of the olive grove and the wheat field (Figure 6). Both indicators increase in the subsoil of the three soil types, especially the cg of the banana plot. It is worth noting that the %IS indicator is the highest in the wheat field topsoil, along with the highest K₂O/Al₂O₃ values in this site. Though the fertilization schedule of the wheat field is unknown, it turns out that the K application was surplus. The %Ka values of the banana plots in both depths are higher than those of the olive grove and the wheat field, reflecting kaolinite neoformation due to the intensive irrigation.

The cg values in Sarid are higher in the subsoil than in the topsoil, indicating natural illitization is taking place in depth, as in Nir Etzion. The cg and %IS values in the topsoil of the orchard are lower than those of the fallow, suggesting a negative K-balance, whereas in the subsoil the %IS in the orchard is slightly lower, but the cg values are equal to those of the fallow, maintaining an almost neutral K balance in depth. The %Ka values are the lowest among all sites studied so no difference between the fallow and the cultivated plots in both depths is detectable.

The sandy Mishmeret soil is unique also for its three clay indicators (Figure 6). The %Ka values are the highest among the five sites studied, and in the tomato plot, they are ~25% higher than in the fallow soil in both depths, indicating that cultivation enhances the natural kaolinite neoformation. Comparably, the %IS values are the highest and increase in topsoil, along with cg values, from fallow to the tomato plot, but decrease in the grass lawn, as no K is added to it. The fallow values show that much of the illitization processes take place mostly at the topsoil, so in the subsoil, the cg pattern returns but that of the %IS does not.

In summary, the three indicators display a variety of responses to irrigation and fertilization. The effect of irrigation is evident in Gilat, Bet Dagan, and Mishmeret. A positive K balance due to K fertilization is evident by increased illitization in Bet Dagan, Mishmeret, and Nir Etzion’s topsoils, whereas in Gilat and Sarid the K balance is negative. Enhancement of natural clay neoformation by cultivation is evident in Bet Dagan, Mishmeret, and Nir Etzion. A mutual increase in %Ka and %IS predominates in Mishmeret and occurs also in Gilat’s subsoil.

4. Discussion

4.1. Composition of Lowland Soils—The Natural Processes

The natural processes that govern the original composition in each place are initially taken into consideration in order to study and comprehend potential changes in soil chemical and mineralogical composition produced by farming. The four fine-textured soils show a gradual compositional change from south to north that reflects both the increasing precipitation and the distance from nearby dust sources. Eventually, the regional geological and geomorphological conditions have an impact on each of them. Gilat’s soil appears to be enriched by sand-sized quartz based on its texture and SiO₂ concentration, which are both roughly 20% higher than the typical concentration of northern Negev Holocene loess captured in archaeological sites [44]. This is caused by the addition of fine sand, winded from the neighboring sand dunes [45], which “dilutes” the loess. The amounts of calcite and feldspars (combined K-feldspar and plagioclase) in Gilat decrease from~20% for each mineral to ~10% in Bet Dagan. The feldspars’ amount further decreases to ≤5% in Nir Etzion and almost vanishes in Sarid. Similarly, the quartz amount decreases from ~50% in Gilat to ~40% in Bet Dagan and Nir Etzion, and to ~30% in Sarid. The calcite amount, unpredictably, increases in the two latter sites. In Nir Etzion, it is probably due to a contribution of calcite from the limestone-built Mt. Carmel, where the colluvial soil is right at its foothill. It is questioned why the calcite at Sarid vertisol has increased further in comparison to nearby vertisols with 10% calcite [4,40]. The soft limestone outcrops above the orchard might be a local supply source.

In parallel to the general decrease in quartz, feldspars, and, in part, calcite, the amount of phyllosilicates increases from south to north. The amount of phyllosilicates increases from 15%–20% in Gilat to 40%–45% in Bet Dagan, 35%–45% in Nir Etzion, and reaches a maximum of ~50% in Sarid. This northward increase in the non-sandy soils is accompanied by changes in the clay fraction composition and, eventually, the phyllosilicates’ assemblage.

