The impact of precipitation and temperature fluctuations on soil salinity in semi-arid regions in Southeast Spain A scenario based on present-day terrace data

(1)

fluctuations on soil salinity in semi-arid regions in

Southeast Spain

A scenario based on present-day terrace data

Author: Fleur van Langen Student number: 10581316 Bachelor thesis

July 3rd, 2016

University of Amsterdam

Supervisor: dhr. dr. Erik Cammeraat

(2)

ABSTRACT

This research focuses on the impact of precipitation and temperature fluctuations on the chemical properties of the soil on a lake-side terrace in Southeast Spain. Influenced by climate change, precipitation and temperature fluctuations are expected to become stronger, which will have implications on the management of the soil and its environment. In this research, the distribution of chemical properties of soils on terraces with different elevations was studied. Soil samples were derived from the field and analyzed in the laboratory. Soil chemical properties that were studied include pH, electrical conductivity, gypsum content and Sodium Adsorption Ratio. The data retrieved in the laboratory show high numbers of pH, electrical conductivity and sodium content indicating a high soil salinity hazard and moderate to high sodium hazard. With this present-day terrace data and climate data retrieved from databases, a model was created that simulates soil salinity in the next 20 years. From this scenario, conclusions can be drawn about the severity of the impact of climate fluctuations on chemical soil properties. Combined with the analysis of the sodium content, this scenario gives an insight in the agricultural use of the soil in the future. The productivity of the soil will decrease further due to high salinity stress.

(3)

CONTENTS

Introduction 4

Relevance and aim 5

Relevance 5 Aim 5 Research questions 5 Main question 5 Sub questions 6 Hypothesis 6 Methods 7 Fieldwork 7 Laboratory work 9

Classification and statistics 13

Modelling 14

Results 15

Chemical properties 15

Classification and statistics 20

Modelling 22

Discussion 25

Laboratory 25

Modelling and future soil use 27

Limitations of the research 27

Recommendations for further research 28

Conclusions 29

Acknowledgements 30

References 31

Appendices 33

Appendix 1: Laboratory results 33

Appendix 2: Modelling results 40

Appendix 3: Additional MATLAB results 42

Appendix 4: Soil classification 43

(4)

INTRODUCTION

Soil salinity might not seem as dramatic as major natural hazards such as landslides or earthquakes but is seen as an important environmental risk. According to Rengasamy (2006), salt affected soils occur in 100 countries in the world. Salinization is a complex process involving the movement of soluble salts and water in soils during seasonal cycles and interactions with groundwater (Rengasamy, 2006). This research will focus on the influence of precipitation and temperature fluctuations on the chemical properties of soils in the semiarid region of Southeast Spain,

taking into account the soil salinity, but also the Sodium Adsorption Ratio and gypsum content. The area that will be investigated is situated in the Betic Cordillera, of which the geological formations form part of the Sub-Betic system. The lithology is dominated by Cretaceous and Tertiary marls and Eocene limestone (Cammeraat, 2004).

Figure 1. The study area in Southeast Spain (Cammeraat, 2002)

Properties of the soils in the study area are calcareous and are therefore prone to erosion. Due to climate change, this region deals with high variability in precipitation and evaporation. Previous climate models predict increasingly arid conditions in most of the Iberian Peninsula. Also, models suggest an increase in extreme precipitation events, with both dry periods and wet periods. Furthermore, an increase in extreme high-temperature events is predicted, specifically in the south (Sánchez & Miguez-Macho, 2010). These fluctuations have an impact on the properties of soils and can cause severe erosion, which is for a large part the driving force of land degradation (Sinoga et al., 2012). Additionally, desertification is an important issue in this area. Therefore, the intensity and duration of precipitation might have an influence on the soil, as leaching may occur, that causes a change in the properties of different soil horizons.

(5)

RELEVANCE AND AIM Relevance

The subject of this research is scientifically relevant as no scenarios for the future have been provided for the link between soil salinity and climate change for this region. Therefore, new insights will be gained and other researches can be carried out. Also from a political vision this research is relevant as climate change is an upcoming topic and governments are interested in the consequences of global warming. Finally, land degradation due to climate change and increasing soil salinity is a socio-economic problem, not only for local famers but also for the society. Soils provide important ecosystem services such as agriculture that is important for reducing food scarcity but also other services such as water maintenance and the ecological sustainability (Soil Science Society of America, 2014).

Aim

The aim of this research is to clarify how chemical properties react to precipitation and temperature fluctuations and include these chemical properties and climate fluctuations in a model to simulate a future scenario. This local information can be applied on larger scale so anticipations can be made for the future. The final product of this research will be a scenario created for 20 years in the future in AquaCrop, in which the achieved information about precipitation, temperature and chemical properties of the soil will be used. The research region has been investigated extensively, but modelling the influence of precipitation and temperature on chemical properties of the soil and creating a scenario for soil salinity is novel in this region. Therefore this research will create a new insight in mitigating land degradation.

Research questions Main question

What is the influence of precipitation and temperature fluctuations on the distribution of chemical properties in soils on terraces near lakesides in the semiarid region of southeast Spain?

(6)

Sub questions

- How large is the variability of chemical properties between terraces and soil horizons regarding pH, electrical conductivity, Sodium Adsorption Ratio and gypsum

content?

- How do fluctuations in precipitation and temperature influence distribution of different chemical properties in the soil?

- How will soil salinity change from the current state in 20 years and to what extent will the soil be suitable for agriculture use?

Hypothesis

The hypothesis of the research is that the distribution of salt in soils on terraces is highly influenced by precipitation and temperature fluctuations as the amount of leaching is determined by variation in precipitation. Leaching will alter the distribution of salt on a terrace. It will decrease the abundance of salt in the upper horizons so more salt accumulates in the lower horizons or even the lower terraces. Furthermore, the model is expected to create scenarios that differ from each other due to the presence of the lake in the lower terraces. Also, the data that comes out of the simulation is expected to be high in salinity which will likely have a negative impact on the agricultural use of the soil.

(7)

METHODS Fieldwork

To gain present-day data, fieldwork was conducted. In the field, three sampling locations were appointed on each terrace. It is decided to do this in triplo so the research data is more evident when carrying out statistical analyses. The terraces that are included in the research are terrace 0, terrace 2, terrace 4, terrace 6, an area that has recently been saturated (recently saturated zone: RSZ) and an area that is

saturated (saturated zone: SZ, see figure 3). The sample spots are on the edge of a gully as it is often hard to make a soil pit in dry land regions. From the sides of the gully, 7 samples were taken at different depths: 10, 20, 30, 40, 50, 75 and 100 centimeters. This way, the chemical composition at different soil depths can be compared. On the areas that are not included in the terrace, a soil pit was made and samples were taken as deep as possible, also with distances of 10 centimeters.

