- Email: [email protected]
Soil tillage effects on monovalent cations (Na+ and K+) in vertisols soil solution
Catena 84 (2011) 61–69
Contents lists available at ScienceDirect
Catena j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a t e n a
Soil tillage effects on monovalent cations (Na+ and K+) in vertisols soil solution B. Lozano-García a,⁎, L. Parras-Alcántara a, J.L. Muriel-Fernández b a b
Dpto. de Química Agrícola y Edafología, Universidad de Córdoba, Campus Universitario de Rabanales, Edificio Marie Curie, 3a planta, 14071, Córdoba, España Instituto de Investigación y Formación Agroalimentaria y Pesquera. Área de Producción Ecológica y Recursos Naturales. “Las Torres-Tomejil”, 41200, Alcalá del Río, Sevilla, España
a r t i c l e
i n f o
Article history: Received 12 May 2010 Received in revised form 10 September 2010 Accepted 16 September 2010 Keywords: Soil solution Soil blocks Conventional tillage Direct drilling Minimum tillage Vertisol
a b s t r a c t Potassium is an essential macronutrient for plants; it is characterized by increased photosynthetic activity by ensuring a better utilization of light energy, also acts as a regulator of cell osmotic pressure, decreasing transpiration and helping to maintain cell turgidity. However, the sodium is not an essential element for plants, although it is beneficial to certain crops, in some instances can replace the potassium and osmotic regulation making and turgidity of the cells, this effect is greatest when the supply of potassium is deficient (Wild, 1992). Both elements, in periods of aridity, delayed the wilting of plants to maintain cellular osmotic potential and in cold periods, they lower the freezing point of sap (Navarro and Navarro, 2000). This is an experiment to study the influence of soil management techniques on the monovalent cations in soil solutions at different depths. The cropping systems studied are conventional tillage, minimum tillage and direct drilling. Conventional tillage releases more Na+ and K+ to the soil solution than the conservative techniques. In the case of Na+, the conventional tillage soil solution has an average concentration of 0.563 meq/L compared to 0.303 meq/L of minimum tillage and 0.340 meq/L of direct drilling. As for the K+, the soil solution concentration of conventional tillage is 0.097 meq/L, compared to 0.079 meq/L of the solution of minimum tillage and 0.056 meq/L of direct drilling. The behavior for the two cations studied is distinct at different depths. The Na+ is more abundant in water samples of soil taken in depth. Therefore, the salinization risk may take place in the subsoil, especially in conventional tillage where the Bw1 horizon values are three times higher than in the Ap horizon, while the K+ is more abundant in the surface horizon. Conventional tillage and minimum tillage techniques, in the Ap horizon have a similar pattern with a K+ concentration average of 0.15 meq/L and 0.14 meq/L, respectively, resulting in lower values for direct drilling. Studies on clay soils have not been performed previously because of the difficulty presented by these soils when soil solution extracted for analysis. We analyzed the monovalent cations (sodium and potassium) from soil solution; because the soil solution is the immediate source of sodium and potassium for plants. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The loss of soil productivity through the accumulation and leaching of salts is a global problem in arid, semiarid and subhumid regions (Szabolcs, 1989). The accumulation of salts in the profile is mainly controlled by the amount of salts that are released from the soil and the amount of salts leaving the soil by percolation (U.S. Salinity Laboratory Staff, 1954; Ayers and Westcot, 1987). Likewise, in the case of clay soils, water infiltration through cracks can cause lateral displacement of salts due to the influence of the horizontal component of infiltration. With respect to production, there are many authors that highlight the benefits of conservation agriculture versus a conventional one, as ⁎ Corresponding author. Tel.: +34 957211092; fax: +34 957212146. E-mail address: [email protected] (B. Lozano-García). 0341-8162/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.catena.2010.09.005
in the case of the Experimental Farm Tomejil (Carmona, Sevilla) in which this work has been carried out (Perea, 2004; Muriel et al. 2005). Studies such as Ordoñez et al. (2007) show an improvement in the physical and chemical properties of the soil under conservative techniques. Jiménez et al. (2005) reveals that the soils under direct drilling, contribute a greater amount of water to cultivation especially in the first 20 cm of the profile. The influence of the management technique on the monovalent cation contents in different soil types has been identified. The results for the case of sodium indicate a greater content when the cultivation techniques are traditional. Thomas et al. (2007) in an Australian luvisol, and Dalal (1989) and Loch and Coughlan (1984), in vertisols, have detected smaller amounts of exchangeable sodium when the cultivation techniques are conservative than when conventional tillage is used. For potassium, the results are not so unanimous. Thomas et al. (2007) claimed that the exchangeable potassium is greater in direct drilling than in conventional tillage in the first 10 cm of the soil surface.
62
B. Lozano-García et al. / Catena 84 (2011) 61–69
Table 1 Physical characterization of the soil. Horizon and depth (cm) Sand (%)a N50 μm Silt (%)a 50–2 μm Clay (%)a b2 μm Bulk density (g.cm−3) Macroporosity (%) Microporosity (%) pF 1/3 atmb pF 15 atmb Direct drilling profile (DD) Ap (0–24) 3.0 Bw1(24–64) 3.2 Bw2k (64–140) 3.5 Ck (N140) 1.3
18.3 21.9 19.1 20.4
78.7 74.9 77.4 78.3
1.27 1.30 1.32 1.36
61.4 56.4 55.7 –
38.6 43.6 44.3 –
59.0 57.4 58.3 58.7
47.2 45.0 46.3 46.7
Minimum tillage profile (MT) Ap (0–24) 3.4 4.1 Bw1 (24–64) Bw2k (64–140) 3.7 Ck (N140) 1.6
20.4 21.3 19.3 20.4
76.2 74.6 77.0 78.0
1.29 1.30 1.32 1.37
58.1 56.0 55.3 –
41.9 44.0 44.7 –
57.9 57.2 58.2 58.6
45.7 44.7 46.1 46.6
Conventional tillage profile (CT) Ap (0–24) 5.7 2.4 Bw1(24–64) Bw2k (64–140) 0.7 Ck (N140) 0.1
18.2 22.4 21.2 22.7
76.1 75.2 78.1 77.2
1.27 1.29 1.32 1.37
60.0 57.4 56.1 –
40.0 42.6 43.9 –
57.6 57.5 58.7 58.8
45.4 45.1 46.8 46.9
a b
Métodos Oficiales de Análisis (1994). Richard's membrane method (1947).
Table 2 Chemical characterization of the soil. pH (H2O)a
Horizon and depth (cm)
O.M. %a
Exchangeable macroelements (cmol (+)/kg)a
Assimilable macroelements (cmol (+)/kg)b
Na+
K+
Na+
K+
Direct drilling profile (DD) Ap (0–24) Bw1(24–64) Bw2k (64–140) Ck (N140)
7.5 7.3 7.6 7.7
1.54 1.21 0.83 0.15
1.46 1.24 1.96 3.64
1.31 1.07 0.42 0.39
8.46 1.59 2.39 5.70
1.69 1.46 0.88 0.74
Minimum tillage profile (MT) Ap (0–24) Bw1 (24-64) Bw2k (64–140) Ck (N140)
7.5 7.3 7.6 7.6
1.32 1.21 0.82 0.03
0.74 0.62 1.83 2.93
1.81 1.22 0.56 0.42
1.26 1.59 2.40 6.10
2.71 1.71 0.98 0.62
Conventional tillage profile (CT) Ap (0–24) Bw1 (24–64) Bw2k (64–140) Ck (N140)
7.3 7.3 7.6 7.7
1.69 1.38 0.80 –
0.98 0.90 1.10 2.89
1.10 1.15 1.00 0.49
1.23 1.51 1.89 2.21
3.27 1.58 0.92 0.87
a b
Métodos Oficiales de Análisis (1994). G.T.N.M.A. (1976).