The palygorskite in Gilat is mostly pedogenic, as its amount is higher than in dust [26]. Pedogenic palygorskite was documented earlier in similar Negev soils [46]. The most drastic changes occur between Gilat and Bet Dagan and include the disappearance of palygorskite, a drop in the amounts of illite and kaolinite, and the replacement of illitic IS by smectitic IS, as is evident by the saddle values. The amount of smectitic IS rises to >75% in Nir Etzion and Sarid, though the saddle values there are somewhat higher than those of Bet Dagan. Thus, the types and abundances of phyllosilicates also reflect long-term geologic factors, including the dissolution or alteration of primary feldspar and mica and the subsequent transformation and dissolution/precipitation reactions that operate within the soil horizons, as has been stated in another study in the USA [18].

The chemical composition reflects the reorganization of the bulk and clay fractions’ mineralogical compositions. As has been noted earlier [32,47] the ratio of Fe₂O₃ to Al₂O₃ is mostly around 0.5–0.6 for the soil studied (Figure 4) and for all soils and surface sediments in Israel, indicating stability of both elements in the soil environment, which is also supported by trace concentrations in groundwater. The relations between K₂O and Al₂O₃ are complex (Figure 5) as neoformation of smectitic IS necessitates all Al₂O₃ and some K, whereas neoformation of kaolinite does not include K and the available K (as cation) may either adsorb to the 2:1 phyllosilicates or drain into groundwater. Thus, the range of K concentrations in Bet Dagan and Sarid is similar to that of Gilat, despite the significant increase in smectitic IS, reflecting the response of phyllosilicates to the local environmental condition. Potassium-rich minerals such as illite, illitic-IS, and K-feldspar dissolve and contribute a part of their K to the neoformed smectitic IS, confirming the importance of structural K release to pedogenic phyllosilicates [11,12,13].

Synthesizing all the trends of compositional changes described above suggests that the clayey nature of vertisols is not just a result of finning due to dust grain size decrease from south to north [48], or alluvial winnowing, but mainly due to the addition of neoformation phyllosilicates at the expense of quartz, feldspars, other primary minerals, and rain solutes. This process has been suggested for vertisols in Israel [29] and the USA [49].

Sandy soils like the Mishmeret soil react differently to the local environment due to their permeability. Soil K, derived from the decomposition of smectitic IS, supports the formation of pedogenic illite, as has been suggested for vegetation-rich soils [17,50]. This process has also been documented in several studies on well-leached red Mediterranean soils in Israel, e.g., [32,34]. The remaining SiO₂ and Al₂O₃ gradually form pedogenic kaolinite, which may turn into the main phyllosilicate. The occurrence of the minor but unique KaS mineral testifies to a transitional mineral through smectite-to-kaolinite weathering, as has been recorded in soils on limestone and other rocks under seasonal humid climates, e.g., [51,52]. Here, the high sand permeability compensates for the lower precipitation than in those humid occurrences and promotes KaS formation even under Mediterranean climate.

4.2. Effect of Irrigation

Irrigation can affect natural processes that are controlled by precipitation, such as the dissolution of relatively soluble minerals, or the transformation of phyllosilicates. The dissolution effect is evaluated here by comparing fallow and cultivated plots for calcite, as dolomite’s amounts are too small. Dissolution of bedrock and dust calcite by rain in Mediterranean soils is a natural process [53,54,55] and is anticipated to be enhanced by adding irrigation waters. Fertilization and manure application can enhance this process owing to added acidity (e.g., ammonium sulfate), enhancing soil solution acidification (via nitrification) leading to calcite dissolution [56], and adding complexing ions (e.g., phosphate) [57]. This is best recorded by the mineralogical and chemical data (Table 5 and Table 6) in the Gilat site where m.a.p. is about half of the other sites (Table 1). On the other end, the Mishmeret soil that developed under enhanced leaching conditions is completely calcite depleted, despite the fact that its parent material (dune sand and desert aeolian dust) contained substantial amounts of calcite [28,30]. The three vertisols are less well drained, especially the Sarid soil, and hence calcite dissolution is less pronounced.

Still, unpublished results of CaCO₃ measurements (estimated from the CO₂ volumes released using a calcimeter) from the Bet Dagan vertisol suggest that carbonates were lost from the topsoil during the about 30 years of experiments in five of the six experimental plots studied. The data from five fertilized soils for each sampling year were statistically weighed with respect to first measurement within each soil and subject to Anova analysis (Figure 7). This manipulation displays the decrease in calcite content in fertilized plots to be highly statistically significant (p < 0.01). The non-fertilized soil was excluded from this weighing manipulation because it showed almost no change in calcite content over the years.