(8)

Figure 3. Sampling locations on the terraces.

(9)

Laboratory work

pH and electrical conductivity

pH-values and electrical conductivity (EC) were measured to get an insight in the soil salinity. To prepare the test tubes, 10 grams of the sieved material was diluted with 25 ml of demi-water and were shaken over night. Afterwards, both the pH and EC were measured from the same prepared sample with a pre-calibrated meter in !S/m. The EC was measured first, directly from the tube. For measuring the pH, approximately 5 ml was decanted into another test tube, as the pH meter contains electrodes that could interfere with the diluted sample.

The data on EC will be interpolated and mapped in a Geographical Information System (GIS) so more conclusions can be drawn about the spatial distribution of soil salinity.

Figure 5. Equipment for measurement of EC and pH. ICP analysis

Using Inductively Coupled Plasma Mass Spectrometry (ICP), the presence of several cations can be found: sulfur (S), phosphorus (P), magnesium (Mg), calcium (Ca), sodium (Na) and potassium (K). To get the samples ready for the ICP analysis, a water extract was prepared. This is done by using the solution used for the EC and pH analyses. The solution was centrifuged first and filtered by a vacuum filter so that a clear solution remains. This solution was transported into clean bottles that were labeled and transferred to the laboratory analyst that has access to the ICP meter. Because the ICP analysis retrieves so many numbers, it is

(10)

chosen to analyze only a selected group of cations so the Sodium Adsorption Ratio (SAR) can be calculated (using equation 1). These include sodium, calcium and magnesium. The SAR is an indicator for the suitability of water for irrigation purposes. It can also be used to indicate the suitability of the soil for agricultural purposes. Based on both the SAR value and the salinity of the soil measured by EC, an indication can be made regarding the suitability for agriculture using the classification system for salinity and sodium hazard (figure 6).

Equation 1. Sodium Adsorption Ratio.

Figure 6. Sodium and salinity hazard (Agriculture Handbook, vol. 60. U.S. Dept. of Agriculture).

Salinity hazard:

C1: Low-salinity water: can be used on virtually all crops, without risk of soil salinization C2: Medium-salinity water: can be used if soil is thoroughly leached

C3: High-salinity water: unsuitable for soils with limited drainage C4: Very high-salinity water: unsuitable for irrigation

(11)

Sodium hazard:

S1: Low-sodium water: can be used for irrigation in virtually all soils

S2: Medium-sodium water: can cause problems in fine-textured soils with bad drainage, no problem in coarse-textured soils with good drainage

S3: High-sodium water: can cause problems but always good drainage, strong leaching; chemical additives may be required

S4: Very high-sodium water: unsuitable Gypsum content

Gypsum content is measured as some saline-alkali soils contain gypsum and this can have an influence on the chemical properties of the soil. For the measurement of the gypsum content, it is assumed that there is gypsum present in the material. This is based on the results of the ICP analysis described in the previous section. Before starting with the experiment, a solution was prepared. This solution is made of potassium chloride (4 gram K/L) and hydrochloric acid (0.02 M HCL). First of all, 50 ml test tubes were filled with 3 grams of sieved soil (1 mm) and 30 milliliters of demi-water. The test tubes were shaken overnight, for 16 hours.

After shaking the samples, they were centrifuged so the water extract is segregated from the soil. From this extract, 20 milliliters was pipetted into another test tube, where 20 milliliters of acetone was added to. These were shaken briefly and centrifuged until the supernatant solution cleared. After this the liquid was decanted, taking care that no precipitate is lost. In the same test tube, another 10 milliliters of acetone was added, pipetting it along the wall of the test tube. Again, this was centrifuged and decanted. The next step was to dry them in the oven at 50 degrees Celsius over the night until the tube is completely dry. Last in the preparation process was to add 12 milliliters of the created solution and shake briefly. This solution was measured by the Spectrometer. This instrument measures in mg Ca/L so a calculation of percentage gypsum or (g/100g) is given in equation 2.

where

s % Gypsum = Ca mg/l x 0,172

(12)

s = air-dry sample [gram]

0,172 =

Equation 2 (van Reeuwijk, 2002).

Loss on Ignition

Loss on Ignition was measured to determine different chemical properties of the soil. It was chosen to carry out this method because it requires relatively minor materials and it gives many results. This method was carried out in a Carbolite furnace CSF 1100. Approximately 2 grams of soil material was placed in crucibles. The crucibles were heated first to 105 degrees Celsius and for 2 hours. Afterwards, the samples were cooled down in desiccators and the weight loss was measured. This was repeated at 150 degrees Celsius to measure the influence of gypsum in the soil material. The crucibles were heated up gradually until 375 degrees Celsius is reached for 16 hours, to combust all organic matter in the samples. The weight loss after this process was measured and the difference between the 150 and the 375 degrees heated samples is the amount of organic matter. Once again this was repeated up to 550 degrees to combust all organic matter. Also calcium was measured by heating the crucibles up to 900 degrees Celsius. The weight loss of this process is the amount of calcium. The following calculations were used to gain the percentages of lost material:

% moisture loss (105ºC) = (b-c) / (c-a) x 100% % gypsum loss (150ºC) = (c-d) / (c-a) x 100% % OM1 loss (375ºC) = (c-e) / (c-a) x 100% % OM2 loss (550ºC) = (c-f) / (c-a) x 100% % Calcium loss (900ºC) = (c-g) / (c-a) x 100% a = empty crucible

b = empty crucible + air-dried sample c = weight of crucible after drying at 105ºC d = weight of crucible after drying at 150ºC e = weight of crucible after drying at 375ºC f = weight of crucible after drying at 550ºC g = weight of crucible after drying at 900ºC

(13)

Classification and statistics Classification

As the extended laboratory analysis retrieved many numbers that provide information about the type of soil in this region, the soils on the terrace were classified. With this information more valuable conclusions can be drawn. This was done using the World reference base for soil resources 2014 created by the IUSS Working group. This is an international soil classification system for naming soils and creating legends for soil maps. It is important that soil description

is done correctly as it provides the basis for soil classification. Therefore, the Guidelines for Soil Description were used to classify soils (FAO, 2006). This report gives a comprehensive description of all analyzed features. Figure 7 provides the steps that were used in describing the soil according to the Guidelines for Soil Description.