In the same way, Martín-Rueda et al. (2007) found that the potassium content is higher in direct drilling than in the other systems in the first 15 cm of the soil surface. But Guzmán et al. (2006), DeMaría et al. (1999), Franzluebbers and Hons (1996) and Hunter and Cowie (1989) found more potassium contents in the systems of conventional tillage than in DD, although these results are only in the surface horizon. There are also studies which determine the nutrient content at different depths. In this sense, Franzluebbers and Hons (1996) found that the exchangeable sodium increased with the depth in the two
techniques that were studied (conventional tillage and direct drilling) in a Fluventic Ustochrept in Texas. With regard to potassium, Thomas et al. (2007) in an Australian luvisol, Guzmán et al. (2006) in a clay soil of Manhattan and Asghar et al. (1996) in a vertisol in Queensland showed that the concentration of potassium is higher in the surface horizon. Also, Lal et al. (1990) and Ismail et al. (1994) found that potassium increased in the surface horizon, even in the case when the management practices diminished. However, Orihuela et al. (2001) found no influence of depth on the amount of potassium when they
Table 3 Long-term climatic conditions for Tomejil, Sevilla (1979–2007). Climatic parameter
January
February
March
April
May
June
July
August
September
October
November
December
Annual
T (°C)a P (mm)b PET (mm)c
9.4 75.4 18.2
10.5 74.8 21.6
12.6 65.3 36.8
14.8 57.8 52.3
18.5 28.8 86.6
21.5 24.5 114.2
26.8 5.1 172.2
26.7 6.9 160.1
22.7 19.6 105.4
17.4 60.8 61.7
13.4 63.8 33.9
10.3 78.0 20.6
17.1d 560.8e 883.6f
a b c d e f
Mean monthly temperature (°C). Mean monthly precipitation (mm). Mean monthly potential evapotranspiration (mm) (Thornthwaite, 1948). Yearly mean temperature (°C). Total precipitation (mm). Total potential evapotranspiration (mm).
B. Lozano-García et al. / Catena 84 (2011) 61–69
63
Fig. 1. Unaltered soil blocks.
studied the leaching of the soluble form of potassium in clay loam texture soils in Huelva (Spain). The previous authors have studied the influence of management technique and sampling depth on monovalent cations, for this purpose, they analyzed the content of exchangeable sodium and potassium soil samples. In this study, we analyzed the monovalent cations (sodium and potassium) from soil solution; because the soil solution is the immediate source of sodium and potassium for plants. Sodium and potassium ions fixed on solid soil particles are easily exchanged with other cations and renew the solution. Therefore, the exchangeable sodium and potassium are index representatives of the amount of cations that have crops in the soil solution. Such studies on clay soils have not been performed previously because of the difficulty presented by these soils when soil solution extracted for analysis. The objective of this study was to determine firstly if there were variations in the content of Na+ and K+ in the soil solution in relation to the cultivation technique and sampling depth. Secondly, we wanted to make a long-term prediction for the behavior of these nutrients in the soil solution.
2. Materials and methods 2.1. Site description This work has been focused on the experimental farm Tomejil, located in Carmona, Seville, Spain (37°24′N, 5°35′W, 79 m.a.s.l.), belonging to IFAPA (Andalusia Institute of Agrarian Research and Training, Fisheries, Food Production and Environment). In this farm, since 1982 different soil management systems have been applied in a normal wheat–sunflower–legume rotation. At present, the soil treatments that are being applied are conventional tillage (CT), minimum tillage (MT) and direct drilling (DD) (Perea et al., 2006). The typical soils in this area are Calcic Vertisol (FAO, 1998), which are locally known as “black lands”. The characterization of the soil used in this study is given in Tables 1 and 2 (Lozano, 2009). From the climatological point of view, the study area is characterized by very dry summers and an irregular distribution of rainfall throughout the year (Table 3). This, together with high temperatures during the summer, leads to an adverse agricultural activity. Rainfed crops are typical in Andalusian soils.
0.25 m
Cup in Ap horizon
0.60 m
Cup in Bw1 horizon
0.40 m Fig. 2. Shape and dimensions of the unaltered soil blocks with cups.
64
B. Lozano-García et al. / Catena 84 (2011) 61–69
2.2. Soil block extraction
a) Ap horizon 1,2
MT CT
0,8
meq/L
This study could not be made in the field due to the uneven distribution of rainfall. Therefore, for this study, three soil blocks (0.40 m × 0.25 m × 0.65 m in deep) were extracted in the three plots on the farm. Each plot had different tillage (CT, MT, and DD). During the short transport to the laboratory, the soil blocks were supported by a wooden frame. The soil blocks were placed on pallets and the wooden frame was removed, following a method similar to the one described by Lozano (2009) (Fig. 1).
DD
1
0,6 0,4 0,2 0 1
2
3
4
5
6
7
8
9
2.3. Soil water status, irrigation and solute sampling
10 11 12 13
b) Bw1 horizon 1,2
During the experiment, the volumetric soil water content was monitored using a TDR (Time Domain Reflectrometry), the model used in this study was ML2x with a dataloger HH2 moisture meter (Delta-T Devices). The three columns were kept under conditions of permanent saturation. Irrigation was applied uniformly to the soil blocks as a fine mist at a rate of 20 mm for only 1 h daily. This irrigation rate although intense was not realistic for the study area, but we needed saturated conditions. Two suction Teflon quartz cups (soil moisture sampler teflon/ PTFE/quartz, Eijkelkamp) were installed in each soil block, one in the Ap horizon and another in the Bw1 horizon, located at depths of 0.12 m and 0.55 m, respectively (Fig. 2). A suction of 70 kPa was applied to each cup. Pressure in the cups was allowed to diminish through time until the samples were collected. The time needed to
DD MT
1
CT
meq/L
0,8 0,6 0,4 0,2 0 1
2
3
4
5
6
8
7
9
10 11 12 13
Fig. 3. Temporal evolution of the sodium concentration in the soil.
Bw1 horizon
DD
Ap horizon 1
1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
0
0
MT
0
3
6
9
12
15
1
1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
0
3
6
9
12
15
0
3
6
9
12
15
0
3
6
9
12
15
0
0 0
3
6
9
12
15 1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
CT
1
0
0 0
3
6
9
12
15
Fig. 4. The analysis of smoothing, using moving averages to five terms for the case of Na+ in the horizons studied (Ap and Bw1) in the different tillage systems.
B. Lozano-García et al. / Catena 84 (2011) 61–69
integrated and moving average parts of the model, respectively (Box and Jenkins, 1976). The Random walk model is the one which offers a better prognosis for future values and is given by the value of the latest available data (Hughes, 1998). The linear trend model assumes that the best forecast for future values is the linear regression line obtained for the known data series (Chai, 1995). Thirdly, the analysis of variance was performed by standard statistical techniques using tillage practices (CT, MT and DD) and depth extraction of the soil solution (Ap and Bw1) (Kruskal and Wallis, 1952). The treatment mean values were compared using the least significant difference (LSD) at P ≤ 0.05.
collect enough sample for laboratory analysis was 12 days, during which time every 24 hours was applied suction of 70 kPa. From each cup, 13 samples of soil solution were extracted where the sodium and the potassium content was determined (Métodos Oficiales de Análisis, 1994). 2.4. Statistical analysis The results were statistically analyzed using Statgraphics Centurion XV (StatPoint, Inc., 2005). Sodium and potassium were considered as variables. The time series of each element was represented according to the type of tillage (CT, MT and DD) and depth. Secondly, the analysis of the time series was performed to find if there were any trends in the data and the prediction of future values. To determine the trend, a smoothing five-term moving averages was applied (Fuller, 1996; Peña, 1998, 2005). For the prediction of future values, a model was adjusted to each time series to give an outlook on future behavior. In each case, we chose the option that best fitted the series (ARIMA model, Random walk model or Linear trend model), using the Akaike criterion (1974) to assign the model. The ARIMA (p,d,q) model is an autoregressive integrated moving average, where p, d and q are integers greater than or equal to zero and less or equal to two referred to the order of the autoregressive,
3. Results and discussion 3.1. Sodium behavior 3.1.1. Sodium concentration in the soil solution In the three techniques, the sodium concentration in the water in the Ap horizon is less than in the Bw1 horizon. However, the sodium concentration remains constant in the distinct soil horizons in DD and MT and increases in the Bw1 horizon when the CT technique is used (Fig. 3a and b).
Bw1 horizon
DD
Ap horizon 1,5
1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1 -0,5
-0,5 0
5
10
15
20
25
0
5
Linear Trend
MT
65
10
15
20
25
20
25
20
25
Linear Trend
1,5
1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1
-0,5
-0,5 0
5
10
15
20
0
25
5
ARIMA (0,2,2)
10
15
ARIMA (0,2,2) 1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1
CT
1,5
-0,5
-0,5 0
5
10
15
ARIMA (1,1,2)
20
25
0
5
10
15
Linear Trend
Fig. 5. Tendency prediction from time series analysis for the case of Na+ in the horizons studied (Ap and Bw1) in the different tillage systems.
