Spatial Gradients of Intensity and Persistence of Soil Water Repellency Under Different Forest Types in Central Mexico

SPATIAL GRADIENTS OF INTENSITY AND PERSISTENCE OF SOIL WATER REPELLENCY UNDER DIFFERENT FOREST TYPES IN CENTRAL MEXICO Nancy Alanís 1,2 , Víctor M. Hernández-Madrigal 1 , Artemi Cerdà 3 , Miriam Muñoz-Rojas 4,5 , Lorena M. Zavala 2 , Antonio Jordán 2 * 1 Instituto de Ciencias de la Tierra, Universidad Michoacana de San Nicolás de Hidalgo, Ciudad Universitaria, Morelia (Mich.), Mexico 2 Departamento de Cristalografía, Mineralogía y Química Agrícola, Universidad de Sevilla, Sevilla, Spain 3 Departament de Geografia, Universitat de València, Valencia, Spain 4 University of Western Australia, School of Plant Biology, Perth, Australia 5 Kings Park and Botanic Garden, Kings Park, Perth, Australia Received: 22 June 2015; Revised: 26 March 2016; Accepted: 16 May 2016 ABSTRACT Organic residues release hydrophobic compounds to the soil that may induce soil water repellency (WR), which may inhibit infiltration and increase runoff and soil loss rates. Although there are many studies on soil WR through the world, very few investigations have been conducted in Mexican areas. This paper studies the natural background of soil WR in soils from central Mexico under representative forest types, analyzing the spatial distribution of soil WR in relation with tree canopy, vegetation cover and main soil chemical (pH, CaCO 3 , organic C content and exchangeable cations) and physical properties (texture). The water drop penetration time and the ethanol tests were used to assess persistence and intensity of soil WR, respectively. Although soil WR was not related with soil properties, it decreased strongly from soil below the canopy of conifers to soil below oaks. When different types of vegetation cover were considered, the proportion of water-repellent soil increased following the sequence: bare soil<shrubs and herbaceous plants <shrubs <trees from fir, fir-pine-oak and pine-oak forest. We found an inverse relation with distance to the tree trunks, contributing to create a patchy pattern of soil WR, with soils under the canopy of conifers showing the most severe WR levels. The spatial distribution of soil WR is also conditioned by microclimatic gradients, as persistence and intensity of soil WR were usually lower in shaded areas (upslope transects from the tree trunks), where soil moisture content is expected to be higher on average through the year. Copyright © 2016 John Wiley & Sons, Ltd. key words: coniferous forest; hydrophobicity; Mexican soils; pine forest; soil hydrology INTRODUCTION Soil water repellency (WR) is a result of the low affinity between soil particles and water. Infiltration into waterrepellent soils may be delayed or inhibited during periods of time ranging from seconds to hours, days or weeks (Doerr et al., 2000). Some impacts of soil WR are increased aggregate stability (Granged et al., 2011a; Mataix-Solera et al., 2011; Jordán et al., 2014), enhanced runoff rates and erosion risk in forest (Doerr et al., 2000; Shakesby et al., 2000; Jordán et al., 2008; Jordán et al., 2015) or agricultural soils (González-Peñaloza et al., 2012; García-Moreno et al., 2013), the development of preferential flow paths (de Rooij, 2000; Jordán et al., 2009; Zavala et al., 2009a; Granged et al., 2011b) and accelerated leaching of soil nutrients and pollutants through preferential flow paths (Blackwell, 2000; Li et al., 2007; Müller et al., 2014). Soil WR is generally induced by plant residues and substances excreted by roots, microorganisms and fungi. Many researchers have observed that soil WR is mostly associated with certain plant species with tissues rich in resins and waxes, as pines or eucalypts (Arcenegui et al., 2008; Hubbert et al., 2006; Martínez-Zavala & Jordán-López, 2009; Mataix-Solera & Doerr, 2004; Mataix-Solera et al., 2007; Rodriguez-Alleres et al., 2007). It has also been reported that soil properties may influence the development of soil WR (Doerr et al., 2000), as soil microbiology, texture, moisture content, acidity and mineralogy (Dekker et al., 2000; Mataix-Solera et al., 2013; Mataix-Solera