Precipitation plays a major role in shaping the soil phyllosilicates’ assemblages, as discussed in the previous section. It is therefore expected to enhance the transformation and neoformation of soil phyllosilicates, though the incorporation of fertilization may mask the net effect. The saddle values in Bet Dagan (Table 6) show enhanced smectitization of source phyllosilicates in the cultivated plots. In addition, the clear decline in the cg indicator between fallow and control plots in both depths suggests that smectitization is aided by the addition of irrigation water. The comparison between the rain-irrigated olive grove and the banana plot in Nir Etzion indicates that the kaolinite increase in the latter in both depths is a result of the heavy irrigation. In addition, the decrease in the %IS indicator in the topsoil, despite the heavy K fertilization has to be also attributed to irrigation. In contrast, the heavy irrigation of the grass loan in the sandy Mishmeret soil appears to be interfering with the natural production of kaolinite and illite IS. Instead, K concentration significantly decreases IS formation at the expense of illite and kaolinite.

To our knowledge, these aspects have not been reported earlier in studies that explored the effects of freshwater irrigation, as mostly they examined parameters such as soil fertility, structure, and organic matter concentration, e.g., [58,59].

4.3. Effect of Fertilization

The current study focuses on how K fertilizer affects soil K balance as indicated by the phyllosilicate indicators. An indirect impact on the phyllosilicates may also occur when fertilizers other than K cause high plant growth and, consequently, enhanced K uptake, as in the Gilat site. This site is dominated by illitic IS phases, as expressed by the highest saddle values, and the subsoil is the main arena for pedogenic processes, as stated above. Basic KCl was applied to all fertilization levels, namely, the control plots, as well as two levels of animal manure application, either alone or together with N (Table 2a). The substantial increase in the %IS indicator that occurred in the subsoil, from the fallow to the control plots (Figure 6) indicates that the K added was much more than needed by the plants, leading to its storage as illitic IS phases. No further change in that indicator occurred under the higher fertilization level, whereas under the highest level, it slightly decreased. The highest N levels caused the highest plant growth and enhanced K uptake beyond its application Table 75 in [36],) causing the transformation of the original illitic IS phases to less illitic ones. Overall, a neutral K balance, as compared to the fallow values, is recorded by the %IS indicator in the topsoil, whereas a slightly positive one is recorded in the subsoil.

Dissimilar K balances between topsoil and subsoil are also recorded in the Bet Dagan site, which is dominated by smectitic IS phases, as expressed by the lowest saddle values. The major decline in the cg indicator from the fallow to the control plots in both depths is caused by soil K consumption, turning the smectitic IS into a more smectitic one (Figure 6). This is in accord with previous studies, which reported that cropping without K fertilization led to the formation of stable smectite in alluvium soil containing carbonates, e.g., [60]. Under a fertilization level of the same K but higher P levels, the slight gradual decline in the cg indicator also indicates higher plant growth, soil K-consumption, and a local negative K balance. However, it became, unexpectedly, even more negative in the topsoil under the addition of manure. This suggests enhanced downward migration of manure-K, as has been shown before, e.g., [61], and a build-up of a K reservoir in the subsoil. Overall, a negative K balance, as compared to the fallow values, is recorded in the topsoil and a neutral one is recorded in the subsoil. Dissimilarity between topsoil and subsoil may reflect, in addition to a reduction in permeability and downward migration, a difference in root activity with depth [62,63].

From the response of the Gilat and Bet Dagan experimental permanent plots to K fertilization over the course of more than thirty years, as expressed by the relative quantitative indicators, it can be deduced that the change in the soil K availability is rather minor. This is a consequence of two main factors: a. the high amount of K-bearing IS phases in the soils and b. the crop rotation system under the same fertilization regime that compensates between more and less K-consuming plants, and between plants of different root depths [36].