66
B. Lozano-García et al. / Catena 84 (2011) 61–69
In the Ap horizon, the plot which has a higher sodium concentration is CT with 0.21 meq/L and the one which has a lower sodium concentration is the DD plot with 0.12 meq/L (Fig. 3). Regarding the soil solution from the Bw1 horizon (Fig. 3), the CT is the one which has more sodium, and also increases its concentration in time, from 0.71 meq/L to 0.97 meq/L. MT has a mean concentration of 0.46 meq/L and DD 0.56 meq/L, respectively.
3.1.3. ANOVA The concentration of sodium in the soil solution depends on the tilling technique and sampling depth by the Kruskal–Wallis test at P ≤ 0.05. The soil solution of CT is richer than those of DD and MT, with mean values of 0.563 meq/L, 0.340 and 0.303 meq/L, respectively (Fig. 6). Also, Loch and Coughlan (1984), Dalal (1989) and Thomas et al. (2007) showed that the evolution of the sodium concentration was similar in vertisols and luvisol under conventional tillage and under conservation techniques. In Fig. 6, the existence of two homogeneous groups is explained more clearly in terms of behavior according to the tilling technique. When comparing the soil solution of the two horizons, sodium is more abundant in the Bw1 horizon, with average values of 0.630 meq/L, compared to 0.174 meq/L of the average of the solutions embodied in the Ap horizon, as shown in Fig. 7. Franzluebbers and Hons (1996) also found that exchangeable sodium increased with depth in the two techniques studied (conventional tillage and direct drilling) in a Fluventic Ustochrept in Texas. Our results suggest that CT produces a greater risk of salinization of the Bw1 horizon than conservation techniques since it releases more sodium to the soil solution throughout the profile and causes an increased accumulation in the Bw1 horizon. Furthermore, DD retains
meq/L
Fig. 7. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the sodium variable.
some degrees of moisture in the soil surface, reducing the mobility of sodium, because it reduces abrupt changes in the wetting drying of the soil.
3.2. Potassium behavior 3.2.1. Potassium concentration in the soil solution With regard to potassium, using the three tillages, the soil solution concentration is greater in the Ap horizon than in the Bw1 horizon, increasing its concentration in time. Therefore, initially, the values are 0.05 meq/L for DD, 0.11 meq/L for MT and 0.09 meq/L at CT, ending the experiment with values of 0.10 meq/L, 0.16 meq/L and 0.18 meq/L, respectively (Fig. 8a). Fig. 8b shows how MT and CT have the same behavior. DD is the one which supplies the least amount of potassium to plant roots, with an average value of 0.09 meq/L. In the case of the Bw1 horizon, the behavior in all three cases is the same. Constant contents in the initial
a) Ap horizon 0,25 DD
0,2
MT CT
meq/L
3.1.2. Time series analysis After performing smoothing using moving averages to five terms to highlight the data trends it was shown that in the case of sodium in the conservation techniques, the tendency was to maintain a constant concentration in soil water (Fig. 4). In fact, both DD and MT with values ranging in the Ap horizon between 0.1 and 0.2 meq/L and in the Bw1 horizon about 0.55 meq/L were observed. Furthermore, in CT the sodium concentration in the soil solution of the Ap horizon decreased throughout the experiment, while the Bw1 horizon concentration increased. Although the forecasts were made from the model that best fitted the data set, it was found that the prediction for sodium was valid only for two situations, MT and CT. In the case of MT the prediction was for the Ap horizon, which conformed to an ARIMA (0,2,2) model and in the case of CT the prediction was for the Bw1 horizon, which adjusted to a linear trend model. In the other cases, the predictions were unreliable (Fig. 5).
Na+
0,15 0,1 0,05 0 1
2
3
4
5
6
7
8
9
10 11 12 13
b) Bw1 horizon 0,25 0,2
meq/L
Na+ meq/L
0,15
DD
0,1
CT
MT
0,05 CT
MT
DD
0 1
Fig. 6. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the sodium variable.
2
3
4
5
6
7
8
9
10 11 12 13
Fig. 8. Temporal evolution of the potassium concentration in the soil.
B. Lozano-García et al. / Catena 84 (2011) 61–69
extractions are followed by an increase in the third or fourth extraction, so this value remains the same until the end. The MT solution is the poorest in potassium, with an average concentration of 0.02 meq/L, CT being the richest with a mean concentration of 0.04 meq/L. 3.2.2. Time series analysis In the case of potassium (Fig. 9), the smoothing technique shows how the Ap horizons of the three techniques increase the potassium concentration, while in the Bw1 horizon the tendency is to keep the concentration constant. The prediction is valid for DD, MT and CT. In the case of DD, the prediction is for the Bw1 horizon, which conforms to a random walk model. For the MT the prediction is for the Bw1 horizon, which conforms to an ARIMA (2,1,2) model, and in the case of CT the prediction is for the Ap and Bw1 horizons, which adjust to a linear trend model in both cases. In the other cases, the predictions are unreliable (Fig. 10). 3.2.3. ANOVA The concentration of potassium in the soil solution depends on the tilling technique and sampling depth by Kruskal–Wallis test at P ≤ 0.05. Firstly, it can be considered that the plots of DD and MT behaved similarly because the averages were homogeneous. On the other hand, MT and CT also behaved similarly. Fig. 11 clearly shows the values of potassium in MT, which overlap at one end with the DD and the other with CT.
67
By contrast, the results of Thomas et al. (2007) and Martín-Rueda et al. (2007) refer to a higher content of exchangeable potassium in the absence of tillage. In the case of comparing only CT and DD, a higher potassium content is seen in traditional cultivation. This is the case of the results of Hunter and Cowie (1989), Franzluebbers and Hons (1996), DeMaría et al. (1999) and Guzmán et al. (2006). However, in this study, the potassium concentration was greater in solutions extracted from the Ap horizons, which reached average values of 0.125 meq/L, whereas the average in Bw1 horizons was 0.030 meq/L (Fig. 12). In this case, the results coincide with those of Thomas et al. (2007) in Australian luvisol, Guzmán et al. (2006) in clay soil of Manhattan and Asghar et al. (1996) in a vertisol in Queensland. Also, Lal et al. (1990) and Ismail et al. (1994) found that potassium increased in the surface horizon, although in this case the tillage practices were reduced. However Orihuela et al. (2001) found no influence in the amount of potassium in the subsurface horizon when they studied the leaching of the soluble form of potassium in clay loam texture soils in Huelva (Spain). A higher concentration of exchangeable K+ in the surface layers of non-tillage treatments was apparently due to the return of crop residue on the soil surface. This is extensively documented for different types of soils (Lal et al., 1990; Franzluebbers and Hons, 1996; Martin-Rueda et al., 2007; Thomas et al., 2007) and also occurs in our study (Table 2). Although our data suggest that the extracted soil solutions between 0 and 0.24 m (Ap horizon) in conservation techniques are richer in
Fig. 9. The analysis of smoothing, using moving averages to five terms for the case of K+ in the horizons studied (Ap and Bw1) in the different tillage systems.
68
B. Lozano-García et al. / Catena 84 (2011) 61–69
DD
Ap horizon
Bw1horizon
0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05
0
0 0
5
10
15
20
25
0
5
MT
ARIMA (0,1,2) 0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05
0
15
20
25
20
25
20
25
0 0
5
10
15
20
25
0
5
ARIMA (0,1,2)
CT
10
Random Walk
10
15
ARIMA (2,1,2)
0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05 0
0 0
5
10
15
20
25
0
5
Linear Trend
10
15
Linear Trend
Fig. 10. Tendency prediction from time series analysis for the case of K+ in the horizons studied (Ap and Bw1) in the different tillage systems.
potassium. Our study shows how CT behaves in that sense, just as MT with a higher intake of potassium in the soil solution than DD. 4. Conclusions The use of undisturbed soil blocks to study the soil solution in the laboratory is a particularly advantageous method when weather conditions are adverse site.
The soil in this study was significantly affected by 25 years of different systems of soil management. CT releases more Na+ and K+ to the soil solution in Ap and Bw1 horizons than the conservative techniques. In depth, Na+ is more abundant in soil solution than in the surface horizon. In CT, there is a risk of salinization in the Bw1 horizon because the concentrations of Na+ in the soil solution are tripled in respect to the surface horizon. In the Bw1 horizon, K+ remains constant in all techniques studied. However, K+ increases in the Ap
0.1 2
K
+
K+
meq/L 0.1
meq/L
0
CT
MT
DD
Fig. 11. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the potassium variable.