et al., 2014; Jiménez-Pinilla et al., 2016). Other factors, like fire, may have no impact, destroy or enhance soil WR (Doerr et al., 2000; Jordán et al., 2013; Pereira et al., 2015). Consequently, different authors have investigated the natural soil WR background in different areas as USA (Pierson et al., 2008; Doerr et al., 2009), Iran (Mirbabaei et al., 2013), Mexico (Jordán et al., 2009), Spain (Rodríguez-Alleres et al., 2012; Zavala et al., 2014), South Africa (Scott, 2000) or Australia (Blackwell, 2000). Although highland forests in central Mexico have played a key role in local economy over centuries, economical and management changes have led to serious deforestation processes (Álvarez-Icaza et al., 1993; Works & Hadley, 2004). In the Angangueo area, some of the most important disturbances include road-building, logging timber for stabilization of mine shafts, clear-cuts because of commercial logging or selective cutting (López-García & Alcántara- Correspondence to: Antonio Jordán, Departamento de Cristalografía, Mineralogía Química Agrícola Facultad de Química (Universidad de Sevilla), C/Profesor García González, 1, 41012 Sevilla, Spain. E-mail: [email protected] Copyright © 2016 John Wiley & Sons, Ltd. land degradation & development Land Degrad. Develop. 28: 317–327 (2017) Published online 17 June 2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ldr.2544 Ayala, 2012). Studies carried out by Jordán et al. (2009) and Jordán et al. (2011) reported relatively high values of soil WR in forest soils from Mexican volcanic highlands, particularly beneath the canopies of fir and pines. In all these cases, soil WR can increase soil erosion risk, especially at the beginning of the rainy season (Jordán et al., 2009; Gabarrón-Galeote et al., 2013). Because of these impacts, information on the natural severity of soil WR is necessary for adequate soil planning and management. Despite the general influence of climate gradients in soil hydrological and geomorphological processes, local factors may be extremely important (Cerdà, 1998; Imeson et al., 1998). Research of soil WR background presents some gaps in areas where it would help to understand hydrological and erosive soil processes. Only a few studies focusing soil WR have been carried out in Mexico. Some of these studies deal with hydrocarbon contaminated soils in the southern Gulf of Mexico region (Adams et al., 2008; Guzmán-Osorio and Adams, 2015). Jordán et al. (2009) studied the hydrological response of different vegetation types (fir, pine and oak forest and shrublands) in hydrophobic and wettable volcanic soils of central Mexico. They observed that soil WR contributed to fast ponding and runoff generation during simulated rainfall experiments. In the same area, Jordán et al. (2011) investigated the effects of fire severity on WR and aggregate stability. In a laboratory experiment, Zavala et al. (2010) studied the effect of burning temperature on WR and aggregate stability in Spanish, Mexican and Australian soils. However, more studies are necessary to increase the number of available data on soil WR and understand its implications on the hydrological and geomorphological soil processes in a region characterized by high ecosystem diversity. For this study, we selected the San Pedro area, near the village of Angangueo (Michoacan, Mexico), in the Mexican Neo-Volcanic Axis Range. Forest types in this area include tree species such as firs, cedars, pines and oaks, which induce different levels of soil WR (Jordán et al., 2009; Jordán et al., 2011). The objectives of this study are (i) to study the intensity and persistence of WR in soils under specific tree species (firs, cedars, pines and oaks); (ii) the relation between soil WR distance upslope (shadowed area) and downslope (sunny area) of the source of organic residues (tree trunk); (iii) the relation between soil WR and plant cover (tree, shrub and herbaceous cover) in different forest types and (iv) the relation between soil WR and physical and chemical soil properties. MATERIAL AND METHODS STUDY AREA The study area is located in the south-facing slope of the San Pedro River catchment (19°37′31″N, 