Figure 7. Guidelines for soil description (FAO, 2006). Statistics

To answer the research questions, statistics were performed in MATLAB to determine whether there is a difference between the layers and terraces. For pH, EC, SAR and gypsum content, the mean values were compared using a Kruskal-Wallis test. This test assesses whether the values come from the same distribution. Also, a boxplot is created that shows the distribution of salt between the different terraces. The boxplots show terraces 0 to 6 as these terraces have complete information to do statistical analyses on.

(14)

Modelling

In order to gain information about the future development of this terrace, a scenario was created. The model used for this part of the research is AquaCrop. AquaCrop is a crop water productivity model developed by the Land and Water Division of FAO. It mainly focuses on simulating yield response but can also be used for carrying out future climate scenario analyses and computing new data with previous obtained data. The data acquired in the lab was the starting data of the scenario. These were inserted in the outset of the model and include soil properties, climate properties, vegetation and groundwater table.

The first terrace that was modelled is terrace 0. It is chosen to do this as this is the upper terrace so it is expected that it has the least influence of the lake. Also terrace 6 was modelled because this is the lowest terrace that has complete data. In order to incorporate the nearby lake to the simulation, a more shallow groundwater table was implemented in the model of terrace 6. Modelling terrace 0 and 6 will create an insight in what influence the presence of the lake has on the development of soil salinity and whether there could be other factors influencing this diversification.

The data that was used in AquaCrop is retrieved from previous climate data provided by Erik Cammeraat. First of all, the environment was designated. The climate data used for this model is the year 2014 in Lorca. The soil profile and groundwater have to be specified by own input gained from the fieldwork and laboratory studies. It was chosen not to incorporate any vegetation as the terrace is abandoned and incorporating crops may alter the outcomes of the simulation. As it is most interesting to look at the development of salinity, this should not be biased. Next, the simulation data was incorporated. The simulation period is one year, as AquaCrop is not able to model for more years in a row. Finally, the initial water and salinity contents were defined by specifying the soil water content in percentages and soil salinity in dS/m. These were retrieved from the laboratory data. After this, the simulation can run for one year and this was repeated 20 times.

(15)

RESULTS

Chemical properties

As is it interesting to look at the distribution of chemical properties, it is chosen to implement only the tables that show the means of the collected data and the graphs created in MATLAB. Also, only the chemical properties that are used to answer the questions stated in the outset of this research were analyzed. All the numbers that are retrieved in the laboratory can be found in appendix 1.

Figure 8 shows the distribution of the pH-value in soil horizons on the different terraces. On the Y-axis the soil horizons are displayed from the surface to 1 meter deep (soil horizon 7). The X-axis shows the pH-value that ranges from 7,6 to 9,2 as high values of pH are found. The graph shows a significant difference between the upper terraces, that are represented by the yellow, orange and blue lines, and the terraces near the lake, that are shown in green and purple. Also, terrace 0 shows a high pH in the upper horizon while terrace 0 and 2 show no disparity in the upper horizons and terrace 6 and 7 show a lower pH in the upper horizons.

Table 1. Values of pH on different terraces.

Depth (cm) Terrace 0 Terrace 2 Terrace 4 Terrace 6 RSZ SZ

10 8,30 7,8 7,81 8,26 9,03 7,83 20 8,00 7,77 7,74 8,46 8,715 30 7,85 7,82 7,72 8,65 8,53 40 7,88 7,93 7,81 8,64 50 7,95 8,25 7,71 8,47 75 7,91 7,88 7,72 8,83 100 7,82 7,84 7,85 8,98 Standard error: 0.3703

(16)

Figure 8. Distribution of pH on different terraces.

Figure 9 shows the distribution of EC in the same coordinate system as figure 8. Similar to the pH-values, the numbers of EC show a significant difference between the upper terraces and the lower terraces. However, a boxplot of the EC shows that EC increases gradually when getting closer to the lake (this can be found in the statistics section). Also, for the upper terraces, the EC increases in depth.

Table 2. Values of EC on different terraces (!S/m).

10 867 1805,33 1338 3963,33 11105 5255 20 1040,67 2300 1790,33 4943,33 6825 30 1880,33 2360,67 2640 8810 4975 40 2046,67 2473,33 3140 9043,33 50 1590,67 3383,33 3746,67 6686,67 75 2209,33 3190 3110 5320 100 2300 3901,67 3238 5246,67 Standard error: 0,0028

(17)

Figure 9. Distribution of EC on different terraces . ICP analysis

From the ICP analysis, different chemical elements are measured. They include Sodium (Na), Potassium (K), Magnesium (Mg), Calcium (Ca), Phosphorus (P) and Sulphur (S). The results of this analysis can be found in appendix 1. As this research is not extended enough to provide information on all chemical elements, the Sodium Adsorption Ratio (SAR) is calculated. The outcomes of the SAR are displayed in table 3 and figure 10.

Table 3. Calculated values of Sodium Adsorption Ratio.

10 0,485 0,221 1,723 8,596 21,103 9,715 20 0,607 0,309 3,566 14,906 42,918 30 0,684 0,841 2,332 26,638 29,800 40 1,462 2,417 3,848 28,439 50 1,620 4,479 3,523 22,523 75 2,016 5,254 3,744 19,991 100 1,106 2,088 5,206 11,996 Standard error: 8.2987

(18)

On the Y-axis, the soil horizons are shown from the soil surface to 1 meter deep. On the X-axis the SAR values are displayed. The SAR values show a clear gradual increase from the upper terraces to the saturated zones. This indicates the presence of sodium in the water and the absence of sodium in the upper terraces.

Figure 10. Sodium Adsorption Ratio distribution on different terraces.

From figure 6 provided by the U.S. department of agriculture the salinity content shows a high hazard. For the SAR, only the lower terraces, terrace 6 and 7, show a high sodium hazard.

(19)

Gypsum content

The outcomes of the measured gypsum are displayed in table 4 and figure 11. They show a large variability in distribution of gypsum in the soil, in the different terraces as well as the different depths.

Figure 11. Gypsum percentage on different terraces

Table 4. Mean gypsum content on different terraces (%)

10 0,954 8,493 5,181 15,222 9,401 7,778 20 4,100 17,272 11,041 12,045 11,367 30 7,929 17,137 12,528 11,539 9,486 40 8,698 17,023 11,481 12,921 50 6,006 17,433 11,149 12,977 75 7,591 15,090 10,501 11,416 100 14,251 12,042 10,234 12,885 Standard error: 4.1057

(20)

Classification and statistics Classification

The lower terraces show high values of EC which indicate they are saline soils. Because the soils have no natric horizon, a dense subsurface horizon with a higher clay content than the overlying horizons, they cannot be classified as Solonetz. However, the soils do have a salic horizon starting <50 cm from the soil surface as well as the absence of a thionic horizon starting <50 cm from the surface. Also, they are not permanently submerged by water and not located below the line by tidal water. With this information, the soil can be classified as a Solonchak. The soils contain a gypsic horizon as there is more than 5 percent gypsum present in the soil. This indicates their principal classifier. The soils on the lower terraces are thus classified as Gypsic Solonchaks.