Fig. 12. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the potassium variable.
B. Lozano-García et al. / Catena 84 (2011) 61–69
horizon and DD is the management that releases less K+ in the soil solution. The applications of statistical models to predict and evaluate the behavior of these cations in the future provide more reliable predictions for sodium than for potassium. We can predict for potassium only for CT while for sodium we can make predictions for the three techniques. According to the data obtained, CT is not advisable because of the large amount of Na+ released into the Bw 1 horizon. In the Conservationist techniques, DD releases less K+ in the soil solution in the Ap horizon. We recommend MT for the best management practice for the edafic–climatic condition studied because there is no risk of salinization and releases more K+ than DD. Acknowledgements The authors are grateful to the Ministry of Education and Science of Andalusia and the University of Cordoba for the formation scholarship granted for the development of the doctoral thesis of Beatriz Lozano Garcia. We would also like to thank the I.F.A.P.A. “Las Torres - Tomejil” of the Agriculture and Fisheries of the Andalusian for the facilities offered. References Akaike, H., 1974. A new look at statistical model identification. Transactions on Automatic Control AC-19, 716–723. Asghar, M., Lack, D.W., Cowie, B.A., Parker, J.C., 1996. Effects of surface soil mixing after long-term zero tillage on soil nutrient concentration and wheat production. In: Asghar, M. (Ed.), Proceedings of the 8th Australian Agronomy Conference, The Australian Society of Agronomy Inc. Toowoomba, Queensland, pp. 88–91. Ayers, R., Westcot, D., 1987. La calidad del agua en la agricultura, vol. 29. Estudio FAO Riego y Drenaje, Roma. Box, G.E.P., Jenkins, G.M., 1976. Time Series Analysis, Forecasting and Control. HoldenDay, San Francisco. Chai, Feng-Shun, 1995. Construction and optimality of nearly linear trend-free designs. Journal of Statistical Planning and Inference 48, 113–129. Dalal, R.C., 1989. Long-term effects of no-tillage, crop residue, and nitrogen application on properties of a Vertisol. Soil Science Society American Journal 53, 1511–1515. DeMaria, I.C., Nnabude, P.C., de Castro, O.M., 1999. Long-term tillage and crop rotation effects on soil chemical properties of a Rhodic Ferralsol in southern Brazil. Soil and Tillage Research 51, 71–79. FAO, ISRIC, ISSS, 1998. World Reference Base of Soil Resources. Soil Map of the world 1:5.000.000. Reports 60. FAO, Rome, Italy. Franzluebbers, A.J., Hons, F.M., 1996. Soil-profiles distribution of primary and secondary plant-available nutrients under conventional and no tillage. Soil and Tillage Research 39, 229–239. Fuller, W.A., 1996. Introduction to Statistical Time Series, 2nd edition. Wiley, New York. G.T.N.M.A. (Grupo de trabajo de normalización de métodos analíticos), 1976. Determinaciones analíticas en suelos II. Anales de Edafología y Agrobiología XXXV 7 (y 8), 813–834. Guzmán, J.G., Godsey, C.B., Pierzynski, G.M., Whitney, D.A., Lamond, R.E., 2006. Effects of tillage and nitrogen management on soil chemical and physical properties after 23 years of continuous sorghum. Soil and Tillage Research 91, 199–206.
69
Hughes, B.D., 1998. Random walks and random environments. Bulletin of the American Mathematical Society 35 (n° 4), 347–349. Hunter, H.M., Cowie, B.A., 1989. Tillage and stubble effects on nutrient distribution. In: Noble, R.M., Radford, B.J., Ruthenberg, S.J., Thorburn, P.J. (Eds.), Biloela Research Station Programme Review. Queensland Department of Primary Industries, Queensland, pp. 61–62. Ismail, L., Blevins, R.L., Frye, W.W., 1994. Long-term no-tillage effects on soil properties and continuous corn yields. Soil Science Society American Journal 58, 193–198. Jiménez, J.A., García, I., Vanderlinden, K., Perea, F., Muriel, J.L., 2005. Balance de agua en suelos arcillosos bajo laboreo convencional y siembra directa. Actas del Congreso Internacional sobre Agricultura de Conservación, Córdoba. 397–402. Kruskal, W.H., Wallis, W.A., 1952. Use of ranks in one-criterion variance analysis. Journal of the American Statistical Association 47 (260), 583–621. Lal, R., Logan, T.J., Fausey, N.R., 1990. Long-term tillage effects on a Mollic Ochraqualf in North-West Ohio. III. Soil nutrient profile. Soil and Tillage Research 15, 371–382. Loch, R.J., Coughlan, K.J., 1984. Effects of zero tillage and stubble retention on some properties of a cracking clay. Australian Journal of Soil Research 22, 91–98. Lozano, B., 2009. Estudio del comportamiento hídrico en ZNS (Zona No Saturada) de un Vertisol del Valle Medio del Guadalquivir. Tesis Doctoral de la Universidad de Córdoba. Servicio de Publicaciones de la Universidad de Córdoba. Córdoba. Martín-Rueda, I., Muñoz-Guerra, L.M., Yunta, F., Esteban, E., Tenorio, J.L., Lucena, J.J., 2007. Tillage and crop rotation effects on barley yield and soil nutrients on a Calciortidic Haploxeralf. Soil and Tillage Research 92, 1–9. Métodos oficiales de análisis, tomo III., 1994. Ministerio de Agricultura, Pesca y Alimentación, Dirección General de Política Agraria. Edita Secretaría General Técnica. Madrid. Muriel, J.L., Vanderlinden, K., Perea, F., García, I., Jiménez, J.A., 2005. Dinámica del agua en suelos sometidos a distintos sistemas de laboreo. In: López-Geta, J.A., Rubio, J.C., Martín, M. (Eds.), VI Simposio del agua en Andalucía. IGME. Sevilla, pp. 727–737. Navarro, S., Navarro, G., 2000. Química Agrícola. El suelo y los elementos químicos esenciales para la vida vegetal. Ed. Munid-Prensa, Madrid. Ordoñez, R., González, P., Giráldez, J.V., Perea, F., 2007. Soil properties and crop yields after 21 years of direct drilling trials in southern Spain. Soil and Tillage Research 94, 47–54. Orihuela, D.L., Hernández-Domínguez, J.C., Romero, E., González, A., 2001. Lixiviación de formas solubles de Ca++, K+ y PO= 4 y su relación con la c.e. y el pH en condiciones experimentales (Huelva. España). In: López Rodriguez, J.J., Quemada Sáez-Badillos, M. (Eds.), En Temas de Investigación en la Zona no Saturada. Pamplona. Peña, D., 1998. Estadística. Modelos y Métodos II. Modelos Lineales y Series Temporales. Ed. Alianza Universidad Textos. Madrid. Sexta edición. Peña, D., 2005. Análisis de Series Temporales. Ed. Alianza Editorial. Madrid. Perea, F., 2004. Estudio de la influencia del manejo del suelo en el régimen térmico de los vertisuelos de la Vega de Carmona. En Carel, Carmona Revista de Estudios Locales, vol. II. 2, pp. 32–45. Sevilla. Perea, F., Jiménez, J.A., García, I., Vanderlinden, K., Muriel, J.L., 2006. Caracterización hidroclimática en vertisuelos de la Campiña de Carmona. En Carel, Carmona Revista de Estudios Locales IV. N° 4, 11-18.Sevilla. Richards, L.A., 1947. Pressure-membrane apparatus construction and use. Agricultural Engineering 28, 451–454. StatPoint, Inc., 2005. Statgraphics Centurion XV User Manual. Printed in USA. Szabolcs, I., 1989. Amelioration of soil in salt affected areas. Soil Technology 2, 331–344. Thomas, G.A., Dalal, R.C., Standley, J., 2007. No-till effects on organic matter, pH, cation exchange capacity and nutrient distribution in a Luvisol in the semi-arid subtropics. Soil and Tillage Research 94, 295–304. Thornthwaite, C.W., 1948. An approach toward a rational classification of climate. Geographical Review 38 (n° 1), 55–94. U.S. Salinity Laboratory Staff, 1954. Diagnosis and improvement of saline and alkali soils. US Dept. Agr. Hanbook, vol. 60. Washintong DC. USA. Wild, A., 1992. Condiciones del suelo y desarrollo de las plantas según Russell. Ed. Mundi-Prensa. Madrid.