100°16′26″W), near Angangueo (Mich., Mexico). Elevation of the study area ranges from 2600 and 3400 masl. The lithological substrate is dominated by Miocene alkaline lavas (andesites). According to the Chincua weather station (19° 45′28·62″ N, 100° 17′5·42″W, 2415 masl, approximately 14km north of the study area), climate is characterized by a dry cool winter and a temperate rainy summer. Mean monthly temperature varies between 11·9 (January) and 17 ·0°C (May), with average annual temperature 14·8°C. Mean monthly rainfall varies between 6 (March) and 211mm (July), with mean annual rainfall 913mm. Main species in the area are sacred fir(Abies religiosa), smooth-bark Mexican pine (Pinus pseudostrobus), oaks (Quercus laurina), shrubs and herbaceous plants. Fir forests also include willows (Salix paradoxa) and white cedar (Cupressus lusitanica). These forests distribute on Andosols in the upper part of slopes, between 3·100 and 3 ·500 masl, forming dense and diverse communities with tree height often reaching 30m (Ramírez, 2001; Rzedowski & Rzedowski, 2001). Cedars in the area come from plantations carried out in the early XXth century (Madrigal, 1994; Ramírez, 2001; Rzedowski & Rzedowski, 2001). Pine-oak forests distribute mostly in the western slopes of the study area, below 3000 masl. These forests are formed by A. religiosa,Alnus acuminata,Alnus jorullensis,Arbutus xalapensis,Clethra mexicana,Pinus leiophylla,Pinus michoacana,Pinus montezumae,P. pseudostrobus,Quercus crassipes,Q. laurina and Quercus rugosa (Madrigal, 1994; Rzedowski & Rzedowski, 2001). Shrubs dominated by sunflowers (Baccharis conferta) and conifers (Juniperus monticola) occupy medium (2800–3300) and upper (3100– 3500 masl) part of slopes, respectively. These shrubs also include Acaena elongata,Senecio cinerarioides,Senecio salignus,Senecio angulifolius and Symphoricarpus microphyllus (Rzedowski & Rzedowski, 2001). EXPERIMENTAL DESIGN The experimental work was carried out during December (2014) and January (2015), part of the dry season. In order to study the influence of tree canopy on soil WR, a set of 80 points was selected in the study area using UTM randomly generated coordinates (20 points for each tree type considered: firs, pines, cedars and oaks). Slope in the selected points ranged between 43·8 and 63·3%, 55·5% on average. At each one of these points, we selected the closest individual of each of the four species considered (firs, pines, cedars and oaks). In order to study differences in the intensity and persistence of soil WR between points at different distance from the source of residues, these parameters were determined at 10 points upslope (U) and 10 points downslope (D) from the trunk (100 cm from each other) for each tree. When canopies from different trees overlapped, the point was discarded and the next closest individual was selected. In order to study the relation between soil WR and plant cover, 10 east-to-west oriented transects (15m long) were established at randomly selected points under fir, fir-pine- oak and pine-oak forest types. At each transect, presence (1) or absence (0) of plant cover (T, tree; S, shrub; H, 318 N. ALANÍS ET AL. Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) herbaceous cover) was recorded every 100cm. The final number of observations was 480 (10 transects ×3 forest types×16 points). Tree (T), shrub (S) and herbaceous cover (H) were expressed as the percentage of observations (points) for each combination of possible covers. We calculated the proportion of bare soil as the percentage of points where no plant cover was observed. Rock fragments (mineral particles larger than 2mm), rocky outcrops and other elements (biological crusts, for example) were included in the “bare soil”class because of their negligible occurrence as compared to bare soil. They were also present in a very small proportion below the canopy of plants. PHYSIC-CHEMICAL CHARACTERIZATION OF SOILS Previously to the assessment of soil WR, soil samples (0– 5cm depth) were collected 200cm upslope and downslope from each individual tree considered. Soil samples were transported in plastic bags to the laboratory for soil analysis, dried