The upper terraces cannot be classified as Solonchaks because of the lower EC content. They do have a high gypsum content which might indicate they are Gypsisols. However, on terrace 0, not all the horizons show a sufficiently high gypsum content to classify them as a gypsic horizon. Therefore, the soils on terrace 0 are classified as Calcisols. Because this soil contains horizons with a high gypsum content the principal classifier is gypsic, which makes it a Gypsic Calsisol. Terraces 2 and 4 are classified as Calcic Gypsisols because of their high CaCO3 content as well as high gypsum content.

Statistics

A boxplot of the pH from terraces 0 to 6 shows that pH-values tend to decrease when approaching the lake, with the exception of terrace 6 which shows significantly higher values (figure 11). In contrast to the pH-values, the EC values show a clear increase lower on the terrace. Also, the EC shows a larger variability between values on the terraces than the pH value, shown in the larger boxplot.

(21)

Figure 12. Boxplot of pH and EC on terraces 0 to terrace 6.

Figure 13 shows the boxplot of the Sodium Adsorption Ratio and gypsum content on the same terraces as the pH and EC. The SAR shows a significant difference between the upper terraces and lower terrace. The values of terrace 6 are high which indicates an abundance of sodium. The gypsum content shows large fluctuations and several outliers. There is no clear distribution of gypsum visible.

Figure 13. Boxplot of SAR and Gypsum content on terraces 0 to terrace 6.

Moreover, a Kruskal-Wallis test in MATLAB accepts the null hypothesis that the terraces come from the same distribution as the P-values retrieved from this test are small. This is the case for all analyzed chemical properties.

(22)

The P-values for this test can be found in table 5. Because the determination of gypsum content can be a difficult test, gypsum content is also tested with loss on ignition. The means of these values can be found in appendix 1. A correlation test is done between the outcomes of the gypsum test of the spectrometer and the loss on ignition test and there is said to be no correlation between the to with a P-value of 0.2727.

Table 5. P-values of pH, EC, SAR and gypsum content of different terraces

Modelling

The outcome of the modelling component of the research are two graphs of terrace 0 and terrace 6 that show the development of the EC of each soil horizon from 0 to 20 years in the future. The Y-axis represents the EC in dS/m, the X-axis shows the simulation in years from 0 to 20. The lines describe the different horizons which can be found in the legend.

Striking about the outcome of the scenario of terrace 0 is the drop in EC value in the upper horizons (0-50 centimeters) first years and their great increase in the years after, while the lower horizons (50-100 centimeters) remain relatively stable.

pH EC (!s/m) SAR Gypsum (%)

(23)

Terrace 0:

Figure 14. The scenario for the next 20 years of terrace 0

The scenario of terrace 6 differs a lot from terrace 0. In the upper horizons (0-10 and 10-20 centimeters) a shift is visible in which the salt of the second horizon is relocated to the upper horizon after 6 years of simulation. The rest of the horizons (20-100 centimeters) remain stable at a high pH value.

(24)

Terrace 6:

(25)

DISCUSSION Laboratory

Soils are generally classified as saline when they have an EC of 400 !S/m or more and sodic when they have an SAR greater than 13 (Sumner, 1995). Hence, soil salinity in the research area is notably high. The increase in salinity with depth is not gradual, as the figures show some sharp increases and decreases. Even though this is not the case, some relations can be made. The high values of pH and EC in the lower terraces can be explained by the presence of the lake that has a high salt content. The higher EC in the lower soil horizons can be related to higher salt concentrations due to relocation processes. These processes are downward processes due to precipitation. However, Khosla et al. (1979) state that 0,4 cm leaching water per centimeter soil depth is required for the water to be passed through the soil in saline-sodic soils. As the investigated area is in a semi-arid region, the relocation of salts could stagnate once in a while when dry periods occur. In addition, in this research only the upper meter of the soil is examined, so no information is present referring to deeper in the soil. In this part of the soil, relocation processes can play an important role as well. The high EC can be explained by the drought in this area, but there are more factors that could play an important role. Darwish et al. (2005) states that salinity is enhanced by a low organic matter content as it deteriorates the buffer capacity of the soil against salinity. Because this is an abandoned field with little vegetation this could be a factor influencing the salinity.

In appendix 5, two maps that show the spatial distribution of salinity can be found as well as one map that shows aggregate stability that is created by Klaver (2016). Because the values are interpolated, the maps are not completely reliable but they show a correlation. This correlation is not completely evident but might indicate that soil salinity has a negative influence on the aggregate stability of the soil, or at least in the investigated research area.

(26)

ICP analysis

From the ICP analysis, the SAR is measured. This ratio shows a clear distribution on the different terraces so the values in study area show a high variety. The higher terraces show a stable distribution in the SAR per horizon. The SAR in the different horizons of the lower terraces fluctuate. According to Sumner (1995) soils are sodic when they have a SAR higher than 13. Hence, only the lower horizons show a high sodium hazard. This might be due to the location of the lake which is expected to have a larger influence on the lower terraces than on the upper terraces. This might be a reason for the relatively stable SAR on the higher terraces, which are more influenced by other factors such as precipitation and evaporation.

Gypsum content

The determination of gypsum in soils is said to be difficult because of different errors that could influence the extraction. They include (1) the solution of calcium from sources other than gypsum; (2) exchange reactions in which soluble calcium replaces other cations; (3) the solution of sulfate from sources other than gypsum (Reitemeier & Christiansen, 1946). These may be causes for the large fluctuations in gypsum content that are found on the examined terraces. The most striking values are the ones found on terrace 2. These values are significantly higher than the other terraces, while this terrace is in the middle of the examined area and no irregularity would be expected here.

Not only the different terraces show large fluctuations, also between the different soil horizons a considerable distribution can be found. On the upper terraces the surface shows a low gypsum content while the subsurface is significantly higher. According to Richards (1954), gypsum commonly occurs at some depth in the soil due to leaching, but in older soils its greatest extent can be found in the surface layers of the soil. The high values in the subsurface horizon might indicate that this horizon has the best conditions for gypsum to form. In the studied area, this might be the case due to precipitation of calcium and sulfate. The surface horizon seems to be unsuitable for gypsum to form which can be caused by the lack of moisture to form gypsum.