Contents lists available at ScienceDirect
Catena j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a t e n a
Soil tillage effects on monovalent cations (Na+ and K+) in vertisols soil solution B. Lozano-García a,⁎, L. Parras-Alcántara a, J.L. Muriel-Fernández b a b
Dpto. de Química Agrícola y Edafología, Universidad de Córdoba, Campus Universitario de Rabanales, Edificio Marie Curie, 3a planta, 14071, Córdoba, España Instituto de Investigación y Formación Agroalimentaria y Pesquera. Área de Producción Ecológica y Recursos Naturales. “Las Torres-Tomejil”, 41200, Alcalá del Río, Sevilla, España
a r t i c l e
i n f o
Article history: Received 12 May 2010 Received in revised form 10 September 2010 Accepted 16 September 2010 Keywords: Soil solution Soil blocks Conventional tillage Direct drilling Minimum tillage Vertisol
a b s t r a c t Potassium is an essential macronutrient for plants; it is characterized by increased photosynthetic activity by ensuring a better utilization of light energy, also acts as a regulator of cell osmotic pressure, decreasing transpiration and helping to maintain cell turgidity. However, the sodium is not an essential element for plants, although it is beneficial to certain crops, in some instances can replace the potassium and osmotic regulation making and turgidity of the cells, this effect is greatest when the supply of potassium is deficient (Wild, 1992). Both elements, in periods of aridity, delayed the wilting of plants to maintain cellular osmotic potential and in cold periods, they lower the freezing point of sap (Navarro and Navarro, 2000). This is an experiment to study the influence of soil management techniques on the monovalent cations in soil solutions at different depths. The cropping systems studied are conventional tillage, minimum tillage and direct drilling. Conventional tillage releases more Na+ and K+ to the soil solution than the conservative techniques. In the case of Na+, the conventional tillage soil solution has an average concentration of 0.563 meq/L compared to 0.303 meq/L of minimum tillage and 0.340 meq/L of direct drilling. As for the K+, the soil solution concentration of conventional tillage is 0.097 meq/L, compared to 0.079 meq/L of the solution of minimum tillage and 0.056 meq/L of direct drilling. The behavior for the two cations studied is distinct at different depths. The Na+ is more abundant in water samples of soil taken in depth. Therefore, the salinization risk may take place in the subsoil, especially in conventional tillage where the Bw1 horizon values are three times higher than in the Ap horizon, while the K+ is more abundant in the surface horizon. Conventional tillage and minimum tillage techniques, in the Ap horizon have a similar pattern with a K+ concentration average of 0.15 meq/L and 0.14 meq/L, respectively, resulting in lower values for direct drilling. Studies on clay soils have not been performed previously because of the difficulty presented by these soils when soil solution extracted for analysis. We analyzed the monovalent cations (sodium and potassium) from soil solution; because the soil solution is the immediate source of sodium and potassium for plants. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The loss of soil productivity through the accumulation and leaching of salts is a global problem in arid, semiarid and subhumid regions (Szabolcs, 1989). The accumulation of salts in the profile is mainly controlled by the amount of salts that are released from the soil and the amount of salts leaving the soil by percolation (U.S. Salinity Laboratory Staff, 1954; Ayers and Westcot, 1987). Likewise, in the case of clay soils, water infiltration through cracks can cause lateral displacement of salts due to the influence of the horizontal component of infiltration. With respect to production, there are many authors that highlight the benefits of conservation agriculture versus a conventional one, as ⁎ Corresponding author. Tel.: +34 957211092; fax: +34 957212146. E-mail address: [email protected] (B. Lozano-García). 0341-8162/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.catena.2010.09.005
in the case of the Experimental Farm Tomejil (Carmona, Sevilla) in which this work has been carried out (Perea, 2004; Muriel et al. 2005). Studies such as Ordoñez et al. (2007) show an improvement in the physical and chemical properties of the soil under conservative techniques. Jiménez et al. (2005) reveals that the soils under direct drilling, contribute a greater amount of water to cultivation especially in the first 20 cm of the profile. The influence of the management technique on the monovalent cation contents in different soil types has been identified. The results for the case of sodium indicate a greater content when the cultivation techniques are traditional. Thomas et al. (2007) in an Australian luvisol, and Dalal (1989) and Loch and Coughlan (1984), in vertisols, have detected smaller amounts of exchangeable sodium when the cultivation techniques are conservative than when conventional tillage is used. For potassium, the results are not so unanimous. Thomas et al. (2007) claimed that the exchangeable potassium is greater in direct drilling than in conventional tillage in the first 10 cm of the soil surface.
62
B. Lozano-García et al. / Catena 84 (2011) 61–69
Table 1 Physical characterization of the soil. Horizon and depth (cm) Sand (%)a N50 μm Silt (%)a 50–2 μm Clay (%)a b2 μm Bulk density (g.cm−3) Macroporosity (%) Microporosity (%) pF 1/3 atmb pF 15 atmb Direct drilling profile (DD) Ap (0–24) 3.0 Bw1(24–64) 3.2 Bw2k (64–140) 3.5 Ck (N140) 1.3
18.3 21.9 19.1 20.4
78.7 74.9 77.4 78.3
1.27 1.30 1.32 1.36
61.4 56.4 55.7 –
38.6 43.6 44.3 –
59.0 57.4 58.3 58.7
47.2 45.0 46.3 46.7
Minimum tillage profile (MT) Ap (0–24) 3.4 4.1 Bw1 (24–64) Bw2k (64–140) 3.7 Ck (N140) 1.6
20.4 21.3 19.3 20.4
76.2 74.6 77.0 78.0
1.29 1.30 1.32 1.37
58.1 56.0 55.3 –
41.9 44.0 44.7 –
57.9 57.2 58.2 58.6
45.7 44.7 46.1 46.6
Conventional tillage profile (CT) Ap (0–24) 5.7 2.4 Bw1(24–64) Bw2k (64–140) 0.7 Ck (N140) 0.1
18.2 22.4 21.2 22.7
76.1 75.2 78.1 77.2
1.27 1.29 1.32 1.37
60.0 57.4 56.1 –
40.0 42.6 43.9 –
57.6 57.5 58.7 58.8
45.4 45.1 46.8 46.9
a b
Métodos Oficiales de Análisis (1994). Richard's membrane method (1947).
Table 2 Chemical characterization of the soil. pH (H2O)a
Horizon and depth (cm)
O.M. %a
Exchangeable macroelements (cmol (+)/kg)a
Assimilable macroelements (cmol (+)/kg)b
Na+
K+
Na+
K+
Direct drilling profile (DD) Ap (0–24) Bw1(24–64) Bw2k (64–140) Ck (N140)
7.5 7.3 7.6 7.7
1.54 1.21 0.83 0.15
1.46 1.24 1.96 3.64
1.31 1.07 0.42 0.39
8.46 1.59 2.39 5.70
1.69 1.46 0.88 0.74
Minimum tillage profile (MT) Ap (0–24) Bw1 (24-64) Bw2k (64–140) Ck (N140)
7.5 7.3 7.6 7.6
1.32 1.21 0.82 0.03
0.74 0.62 1.83 2.93
1.81 1.22 0.56 0.42
1.26 1.59 2.40 6.10
2.71 1.71 0.98 0.62
Conventional tillage profile (CT) Ap (0–24) Bw1 (24–64) Bw2k (64–140) Ck (N140)
7.3 7.3 7.6 7.7
1.69 1.38 0.80 –
0.98 0.90 1.10 2.89
1.10 1.15 1.00 0.49
1.23 1.51 1.89 2.21
3.27 1.58 0.92 0.87
a b
Métodos Oficiales de Análisis (1994). G.T.N.M.A. (1976).