on paper tries at laboratory room temperature (25 °C) to constant weight and sieved (2mm) to eliminate coarse soil particles. Soil acidity (pH) was determined in 1:2·5 soil: water extracts. Carbonate content was determined in soil samples with pH>6·0 using the Bernard’s calcimeter. Organic carbon (OC) content was determined by oxidation with acid-dichromate potassium and titration of dichromate excess with ferrous sulfate (Walkley & Black, 1934). Exchangeable cations were determined by the ammonium acetate method (Thomas, 1982) and cation exchange capacity (CEC) was determined by the BaCl 2 test (Rhoades, 1982). Previously to textural analysis, soil subsamples (about 100g) were pre-treated with H 2 O 2 (6%) to remove organic matter and soluble salts and, later, with HCl (35%) to remove CaCO 3 if present. These pre-treated samples were dried in the oven to obtain the initial weight, dispersed with a (NaPO 3 ) 6 solution (5%), and mechanically shaken. Soil material was then sieved to separate the sand (0·05–2mm) fraction. The clay (<0· 002mm) fraction was determined by the Bouyoucos method (Bouyoucos, 1962). SOIL WATER REPELLENCY Persistence and intensity of soil WR were assessed under field conditions on the soil surface, after gently removing the litter manually. Intensity of soil WR was determined using the ethanol percentage test (EPT). Drops (50μL) of decreasing ethanol concentrations (increasing surface tensions) were applied onto the soil surface using a micropipet until one of the drops infiltrated within 5s after application. This allows the classification of the soil surface into a surface tension category between two consecutive ethanol concentrations. EPT classes were classified as in Doerr (1998): very wettable (1, 0·0% ethanol), wettable (2, 3·0%), slightly water repellent (3, 5· 0%), moderately water repellent (4, 8·5%), strongly water repellent (5, 13·0%), very strongly water repellent (6, 24·0%) and extremely water repellent (7, 36·0%). Persistence of soil WR was determined using the water drop penetration time (WDPT) test (Wessel, 1988). At each point, 10 drops (0·5μL) of distilled water were applied in a circular area (10cm in diameter) of the soil surface using a micro-pipet from a height of approximately 5mm to avoid excess kinetic energy affecting soil-droplet interactions, and time for complete infiltration was recorded. The median WDPT was considered representative for each case and soil was classified as wettable (1, WDPT ≤5s), slightly water repellent (2, 5s<WDPT ≤60s), strongly water repellent (3, 60s<WDPT≤600 s), severely water repellent (4, 600s<WDPT≤1h) and extremely water repellent (5, WDPT>1h). DATA ANALYSIS The normal distribution of data and homogeneity of variances was tested using the Shapiro–Wilk’s and Levene’s test, respectively. When both null-hypotheses were accepted, parametric statistics (mean, standard deviation) and tests (ANOVA) were used. Otherwise, non-parametric statistics (median, range and Spearman rank correlation coefficient) and tests (Kruskal–Wallis and Mann–Whitney Utests) were used. The Spearman rank correlation coefficient (r s ) was used to study the relation between EPT and WDPT data and among them and the rest of soil variables. A principal component analysis (PCA) was performed using intensity of WR data (transformed as ln(EPT)) and soil chemical and physical parameters (pH, OC content, exchangeable Ca 2+ ,Mg 2+ ,K + and Na + , CEC, and sand and clay contents). All computations were performed using SPSS v.23 (IBM Corp., 2015). RESULTS Soil characterization The ANOVA test did not found significant differences between soil chemical and physical properties in U- and D-points. A summary of soil characteristics is shown in Table I. Studied soils were slightly to strongly acidic, with mean pH values ranging from 5·4±0·4 (soils below oaks) to 6·2±0·4 (below cedars). Mean soil CaCO 3 content per tree species and orientation varied between 0·0 ±0·0 (oaks and pines) and 0·07±0·30% (cedars); it was 0·00% in all cases except in three soil plots under the canopy of cedars and three below firs with pH6 ·8–7·0 (0·01–1·44% CaCO 3 ). Organic C content varied