(27)

Modelling and future soil use

The modeled scenarios of terrace 0 and terrace 6 give different outcomes. The most important difference are the fluctuations on terrace 0 while the values of terrace are less oscillating. Terrace 0 shows a relocation from the lower horizons to the upper horizons. This takes some years which clarifies the drop in the beginning. On terrace 6, the horizons from 20 centimeters and lower remain stable through the whole simulation. This may indicate a stronger influence from the lake as there is a higher moisture content in the soil in the lower horizons.

The studied area is an abandoned agricultural field which will presumably not be used as an agricultural field in the future. However, this terrace can provide a representation of other agricultural fields in semi-arid regions. The scenario for the future indicates that salt will relocate from the lower horizons to the upper horizons. This will create a high salinity stress. As stated in the relevance part of this research, soils provide important ecosystem services of which food scarcity is major. Therefore, measures should be taken to increase the productivity of salt- and sodium-affected soils. Salt- and sodium-affected soils are usually not suitable for agriculture but low concentrations of salt-tolerant vegetation can sometimes be cultivated, such as barley, cotton and alfalfa (Metternicht & Zinck, 2003). Consequently, salinity decreases the productivity of the soil but there are some solutions to increase this productivity.

Limitations of the research

As there were only three months to execute this research, decisions are made on the chemical properties that are used in the research. Therefore, several parts of the information gained in the laboratory is not used in this research. This creates a less extended analysis on the soil salinity in Spain than what was first intended.

Furthermore, due to the strict time schedule, the simulations made in AquaCrop are extremely simplified. First of all, the climate data that was used was the same for all years. As a result, a simulation is created with the same climate data through 20 years. Secondly, extreme events have not been taken into account. These can be for example flash floods, extreme droughts and extreme lake level rises.

(28)

Recommendations for further research

As the outcome of this research shows high values and the scenario suggests that these will increase further in the next 20 years, this research should be extended in the future. Since this research is highly delimited due to the time restrictions, there are many recommendations that could be made for further research.

First of all, the fieldwork can be executed more precise so each terrace is considered. Additionally, when sampling deeper in the soil, also lower horizons can be investigated. Taken these deeper depths into account, it could give a different insight on the salt relocation of the soil. Also, the sampling date can have an influence on the outcome of the research. Therefore a recommendation is made to sample the exact same samples 6 months later to see how much influence the seasons have on the distribution of soil salinity.

Second, the modelling component of the research should be extended as it is now a highly simplified representation of reality. The main component that should be incorporated correctly is the lake level fluctuation. Also other large disturbances such as floods, fires and extreme droughts should be incorporated. Lastly, vegetation changes could be incorporated because the research by Klaver (2016) shows that vegetation has a significant positive influence on the stability of the soil which could have a link with soil salinity.

(29)

CONCLUSIONS

Because of the extended laboratory analysis a valuable conclusion can be drawn on the variability of the chemical properties. The measured chemical values showed large variability within soil horizons and between different terraces. Due to the drought in the area and the presence of the lake, the pH and EC were higher than expected. Also, both pH and EC but also SAR showed general trends that increased towards the lake. Only in the gypsum content no particular trend was found.

Fluctuations in precipitation and temperature will likely have more influence on the upper soil horizons as the lower horizons are saturated from the groundwater. Therefore both evaporation and precipitation will control the salt relocation and alter the distribution of salt in the soil. The maps created in GIS show that the salinity of the soil has influence on the stability of aggregates. This is an uncertain yet noticeable and interesting conclusion for further research.

The modelling component of the research showed some interesting results that should be further elaborated. The main conclusion drawn from the scenario is that the salt will relocate from the lower horizons to the upper horizons. Combining these outcomes with the results of the sodium adsorption ratio, the soil experiences severe salinity and sodium hazard, especially in the lower terraces of the research area. Therefore the field is not suitable for high agricultural productivity and according to the scenario this productivity will decrease in the future.

To conclude, precipitation and temperature fluctuations are likely to have an important influence of the distribution of chemical properties in soils in this research area. This research presents a scenario that shows a basic insight in the future of agricultural use of saline soils. It is important that this subject is further researched as future climate change will have an impact on the suitability of soils for agriculture, especially in semi-arid regions.

(30)

ACKNOWLEDGEMENTS

This research was guided by Erik Cammeraat. He is thanked for his guidance in the field as well as in executing the research. Chiara Cerli and John Visser are appreciated for their helpful assistance in the laboratory, as well as Leen de Lange who is thanked for executing the ICP analysis. Lastly, Niels Klaver is valued for sharing the erodibility map.

(31)

REFERENCES

Cammeraat, L. H. (2002). A review of two strongly contrasting geomorphological systems within the context of scale. Earth Surface Processes and Landforms, 27(11), 1201-1222.

Cammeraat, L.H. (2004). Scale dependent thresholds in hydrological and erosion response of a semi- arid catchment in southeast Spain, Agriculture, Ecosystems and

Environment, vol. 104, no. 2, pp. 317-332.

Darwish, T., Atallah, T., El Moujabber, M., & Khatib, N. (2005). Salinity evolution and crop response to secondary soil salinity in two agro-climatic zones in Lebanon. Agricultural

Water Management, 78(1), 152-164.

Khosla, B. K., Gupta, R. K., & Abrol, I. P. (1979). Salt leaching and the effect of gypsum application in a saline-sodic soil. Agricultural Water Management, 2(3), 193-202.

Klaver, N.R. (2016). The erodibility of abandoned agricultural terraces in the semi-arid environment of southern Spain: a case study. Bachelorthesis, University of Amsterdam, Amsterdam, The Netherlands.

Metternicht, G. I., & Zinck, J. A. (2003). Remote sensing of soil salinity: potentials and constraints. Remote sensing of Environment, 85(1), 1-20.

Reitemeier, R. F., & Christiansen, J. E. (1946). The Effect of organic matter, gypsum, and drying on the infiltration rate and permeability of a soil irrigated with a high sodium water. Eos, Transactions American Geophysical Union, 27(2), 181-186.

Rengasamy, P. (2006). World salinization with emphasis on Australia. Journal of

experimental botany, 57(5), 1017-1023.

Richards, L. A. (1954). Diagnosis and improvement of saline and alkali soils. Soil

Science, 78(2), 154.