The three single-plant sites, which are heavily fertilized by K, display an intricate behavior with respect to K balance. K uptake and release in the Nir Etzion banana plantation are mainly reflected by the cg indicator (Figure 6); a neutral to negative K balance occurs in the topsoil, whereas the subsoil displays a neutral to positive K balance. Moreover, the higher average illite amount in both depths (Table 6) records pedogenic illite formation, apparently due to plant activity, as has been suggested before [50]. This dynamic K behavior in the banana plots, with the intensive application and consumption by the plants, is suggested to reflect seasonal changes along the growth cycle. Surplus K during K application time leads to illite formation, which seems to be rather stable [8]. Nevertheless, during maximum growth and fruit-bearing, the illitic IS is partially decomposed, resulting in an overall neutral to slightly negative K balance. Such seasonal variability in the composition of IS phases due to plant activities has been already recorded in previous studies [18,64,65]. Similarly, it is suggested that the highest %IS and K₂O/Al₂O₃ values in the wheat field topsoil, which clearly indicate K application in excess, is also a seasonal phenomenon, as it was sampled after the harvest and before the rainy season.

The Sarid grapefruit orchard is the only site with a distinctly higher K₂O concentration than the fallow soil, and with cg and %IS indicators (Figure 6) that apparently represent a negative to neutral K balance. This is suggested to reflect a seasonal imbalance due to a time gap between the K application when K temporarily occurs as a water-soluble and exchangeable ion and is later fixed by the IS phases. Despite the deep roots, the subsoil is more illitized than the topsoil, which is suggested to be associated with wastewater irrigation, which has been shown to enhance downward K migration [66,67]. It has also been shown in experiments with green-manure plants that even deep-rooting crops extract most of their K demand from the upper soil layers [60].

Though K leaching should be anticipated in permeable sandy soils, the heavy manure application in the Mishmeret site strengthens the natural process of pedogenic illitization of former IS phases. This is expressed mainly in the topsoil by significant illite formation, along with an increase in the cg and %IS values, much more than in the other sites. The small amount of the clay fraction enables an intense response to K addition and immediate new chemical equilibrium with respect to pedogenic illite, as has been modeled by [50]. K application to nearby sandy soil significantly increased corn yield even under high salinity levels, in contrast to fine-textured soils [68].

The mutual formation of pedogenic kaolinite and illite has not been modeled, or reported before, aside from in some cases in non-cultivated soils in Israel [32,33,34]. The unique positive correlation between the amounts of kaolinite and illite at Mishmeret (Figure 8), well demonstrates their mutual pedogenic formation.

The main control on the small or undetectable modifications of phyllosilicates in the clayey soils after so many years of K application is the dominance of 2:1 phyllosilicates, which locally react to temporal changes in fertilization and plant growth. The dynamic K intake or uptake leads to a complex and heterogeneous clay assemblage at every depth within a profile [8,18]. Two additional controls are the crop rotation in the permanent plot experiments, as stated above, and proper fertilization scheme in the single-plant plots. These factors also probably hold for earlier studies that found that XRD patterns remained unaltered even after extensive cultivation [23,69,70].

5. Conclusions

Lowland long-term cultivated soils in Israel, four fine-textured and one sandy, were studied for the impact of K-fertilization and irrigation on their bulk mineralogical and chemical composition and the clay fraction mineralogical composition. The main conclusions are:

• Natural changes in the composition of the clayey soils studied from the semi-arid to the Mediterranean climate regimes, from south to north, included reduced amounts of calcite, quartz, and feldspars and increased phyllosilicates. Mixed layer IS phases became more smectitic northward and palygorskite disappeared.
• Application of K caused diverse behavior of the IS phases in the fine-textured soils. Quantitative XRD analysis estimated either a negative or a positive K balance in certain soil depths and fertilization levels at the same site. However, under high fertilization levels in the permanent plots, which led to high plant mass production and high K consumption, illitization declined.
• Irrigation affected soils by dissolving calcite and transforming phyllosilicates, enhancing the smetitization of source IS phases at one site and kaolinite formation at another two sites.
• The rather balanced K in the two permanent plot experiments after more than 30 years is due to the crop rotation applied and the dominance of IS phases in the clay fraction. The latter parameter is also the main reason for the small changes in K balance in the single-plant plots. This emphasizes the importance of IS minerals as a dynamic K pool that responds to plant needs.
• An exceptional pedological system took place in the sandy soil with a small clay fraction, where kaolinite and illite amounts were higher than in the clayey soils. Fertilization enhanced the natural process of contemporaneous transformation of source IS phases into both kaolinite and illite, resulting in a positive K balance.