In the same way, Martín-Rueda et al. (2007) found that the potassium content is higher in direct drilling than in the other systems in the first 15 cm of the soil surface. But Guzmán et al. (2006), DeMaría et al. (1999), Franzluebbers and Hons (1996) and Hunter and Cowie (1989) found more potassium contents in the systems of conventional tillage than in DD, although these results are only in the surface horizon. There are also studies which determine the nutrient content at different depths. In this sense, Franzluebbers and Hons (1996) found that the exchangeable sodium increased with the depth in the two
techniques that were studied (conventional tillage and direct drilling) in a Fluventic Ustochrept in Texas. With regard to potassium, Thomas et al. (2007) in an Australian luvisol, Guzmán et al. (2006) in a clay soil of Manhattan and Asghar et al. (1996) in a vertisol in Queensland showed that the concentration of potassium is higher in the surface horizon. Also, Lal et al. (1990) and Ismail et al. (1994) found that potassium increased in the surface horizon, even in the case when the management practices diminished. However, Orihuela et al. (2001) found no influence of depth on the amount of potassium when they
Table 3 Long-term climatic conditions for Tomejil, Sevilla (1979–2007). Climatic parameter
January
February
March
April
May
June
July
August
September
October
November
December
Annual
T (°C)a P (mm)b PET (mm)c
9.4 75.4 18.2
10.5 74.8 21.6
12.6 65.3 36.8
14.8 57.8 52.3
18.5 28.8 86.6
21.5 24.5 114.2
26.8 5.1 172.2
26.7 6.9 160.1
22.7 19.6 105.4
17.4 60.8 61.7
13.4 63.8 33.9
10.3 78.0 20.6
17.1d 560.8e 883.6f
a b c d e f
Mean monthly temperature (°C). Mean monthly precipitation (mm). Mean monthly potential evapotranspiration (mm) (Thornthwaite, 1948). Yearly mean temperature (°C). Total precipitation (mm). Total potential evapotranspiration (mm).
B. Lozano-García et al. / Catena 84 (2011) 61–69
63
Fig. 1. Unaltered soil blocks.
studied the leaching of the soluble form of potassium in clay loam texture soils in Huelva (Spain). The previous authors have studied the influence of management technique and sampling depth on monovalent cations, for this purpose, they analyzed the content of exchangeable sodium and potassium soil samples. In this study, we analyzed the monovalent cations (sodium and potassium) from soil solution; because the soil solution is the immediate source of sodium and potassium for plants. Sodium and potassium ions fixed on solid soil particles are easily exchanged with other cations and renew the solution. Therefore, the exchangeable sodium and potassium are index representatives of the amount of cations that have crops in the soil solution. Such studies on clay soils have not been performed previously because of the difficulty presented by these soils when soil solution extracted for analysis. The objective of this study was to determine firstly if there were variations in the content of Na+ and K+ in the soil solution in relation to the cultivation technique and sampling depth. Secondly, we wanted to make a long-term prediction for the behavior of these nutrients in the soil solution.
2. Materials and methods 2.1. Site description This work has been focused on the experimental farm Tomejil, located in Carmona, Seville, Spain (37°24′N, 5°35′W, 79 m.a.s.l.), belonging to IFAPA (Andalusia Institute of Agrarian Research and Training, Fisheries, Food Production and Environment). In this farm, since 1982 different soil management systems have been applied in a normal wheat–sunflower–legume rotation. At present, the soil treatments that are being applied are conventional tillage (CT), minimum tillage (MT) and direct drilling (DD) (Perea et al., 2006). The typical soils in this area are Calcic Vertisol (FAO, 1998), which are locally known as “black lands”. The characterization of the soil used in this study is given in Tables 1 and 2 (Lozano, 2009). From the climatological point of view, the study area is characterized by very dry summers and an irregular distribution of rainfall throughout the year (Table 3). This, together with high temperatures during the summer, leads to an adverse agricultural activity. Rainfed crops are typical in Andalusian soils.
0.25 m
Cup in Ap horizon
0.60 m
Cup in Bw1 horizon
0.40 m Fig. 2. Shape and dimensions of the unaltered soil blocks with cups.
64
B. Lozano-García et al. / Catena 84 (2011) 61–69
2.2. Soil block extraction
a) Ap horizon 1,2
MT CT
0,8
meq/L
This study could not be made in the field due to the uneven distribution of rainfall. Therefore, for this study, three soil blocks (0.40 m × 0.25 m × 0.65 m in deep) were extracted in the three plots on the farm. Each plot had different tillage (CT, MT, and DD). During the short transport to the laboratory, the soil blocks were supported by a wooden frame. The soil blocks were placed on pallets and the wooden frame was removed, following a method similar to the one described by Lozano (2009) (Fig. 1).
DD
1
0,6 0,4 0,2 0 1
2
3
4
5
6
7
8
9
2.3. Soil water status, irrigation and solute sampling
10 11 12 13
b) Bw1 horizon 1,2
During the experiment, the volumetric soil water content was monitored using a TDR (Time Domain Reflectrometry), the model used in this study was ML2x with a dataloger HH2 moisture meter (Delta-T Devices). The three columns were kept under conditions of permanent saturation. Irrigation was applied uniformly to the soil blocks as a fine mist at a rate of 20 mm for only 1 h daily. This irrigation rate although intense was not realistic for the study area, but we needed saturated conditions. Two suction Teflon quartz cups (soil moisture sampler teflon/ PTFE/quartz, Eijkelkamp) were installed in each soil block, one in the Ap horizon and another in the Bw1 horizon, located at depths of 0.12 m and 0.55 m, respectively (Fig. 2). A suction of 70 kPa was applied to each cup. Pressure in the cups was allowed to diminish through time until the samples were collected. The time needed to
DD MT
1
CT
meq/L
0,8 0,6 0,4 0,2 0 1
2
3
4
5
6
8
7
9
10 11 12 13
Fig. 3. Temporal evolution of the sodium concentration in the soil.
Bw1 horizon
DD
Ap horizon 1
1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
0
0
MT
0
3
6
9
12
15
1
1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
0
3
6
9
12
15
0
3
6
9
12
15
0
3
6
9
12
15
0
0 0
3
6
9
12
15 1
0,8
0,8
0,6
0,6
0,4
0,4
0,2
0,2
CT
1
0
0 0
3
6
9
12
15
Fig. 4. The analysis of smoothing, using moving averages to five terms for the case of Na+ in the horizons studied (Ap and Bw1) in the different tillage systems.
B. Lozano-García et al. / Catena 84 (2011) 61–69
integrated and moving average parts of the model, respectively (Box and Jenkins, 1976). The Random walk model is the one which offers a better prognosis for future values and is given by the value of the latest available data (Hughes, 1998). The linear trend model assumes that the best forecast for future values is the linear regression line obtained for the known data series (Chai, 1995). Thirdly, the analysis of variance was performed by standard statistical techniques using tillage practices (CT, MT and DD) and depth extraction of the soil solution (Ap and Bw1) (Kruskal and Wallis, 1952). The treatment mean values were compared using the least significant difference (LSD) at P ≤ 0.05.
collect enough sample for laboratory analysis was 12 days, during which time every 24 hours was applied suction of 70 kPa. From each cup, 13 samples of soil solution were extracted where the sodium and the potassium content was determined (Métodos Oficiales de Análisis, 1994). 2.4. Statistical analysis The results were statistically analyzed using Statgraphics Centurion XV (StatPoint, Inc., 2005). Sodium and potassium were considered as variables. The time series of each element was represented according to the type of tillage (CT, MT and DD) and depth. Secondly, the analysis of the time series was performed to find if there were any trends in the data and the prediction of future values. To determine the trend, a smoothing five-term moving averages was applied (Fuller, 1996; Peña, 1998, 2005). For the prediction of future values, a model was adjusted to each time series to give an outlook on future behavior. In each case, we chose the option that best fitted the series (ARIMA model, Random walk model or Linear trend model), using the Akaike criterion (1974) to assign the model. The ARIMA (p,d,q) model is an autoregressive integrated moving average, where p, d and q are integers greater than or equal to zero and less or equal to two referred to the order of the autoregressive,
3. Results and discussion 3.1. Sodium behavior 3.1.1. Sodium concentration in the soil solution In the three techniques, the sodium concentration in the water in the Ap horizon is less than in the Bw1 horizon. However, the sodium concentration remains constant in the distinct soil horizons in DD and MT and increases in the Bw1 horizon when the CT technique is used (Fig. 3a and b).
Bw1 horizon
DD
Ap horizon 1,5
1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1 -0,5
-0,5 0
5
10
15
20
25
0
5
Linear Trend
MT
65
10
15
20
25
20
25
20
25
Linear Trend
1,5
1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1
-0,5
-0,5 0
5
10
15
20
0
25
5
ARIMA (0,2,2)
10
15
ARIMA (0,2,2) 1,5
1,1
1,1
0,7
0,7
0,3
0,3
-0,1
-0,1
CT
1,5
-0,5
-0,5 0
5
10
15
ARIMA (1,1,2)
20
25
0
5
10
15
Linear Trend
Fig. 5. Tendency prediction from time series analysis for the case of Na+ in the horizons studied (Ap and Bw1) in the different tillage systems.