Sánchez E. and Miguez-Macho, G. (2010). Regional climate projections over the Iberian Peninsula: climate change scenarios modeling. Report: Climate in Spain: Past, Present and Future (Editors: Pérez F. Fiz and Boscolo Roberta) pp. 69-80.

Sinoga, J. D. R., Pariente, S., Diaz, A. R., & Murillo, J. F. M. (2012). Variability of relationships between soil organic carbon and some soil properties in Mediterranean rangelands under different climatic conditions (South of Spain). Catena, 94, 17-25.

(32)

Soil Science Society of America, 2014. Soil’s role in restoring ecosystem services. Sumner, M.E., 1995. Sodic soils: new perspectives. In: Naidu, R., Sumner, M.E.

Reeuwijk, L.P., van (2002). Procedures for soil analysis, International Soil Reference and Information Centre.

World Reference Base for Soil Resources (2014). International soil classification system. World Soil Resources Report 106. FAO, Rome, 181 pp.

World Reference Base for Soil Resources (2006b). Guidelines for Soil Description. Fourth Edition, FAO, Rome, 109 pp.

(33)

APPENDICES

Appendix 1: Laboratory results

pH and electrical conductivity Terrace 0 pH EC Terrace 2 pH EC Terrace 4 pH EC

Depth (cm) Location 1 Location 2 Location 3

10 8,16 8,45 8,3 20 7,95 8,49 7,57 30 7,78 7,89 7,88 40 7,81 7,89 7,93 50 7,89 7,94 8,02 75 8,02 7,85 7,87 100 7,85 7,86 7,75

10 1812 230 559 20 1455 281 1386 30 2320 1978 1343 40 2480 2340 1320 50 2440 1374 958 75 2530 2330 1768 100 2180 2530 2190

10 7,77 7,83 7,8 20 7,72 7,83 7,76 30 7,78 7,87 7,8 40 7,89 7,92 7,99 50 8,7 7,84 8,21 75 7,88 7,82 7,94 100 7,8 7,83 7,9

10 2140 1106 2170 20 2280 2310 2310 30 1952 2310 2820 40 2640 2440 2340 50 3250 2560 4340 75 3230 2180 4160 100 3345 4150 4210

10 7,79 8,1 7,53 20 7,8 7,92 7,51 30 7,88 7,53 7,75 40 7,8 7,78 7,86 50 7,7 7,69 7,74 75 7,53 7,82 7,81 100 7,82 7,86 7,86

10 1522 412 2080 20 2510 591 2270 30 2880 2410 2630 40 4180 2270 2970 50 4530 2450 4260 75 3080 2830 3420 100 3130 2924 3660

(34)

Terrace 6

pH EC

Recently Saturated Zone

pH EC

Saturated Zone

pH EC

10 8,09 8,49 8,2 20 8,36 8,44 8,57 30 8,91 8,29 8,76 40 8,82 8,34 8,77 50 8,62 8,28 8,52 75 8,38 8,93 9,19 100 8,97 8,99 8,98

10 2440 6120 3330 20 3150 5980 5700 30 9580 4450 12400 40 9560 4810 12760 50 6000 2760 11300 75 4970 5190 5800 100 5080 5360 5300

Depth (cm) Location 1 Location 2

opp 8,53 8,81

10 8,87 9,19

20 8,37 9,06

30 8,12 8,94

opp 4400 2930

10 13460 8750

20 6300 7350

30 4800 5150

Surface 7,96 7,7

(35)

ICP analysis

Sample Id S_(mmol/L) P_(mmol/L) Mg_(mmol/L) Ca_(mmol/L) Na_(mmol/L) K_(mmol/L)

Blanco 1 0,02 0,0000 0,01 0,04 0,17 0,01 Blanco 2 0,02 0,0000 0,00 0,03 0,02 0,01 Blanco 3 0,02 0,0000 0,00 0,07 0,00 0,01 T0M1-10 0,15 0,0000 0,09 0,49 0,19 0,13 T0M1-20 8,62 0,0005 1,11 7,25 0,66 0,21 T0M1-30 16,57 0,0003 2,63 13,18 1,43 0,27 T0M1-40 18,11 0,0026 2,96 13,09 3,83 0,28 T0M1-50 18,31 0,0021 2,84 13,59 2,96 0,24 T0M1-75 18,09 0,0004 1,93 13,36 6,92 0,24 T0M1-100 16,62 0,0002 1,61 13,84 1,82 0,29 T0M2-10 0,46 0,0039 0,17 0,85 0,26 0,17 T0M2-20 0,38 0,0010 0,16 0,67 0,38 0,14 T0M2-30 13,73 0,0021 2,01 11,06 1,41 0,32 T0M2-40 17,12 0,0024 2,34 13,33 3,36 0,30 T0M2-50 5,86 0,0026 1,19 5,05 3,02 0,27 T0M2-75 15,16 0,0000 2,55 12,84 3,16 0,30 T0M2-100 17,07 0,0017 2,00 13,81 4,56 0,26 T0M3-10 1,21 0,0000 0,21 1,70 0,31 0,14 T0M3-20 6,97 0,0030 0,92 7,38 0,76 0,35 T0M3-40 7,21 0,0019 0,73 6,78 0,76 0,19 T0M3-50 4,17 0,0000 0,48 4,09 0,83 0,13 T0M3-75 10,86 0,0018 0,91 9,78 1,61 0,15 T0M3-100 17,07 0,0019 1,85 13,98 1,78 0,25 T2M1-10 16,13 0,0006 0,88 14,33 0,20 0,81 T2M1-20 17,04 0,0013 2,34 13,77 0,55 0,79 T2M1-30 18,47 0,0044 3,73 13,40 1,46 0,87 T2M1-40 19,76 0,0049 4,07 13,22 5,25 0,84 T2M1-50 20,38 0,0060 5,86 13,81 7,64 0,66

(36)