66
B. Lozano-García et al. / Catena 84 (2011) 61–69
In the Ap horizon, the plot which has a higher sodium concentration is CT with 0.21 meq/L and the one which has a lower sodium concentration is the DD plot with 0.12 meq/L (Fig. 3). Regarding the soil solution from the Bw1 horizon (Fig. 3), the CT is the one which has more sodium, and also increases its concentration in time, from 0.71 meq/L to 0.97 meq/L. MT has a mean concentration of 0.46 meq/L and DD 0.56 meq/L, respectively.
3.1.3. ANOVA The concentration of sodium in the soil solution depends on the tilling technique and sampling depth by the Kruskal–Wallis test at P ≤ 0.05. The soil solution of CT is richer than those of DD and MT, with mean values of 0.563 meq/L, 0.340 and 0.303 meq/L, respectively (Fig. 6). Also, Loch and Coughlan (1984), Dalal (1989) and Thomas et al. (2007) showed that the evolution of the sodium concentration was similar in vertisols and luvisol under conventional tillage and under conservation techniques. In Fig. 6, the existence of two homogeneous groups is explained more clearly in terms of behavior according to the tilling technique. When comparing the soil solution of the two horizons, sodium is more abundant in the Bw1 horizon, with average values of 0.630 meq/L, compared to 0.174 meq/L of the average of the solutions embodied in the Ap horizon, as shown in Fig. 7. Franzluebbers and Hons (1996) also found that exchangeable sodium increased with depth in the two techniques studied (conventional tillage and direct drilling) in a Fluventic Ustochrept in Texas. Our results suggest that CT produces a greater risk of salinization of the Bw1 horizon than conservation techniques since it releases more sodium to the soil solution throughout the profile and causes an increased accumulation in the Bw1 horizon. Furthermore, DD retains
meq/L
Fig. 7. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the sodium variable.
some degrees of moisture in the soil surface, reducing the mobility of sodium, because it reduces abrupt changes in the wetting drying of the soil.
3.2. Potassium behavior 3.2.1. Potassium concentration in the soil solution With regard to potassium, using the three tillages, the soil solution concentration is greater in the Ap horizon than in the Bw1 horizon, increasing its concentration in time. Therefore, initially, the values are 0.05 meq/L for DD, 0.11 meq/L for MT and 0.09 meq/L at CT, ending the experiment with values of 0.10 meq/L, 0.16 meq/L and 0.18 meq/L, respectively (Fig. 8a). Fig. 8b shows how MT and CT have the same behavior. DD is the one which supplies the least amount of potassium to plant roots, with an average value of 0.09 meq/L. In the case of the Bw1 horizon, the behavior in all three cases is the same. Constant contents in the initial
a) Ap horizon 0,25 DD
0,2
MT CT
meq/L
3.1.2. Time series analysis After performing smoothing using moving averages to five terms to highlight the data trends it was shown that in the case of sodium in the conservation techniques, the tendency was to maintain a constant concentration in soil water (Fig. 4). In fact, both DD and MT with values ranging in the Ap horizon between 0.1 and 0.2 meq/L and in the Bw1 horizon about 0.55 meq/L were observed. Furthermore, in CT the sodium concentration in the soil solution of the Ap horizon decreased throughout the experiment, while the Bw1 horizon concentration increased. Although the forecasts were made from the model that best fitted the data set, it was found that the prediction for sodium was valid only for two situations, MT and CT. In the case of MT the prediction was for the Ap horizon, which conformed to an ARIMA (0,2,2) model and in the case of CT the prediction was for the Bw1 horizon, which adjusted to a linear trend model. In the other cases, the predictions were unreliable (Fig. 5).
Na+
0,15 0,1 0,05 0 1
2
3
4
5
6
7
8
9
10 11 12 13
b) Bw1 horizon 0,25 0,2
meq/L
Na+ meq/L
0,15
DD
0,1
CT
MT
0,05 CT
MT
DD
0 1
Fig. 6. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the sodium variable.
2
3
4
5
6
7
8
9
10 11 12 13
Fig. 8. Temporal evolution of the potassium concentration in the soil.
B. Lozano-García et al. / Catena 84 (2011) 61–69
extractions are followed by an increase in the third or fourth extraction, so this value remains the same until the end. The MT solution is the poorest in potassium, with an average concentration of 0.02 meq/L, CT being the richest with a mean concentration of 0.04 meq/L. 3.2.2. Time series analysis In the case of potassium (Fig. 9), the smoothing technique shows how the Ap horizons of the three techniques increase the potassium concentration, while in the Bw1 horizon the tendency is to keep the concentration constant. The prediction is valid for DD, MT and CT. In the case of DD, the prediction is for the Bw1 horizon, which conforms to a random walk model. For the MT the prediction is for the Bw1 horizon, which conforms to an ARIMA (2,1,2) model, and in the case of CT the prediction is for the Ap and Bw1 horizons, which adjust to a linear trend model in both cases. In the other cases, the predictions are unreliable (Fig. 10). 3.2.3. ANOVA The concentration of potassium in the soil solution depends on the tilling technique and sampling depth by Kruskal–Wallis test at P ≤ 0.05. Firstly, it can be considered that the plots of DD and MT behaved similarly because the averages were homogeneous. On the other hand, MT and CT also behaved similarly. Fig. 11 clearly shows the values of potassium in MT, which overlap at one end with the DD and the other with CT.
67
By contrast, the results of Thomas et al. (2007) and Martín-Rueda et al. (2007) refer to a higher content of exchangeable potassium in the absence of tillage. In the case of comparing only CT and DD, a higher potassium content is seen in traditional cultivation. This is the case of the results of Hunter and Cowie (1989), Franzluebbers and Hons (1996), DeMaría et al. (1999) and Guzmán et al. (2006). However, in this study, the potassium concentration was greater in solutions extracted from the Ap horizons, which reached average values of 0.125 meq/L, whereas the average in Bw1 horizons was 0.030 meq/L (Fig. 12). In this case, the results coincide with those of Thomas et al. (2007) in Australian luvisol, Guzmán et al. (2006) in clay soil of Manhattan and Asghar et al. (1996) in a vertisol in Queensland. Also, Lal et al. (1990) and Ismail et al. (1994) found that potassium increased in the surface horizon, although in this case the tillage practices were reduced. However Orihuela et al. (2001) found no influence in the amount of potassium in the subsurface horizon when they studied the leaching of the soluble form of potassium in clay loam texture soils in Huelva (Spain). A higher concentration of exchangeable K+ in the surface layers of non-tillage treatments was apparently due to the return of crop residue on the soil surface. This is extensively documented for different types of soils (Lal et al., 1990; Franzluebbers and Hons, 1996; Martin-Rueda et al., 2007; Thomas et al., 2007) and also occurs in our study (Table 2). Although our data suggest that the extracted soil solutions between 0 and 0.24 m (Ap horizon) in conservation techniques are richer in
Fig. 9. The analysis of smoothing, using moving averages to five terms for the case of K+ in the horizons studied (Ap and Bw1) in the different tillage systems.
68
B. Lozano-García et al. / Catena 84 (2011) 61–69
DD
Ap horizon
Bw1horizon
0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05
0
0 0
5
10
15
20
25
0
5
MT
ARIMA (0,1,2) 0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05
0
15
20
25
20
25
20
25
0 0
5
10
15
20
25
0
5
ARIMA (0,1,2)
CT
10
Random Walk
10
15
ARIMA (2,1,2)
0,3
0,3
0,25
0,25
0,2
0,2
0,15
0,15
0,1
0,1
0,05
0,05 0
0 0
5
10
15
20
25
0
5
Linear Trend
10
15
Linear Trend
Fig. 10. Tendency prediction from time series analysis for the case of K+ in the horizons studied (Ap and Bw1) in the different tillage systems.
potassium. Our study shows how CT behaves in that sense, just as MT with a higher intake of potassium in the soil solution than DD. 4. Conclusions The use of undisturbed soil blocks to study the soil solution in the laboratory is a particularly advantageous method when weather conditions are adverse site.
The soil in this study was significantly affected by 25 years of different systems of soil management. CT releases more Na+ and K+ to the soil solution in Ap and Bw1 horizons than the conservative techniques. In depth, Na+ is more abundant in soil solution than in the surface horizon. In CT, there is a risk of salinization in the Bw1 horizon because the concentrations of Na+ in the soil solution are tripled in respect to the surface horizon. In the Bw1 horizon, K+ remains constant in all techniques studied. However, K+ increases in the Ap
0.1 2
K
+
K+
meq/L 0.1
meq/L
0
CT
MT
DD
Fig. 11. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the potassium variable.
Fig. 12. Mean values and intervals LSD (less significant difference) at P ≤ 0.05 for the potassium variable.
B. Lozano-García et al. / Catena 84 (2011) 61–69
horizon and DD is the management that releases less K+ in the soil solution. The applications of statistical models to predict and evaluate the behavior of these cations in the future provide more reliable predictions for sodium than for potassium. We can predict for potassium only for CT while for sodium we can make predictions for the three techniques. According to the data obtained, CT is not advisable because of the large amount of Na+ released into the Bw 1 horizon. In the Conservationist techniques, DD releases less K+ in the soil solution in the Ap horizon. We recommend MT for the best management practice for the edafic–climatic condition studied because there is no risk of salinization and releases more K+ than DD. Acknowledgements The authors are grateful to the Ministry of Education and Science of Andalusia and the University of Cordoba for the formation scholarship granted for the development of the doctoral thesis of Beatriz Lozano Garcia. We would also like to thank the I.F.A.P.A. “Las Torres - Tomejil” of the Agriculture and Fisheries of the Andalusian for the facilities offered. References Akaike, H., 1974. A new look at statistical model identification. Transactions on Automatic Control AC-19, 716–723. Asghar, M., Lack, D.W., Cowie, B.A., Parker, J.C., 1996. Effects of surface soil mixing after long-term zero tillage on soil nutrient concentration and wheat production. In: Asghar, M. (Ed.), Proceedings of the 8th Australian Agronomy Conference, The Australian Society of Agronomy Inc. Toowoomba, Queensland, pp. 88–91. Ayers, R., Westcot, D., 1987. La calidad del agua en la agricultura, vol. 29. Estudio FAO Riego y Drenaje, Roma. Box, G.E.P., Jenkins, G.M., 1976. Time Series Analysis, Forecasting and Control. HoldenDay, San Francisco. Chai, Feng-Shun, 1995. Construction and optimality of nearly linear trend-free designs. Journal of Statistical Planning and Inference 48, 113–129. Dalal, R.C., 1989. Long-term effects of no-tillage, crop residue, and nitrogen application on properties of a Vertisol. Soil Science Society American Journal 53, 1511–1515. DeMaria, I.C., Nnabude, P.C., de Castro, O.M., 1999. Long-term tillage and crop rotation effects on soil chemical properties of a Rhodic Ferralsol in southern Brazil. Soil and Tillage Research 51, 71–79. FAO, ISRIC, ISSS, 1998. World Reference Base of Soil Resources. Soil Map of the world 1:5.000.000. Reports 60. FAO, Rome, Italy. Franzluebbers, A.J., Hons, F.M., 1996. Soil-profiles distribution of primary and secondary plant-available nutrients under conventional and no tillage. Soil and Tillage Research 39, 229–239. Fuller, W.A., 1996. Introduction to Statistical Time Series, 2nd edition. Wiley, New York. G.T.N.M.A. (Grupo de trabajo de normalización de métodos analíticos), 1976. Determinaciones analíticas en suelos II. Anales de Edafología y Agrobiología XXXV 7 (y 8), 813–834. Guzmán, J.G., Godsey, C.B., Pierzynski, G.M., Whitney, D.A., Lamond, R.E., 2006. Effects of tillage and nitrogen management on soil chemical and physical properties after 23 years of continuous sorghum. Soil and Tillage Research 91, 199–206.
69
Hughes, B.D., 1998. Random walks and random environments. Bulletin of the American Mathematical Society 35 (n° 4), 347–349. Hunter, H.M., Cowie, B.A., 1989. Tillage and stubble effects on nutrient distribution. In: Noble, R.M., Radford, B.J., Ruthenberg, S.J., Thorburn, P.J. (Eds.), Biloela Research Station Programme Review. Queensland Department of Primary Industries, Queensland, pp. 61–62. Ismail, L., Blevins, R.L., Frye, W.W., 1994. Long-term no-tillage effects on soil properties and continuous corn yields. Soil Science Society American Journal 58, 193–198. Jiménez, J.A., García, I., Vanderlinden, K., Perea, F., Muriel, J.L., 2005. Balance de agua en suelos arcillosos bajo laboreo convencional y siembra directa. Actas del Congreso Internacional sobre Agricultura de Conservación, Córdoba. 397–402. Kruskal, W.H., Wallis, W.A., 1952. Use of ranks in one-criterion variance analysis. Journal of the American Statistical Association 47 (260), 583–621. Lal, R., Logan, T.J., Fausey, N.R., 1990. Long-term tillage effects on a Mollic Ochraqualf in North-West Ohio. III. Soil nutrient profile. Soil and Tillage Research 15, 371–382. Loch, R.J., Coughlan, K.J., 1984. Effects of zero tillage and stubble retention on some properties of a cracking clay. Australian Journal of Soil Research 22, 91–98. Lozano, B., 2009. Estudio del comportamiento hídrico en ZNS (Zona No Saturada) de un Vertisol del Valle Medio del Guadalquivir. Tesis Doctoral de la Universidad de Córdoba. Servicio de Publicaciones de la Universidad de Córdoba. Córdoba. Martín-Rueda, I., Muñoz-Guerra, L.M., Yunta, F., Esteban, E., Tenorio, J.L., Lucena, J.J., 2007. Tillage and crop rotation effects on barley yield and soil nutrients on a Calciortidic Haploxeralf. Soil and Tillage Research 92, 1–9. Métodos oficiales de análisis, tomo III., 1994. Ministerio de Agricultura, Pesca y Alimentación, Dirección General de Política Agraria. Edita Secretaría General Técnica. Madrid. Muriel, J.L., Vanderlinden, K., Perea, F., García, I., Jiménez, J.A., 2005. Dinámica del agua en suelos sometidos a distintos sistemas de laboreo. In: López-Geta, J.A., Rubio, J.C., Martín, M. (Eds.), VI Simposio del agua en Andalucía. IGME. Sevilla, pp. 727–737. Navarro, S., Navarro, G., 2000. Química Agrícola. El suelo y los elementos químicos esenciales para la vida vegetal. Ed. Munid-Prensa, Madrid. Ordoñez, R., González, P., Giráldez, J.V., Perea, F., 2007. Soil properties and crop yields after 21 years of direct drilling trials in southern Spain. Soil and Tillage Research 94, 47–54. Orihuela, D.L., Hernández-Domínguez, J.C., Romero, E., González, A., 2001. Lixiviación de formas solubles de Ca++, K+ y PO= 4 y su relación con la c.e. y el pH en condiciones experimentales (Huelva. España). In: López Rodriguez, J.J., Quemada Sáez-Badillos, M. (Eds.), En Temas de Investigación en la Zona no Saturada. Pamplona. Peña, D., 1998. Estadística. Modelos y Métodos II. Modelos Lineales y Series Temporales. Ed. Alianza Universidad Textos. Madrid. Sexta edición. Peña, D., 2005. Análisis de Series Temporales. Ed. Alianza Editorial. Madrid. Perea, F., 2004. Estudio de la influencia del manejo del suelo en el régimen térmico de los vertisuelos de la Vega de Carmona. En Carel, Carmona Revista de Estudios Locales, vol. II. 2, pp. 32–45. Sevilla. Perea, F., Jiménez, J.A., García, I., Vanderlinden, K., Muriel, J.L., 2006. Caracterización hidroclimática en vertisuelos de la Campiña de Carmona. En Carel, Carmona Revista de Estudios Locales IV. N° 4, 11-18.Sevilla. Richards, L.A., 1947. Pressure-membrane apparatus construction and use. Agricultural Engineering 28, 451–454. StatPoint, Inc., 2005. Statgraphics Centurion XV User Manual. Printed in USA. Szabolcs, I., 1989. Amelioration of soil in salt affected areas. Soil Technology 2, 331–344. Thomas, G.A., Dalal, R.C., Standley, J., 2007. No-till effects on organic matter, pH, cation exchange capacity and nutrient distribution in a Luvisol in the semi-arid subtropics. Soil and Tillage Research 94, 295–304. Thornthwaite, C.W., 1948. An approach toward a rational classification of climate. Geographical Review 38 (n° 1), 55–94. U.S. Salinity Laboratory Staff, 1954. Diagnosis and improvement of saline and alkali soils. US Dept. Agr. Hanbook, vol. 60. Washintong DC. USA. Wild, A., 1992. Condiciones del suelo y desarrollo de las plantas según Russell. Ed. Mundi-Prensa. Madrid.