T2M1-75 21,95 0,0079 6,63 13,79 8,36 0,50 T2M2-10 4,29 0,0000 0,56 4,70 0,46 0,70 T2M2-20 16,48 0,0000 1,60 14,24 0,80 0,59 T2M2-30 17,26 0,0000 2,36 13,80 1,21 0,42 T2M2-40 18,32 0,0022 3,64 13,45 1,98 0,34 T2M2-50 20,20 0,0061 4,38 13,99 3,38 0,33 T2M2-75 20,67 0,0046 4,45 13,25 6,44 0,30 T2M2-100 19,35 0,0046 7,76 15,23 13,11 0,32 T2M3-10 16,36 0,0026 1,03 14,47 0,32 0,73 T2M3-20 18,02 0,0038 2,99 14,12 0,52 0,56 T2M3-30 21,49 0,0056 6,99 13,29 2,74 0,53 T2M3-40 25,40 0,0079 8,78 12,86 8,77 0,46 T2M3-50 32,40 0,0105 12,01 12,27 20,73 0,41 T2M3-75 30,62 0,0106 10,96 11,93 21,54 0,37 T4M1-10 9,73 0,0024 1,07 8,52 0,62 0,54 T4M1-20 17,17 0,0006 3,69 14,10 1,28 0,25 T4M1-30 20,71 0,0041 5,93 13,43 5,61 0,16 T4M1-40 21,13 0,0047 4,91 15,31 22,03 0,15 T4M1-50 18,82 0,0048 4,18 17,44 22,17 0,16 T4M1-75 9,76 0,0042 2,44 12,53 9,04 0,28 T4M1-100 10,64 0,0029 2,58 12,68 9,68 0,30 T4M2-10 0,55 0,0000 0,25 1,32 0,56 0,51 T4M2-20 21,70 0,0031 6,70 15,10 20,06 0,44 T4M2-30 21,60 0,0013 5,81 13,51 8,26 0,40 T4M2-40 20,58 0,0024 5,86 13,97 3,12 0,41 T4M2-50 17,28 0,0005 2,63 14,43 0,78 0,38 T4M2-75 15,91 0,0006 1,16 14,61 0,42 0,65 T4M3-10 21,24 0,0022 4,29 13,19 8,10 0,30 T4M3-20 19,33 0,0022 3,59 14,03 3,14 0,34 T4M3-30 17,98 0,0020 3,35 13,50 1,41 0,31 T4M3-40 16,09 0,0013 1,78 14,18 0,69 0,66

(37)

T4M3-50 1,10 0,0005 0,36 1,94 0,50 0,28 T4M3-75 21,77 0,0035 4,68 14,93 14,06 0,38 T4M3-100 21,09 0,0042 4,96 15,14 15,21 0,39 T6M1-10 21,76 0,0047 5,42 14,03 3,68 0,87 T6M1-20 25,29 0,0056 6,62 13,68 12,23 1,04 T6M1-30 53,70 0,0153 23,23 13,70 104,09 1,66 T6M1-40 49,00 0,0124 19,95 13,32 96,27 1,48 T6M1-50 35,21 0,0090 9,30 13,55 53,40 0,84 T6M1-75 33,41 0,0072 7,01 12,69 42,55 0,41 T6M1-100 33,49 0,0067 6,87 12,40 43,56 0,42 T6M2-10 34,44 0,0078 10,36 13,89 47,73 0,92 T6M2-20 33,55 0,0089 10,33 13,82 51,12 0,84 T6M2-30 30,32 0,0075 7,02 13,30 33,95 0,67 T6M2-40 30,82 0,0087 8,50 13,98 35,89 0,69 T6M2-50 26,42 0,0078 5,86 14,56 22,91 0,45 T6M2-75 31,24 0,0043 7,14 12,71 41,60 0,60 T6M3-10 23,13 0,0057 6,93 14,99 11,09 2,11 T6M3-20 29,05 0,0080 11,46 16,01 48,44 2,47 T6M3-30 41,16 0,0190 33,53 19,50 111,41 2,81 T6M3-40 43,18 0,0186 29,80 19,36 128,99 2,17 T6M3-50 48,94 0,0171 28,07 15,98 116,45 1,28 T6M3-75 34,56 0,0083 10,03 14,03 54,26 0,69 T7M1 opp 239,92 0,0247 171,04 13,60 77,51 3,54 T7M1-10 68,01 0,0173 38,68 13,15 163,55 1,21 T7M1-20 30,40 0,0093 11,72 5,16 63,09 0,57 T7M1-30 35,73 0,0097 9,53 13,26 39,15 0,87 T7M2 opp 157,42 0,0183 56,58 12,62 128,09 3,59 T7M2-10 57,57 0,0084 14,28 11,78 103,13 1,57 T7M2-20 44,93 0,0098 13,07 13,05 73,81 1,17 Meer 1 40,32 0,0178 19,35 15,75 76,92 1,46 Meer 2 11,94 0,0013 4,13 7,30 16,91 0,96

(38)

Gypsum content

Terrace 0 Terrace 2

Terrace 4 Terrace 6

Recently Saturated Zone Saturated Zone

Loss on Ignition Terrace 0 Depth (cm) 105 M 150 M 375 M 550 M 950 M 10 3,90188666 0,81100476 2,7308627 5,381952308 27,52255398 20 3,299764562 0,810310046 2,808191892 5,416975674 27,8937995 30 1,960868345 0,945020871 2,426378226 4,065621515 28,10347646 40 2,226336087 1,011745755 2,70506241 5,109483147 35,42926048 50 2,433758944 0,898716673 2,586699428 4,389810828 27,86439418 75 2,484048683 0,941833144 2,652012851 4,501288943 27,80765035 100 3,36994417 0,94711182 2,715708123 4,563297102 25,24339303 Depth (cm) Location 1 Location 2 Location 3

10 1,1530805 1,0509053 0,6586587 20 5,2379480 3,6458571 3,4175918 30 12,284345 7,8263180 3,6763048 40 10,993578 9,6650095 5,4358633 50 12,827937 3,2730232 1,9163461 75 9,2283830 7,0474384 6,4984390 100 13,040096 11,836232 17,875299

Depth (cm) Location 1 Location 2 Location 3 10 17,263834 0,9588127 7,2553293 20 17,242466 17,742363 16,829863 30 17,525349 18,279248 15,606428 40 17,816385 17,342307 15,909413 50 18,166700 17,631759 16,501032 75 17,102220 17,097799 11,070187 100 17,101422 7,9920995 11,033602

Depth (cm) Location 1 Location 2 opp 11,09752644 7,703727636 10 11,22823617 11,50545625 20 4,995034153 13,97761464

30 14,64876826

Depth (cm) Location 1 Location 2 Surface 13,90099678 1,655768287

(39)

Terrace 2

Terrace 4

Terrace 6

Recently Saturated Zone

Depth (cm) 105 M 150 M 375 M 550 M 950 M 10 3,960380722 0,983322303 3,252001281 5,558080905 26,38202818 20 4,272024862 0,941177709 2,588895616 5,605093049 24,36588185 30 3,322947869 0,982810652 2,477581497 4,912671101 16,81194505 40 3,587397035 0,82389211 2,20753972 4,970367073 24,43644597 50 4,678004002 1,149016796 2,328875297 5,441564235 23,94040333 75 2,553933769 1,072722969 1,889866826 6,227414594 24,1111605 100 2,19610865 1,184742824 1,849354652 7,05273386 25,10507432 Depth 105 M 150 M 375 M 550 M 950 M 10 1,734501483 1,042876171 2,881038948 5,564896307 26,54231093 20 2,234356371 0,829147311 2,427799506 5,863282424 26,17060825 30 2,043577026 0,963178095 2,513394417 4,718925251 26,97800123 40 2,144389501 1,144438552 2,757285621 6,457202204 25,65625105 50 2,425209001 1,023331106 2,327444099 6,358019548 25,70199353 75 2,398550819 1,249926628 2,493034203 6,886152182 25,05712024 100 2,49720949 1,11872336 2,469039888 6,670008685 25,08882433 Depth 105 M 150 M 375 M 550 M 950 M 10 3,058693308 1,05835939 2,121562347 4,233237913 29,19791053 20 3,056178795 1,791761054 2,526598978 5,303719121 30,09421383 30 4,2629752 1,355278094 2,174102965 5,089657483 29,90284424 40 5,678356812 1,273022138 2,064881397 5,816483703 28,65026944 50 3,811326317 1,274446016 1,911457731 4,025813938 31,86770205 75 2,638990267 1,338969071 2,030117443 5,020739503 28,14209897 100 0,899209486 0,849802372 1,729249012 4,88287798 27,58798107 105 M 150 M 375 M 550 M 950 M opp 5,232626482 2,830321542 6,010049866 12,0252127 19,58659574 10 1,142121163 1,699774866 3,594302667 9,351104405 22,57803142 20 0,689178021 1,564696647 2,624701259 5,503853589 27,47234652

(40)

Saturated Zone

Appendix 2: Modelling results

Terrace 0 In dS/m In dS/m 105 M 150 M 375 M 550 M 950 M Location 1 0,798816568 1,972386588 5,641025641 9,04651791 24,256735 Location 2 0,462133358 1,537300764 4,130906347 6,676574931 28,57854368 Depth S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 10 21,22 23,71 25,7 27,35 28,78 30,05 31,21 32,28 33,27 34,18 20 13,23 14,74 17,22 19,81 22,15 24,19 25,96 27,52 28,91 30,18 30 18,88 16,06 15,42 16,32 18,07 20,11 22,15 24,06 25,79 27,35 40 24 21,44 18,77 17,09 16,7 17,38 18,74 20,44 22,25 24,02 50 24,6 24,3 22,87 20,82 18,95 17,82 17,6 18,17 19,3 20,77 60 24,94 24,8 24,55 23,89 22,66 22,56 22,4 22,22 22,08 21,99 70 23,22 23,9 24,3 24,46 24,35 24,01 23,72 23,44 23,2 22,83 Depth S10 S12 S13 S14 S15 S16 S17 S18 S19 S20 10 35,04 35,85 36,62 37,35 38,05 38,72 39,37 39,99 40,59 41,17 20 31,32 32,37 33,34 34,25 35,1 35,9 36,66 37,39 38,09 38,76 30 28,77 30,05 31,21 32,28 33,27 34,19 35,05 35,86 36,63 37,36 40 25,68 27,22 28,63 29,92 31,1 32,18 33,18 34,11 34,98 35,8 50 22,39 24,03 25,63 27,13 28,52 29,81 30,99 32,08 33,09 34,03 60 21,97 22,03 22,15 22,34 22,58 22,87 23,19 23,54 23,93 24,33 70 22,66 22,52 22,42 22,37 22,36 22,4 22,49 22,63 22,81 23,03

(41)

Terrace 6 In dS/m In dS/m Depth S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 10 13,91 14,43 14,94 15,44 15,93 16,41 16,88 17,34 17,88 18,38 20 17,37 16,85 16,34 15,84 15,35 14,87 14,4 13,94 13,40 12,90 30 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 40 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 50 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 60 31,74 32,01 32,01 32,01 32,01 32,01 32,01 32,01 32,01 32,01 70 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28 Depth S10 S12 S13 S14 S15 S16 S17 S18 S19 S20 10 18,87 19,37 19,86 20,36 20,85 21,35 21,84 22,34 22,83 23,33 20 12,41 11,91 11,42 10,92 10,43 9,93 9,44 8,94 8,45 7,95 30 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 26,4 40 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 31,4 50 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 37,47 60 32,01 32,01 32,01 32,01 32,01 32,01 32,01 32,01 32,01 32,01 70 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28 32,28

(42)

Appendix 3: Additional MATLAB results

(43)

Appendix 4: Soil classification

Terrace 0

Profile number SP/EP/T0

Soil profile description status 2.1 Date of description 280616

Authors F.J. van Langen

Location Abandoned agricultural terrace near Embalse de Puentes

Elevation 476 meters

Coordinates 37°45’23.79’’N

1°51’21.71’’W Atmospheric climate and weather

conditions

PC Landform and topography

Major landform Position

Slope form

Slope gradient and orientation

SH MS SS 06 Land use and vegetation

Land use Crops Human influence Vegetation U -N -Parent material WE

Age of land surface SO1 - T Surface characteristics

Rock outcrops

Coarse surface fragments Erosion Surface sealing C F WG M - H Horizon boundary G

Primary constituents Silt loam

Mottling N

Carbonates ST

Gypsum MO

Readily soluble salts EX

Soil pH 7,77

Organization of soil constituents VF granular

(44)

Terrace 2

conditions

Slope form

Rock outcrops

Mottling N

Carbonates ST

Gypsum MO

Soil pH 7,90

(45)

Terrace 4

conditions

Slope form

Rock outcrops

Mottling N

Carbonates ST

Gypsum MO

Soil pH 7,77

(46)

Terrace 6

conditions

Slope form

Rock outcrops

Mottling N

Carbonates ST

Gypsum MO

Soil pH 8,61

(47)

Recently saturated zone (Terrace 7)

Profile number SP/EP/RSZ

conditions

Slope form

SH LS SS 04 Land use and vegetation

Land use Crops Human influence Vegetation U -N SX Parent material WE

Rock outcrops

Mottling N

Carbonates ST

Gypsum MO

Soil pH 8,74

(48)

Appendix 5: GIS maps

GIS map 1: spatial distribution of soil salinity expressed in electrical conductivity in !S/m of the total research area.

(49)

GIS map 2: spatial distribution of soil salinity expressed in electrical conductivity in !S/m of a part of the research area.

(50)

GIS map 3: spatial distribution of erodibility of the soil expressed in water drops created by Klaver (2016).