between 7 ·76±0·89 (cedars) and 11·31±0·43% (oaks), 9·39± 1·52% on average. The content of exchangeable cations was relatively high for Ca 2+ (6·17±3·07 cmol (+) kg 1 , on average) and Mg 2+ (2·21±1·24 cmol (+) kg 1 ). Mean values for K + and Na + were 0·45±0· 26 and 0·09±0· 06 cmol (+) kg 1 . Generally, soil CEC decreased following the sequence oaks>pines>firs>cedars, with mean values ranging 319 SPATIAL GRADIENTS OF SOIL WATER REPELLENCY IN MEXICAN FORESTS Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) between 14·43±7·98 (cedars) and 38·23±19· 82 cmol (+) kg 1 (oaks). Soil texture was silt loam, with average sand and clay contents 31· 1±6·3 and 12·6±1 ·9%, respectively. Soil Water Repellency Under Different Tree Species Intensity of soil water repellency EPT classes varied between 1–2 and 7 in U- and D-points below firs and cedars. We observed extreme intensity (EPT=7) below pines only in D-points. In the case of oaks, EPT classes only varied between 1 and 3 (U-points) or 1 and 5 (D-points). Generally, EPT decreased with distance from the trunk (Table II). In U-points below firs, for example, EPT decreased from 7 (1m from the trunk) to 3(10m from the trunk). Although soil below the canopy of firs, pines and cedars was usually very strongly or extremely waterrepellent in a 2-m radius he trunk, EPT classes below the canopy of oaks varied between 3 and 4 near the trunk, quickly decreasing to 1 or 2 with distance. Median intensities of soil WR in U- and D-points below the canopy of studied trees were not significantly different, yet we found some differences between them at different distances from the tree trunks. EPT decreased one class at 1-m position from U- to D-points (p<0·001), and increased one class at 6 and 7m (p<0·001 in both cases) below firs. Median EPTs below pines, cedars and oaks increased in one class at least for each distance from the trunk in D-points (p<0 ·001) except in one case. In some soil plots below oaks, this only meant changing from wettable to very wettable. Persistence of soil water repellency WDPT classes varied between 1 and 7. The proportion of points with extreme persistence of soil WR (WDPT=7) varied between 35 (pines, U-points) and 72% (firs, D-points) for soil below the canopy of firs, cedars and pines. Below oaks, the proportion of extremely persistent WR was only 3 (U) and 6% (D), and very wettable or wettable soil points dominated (71%, U, and 54%, D). As seen for EPT values, WDPT significantly decreased with distance from the trunk of trees, varying between 4 and 7 (firs), 3 and 7 (pines and cedars), and 1 and 3 (oaks) (Table III). Generally, the influence of the relative position (U- and D-points) was not important when all cases were considered together. However, we found 1-class increments between U and D equivalent positions in 3 points below the canopy of firs, 3 points below pines and 5 points below cedars and oaks (p<0·001 in all cases). In the latter case, this only meant jumping from wettable to slightly water repellent. Wettable soil points were not observed below the canopy of firs, and only in some cases below pines and cedars (between 7 and 10m from the trunk). On the other hand, most of soil points below oaks were classified as wettable (WDPT=1; 41·5 and 16·5%, U- and D-points, respectively) or slightly water repellent (WDPT=2; 28·5 and 36·0%, U- and D-points, respectively). Table I. Soil characterization (mean ± standard deviation) under different tree species: pH; CaCO 3 content (%); organic C content; exchangeable cations (Ca 2+ ,Mg 2+ ,K + and Na + ); cation exchange capacity (CEC) and sand and clay contents. N= 40 in all cases Species pH CaCO 3 Organic C Ca 2+ Mg 2+ K + Na + CEC Sand Clay % % cmol(+) kg 1 cmol(+) kg 1 cmol(+) kg 1 cmol(+) kg 1 cmol(+) kg 1 %% Cedar 6 · 2 ± 0 · 4 0 · 07 ± 0 · 30 7 · 76 ± 0 · 89 5 · 82 ± 3 · 08 2 · 08 ± 1 · 28 0 · 42 ± 0 · 26 0 · 09 ± 0 · 06 14 · 43 ± 7 · 98 29 · 3 ± 6 · 4 12 · 4 ± 2 · 1 Fir 6 · 1 ± 0 · 4 0 · 05 ± 0 · 17 8 · 47 ± 0 · 69 7 · 08 ± 3 · 20 2 · 58 ± 1 · 30 0 · 51 ± 0 · 27 0 · 11 ± 0 · 061 24 · 15 ± 11 · 70 29 · 3 ± 7 · 1 12 · 7 ± 2 · 0 Oak 5 · 4 ± 0 · 4 0 · 00 ± 0 · 00 11 · 31 ± 0 · 43 5 · 88 ± 2 · 91 2 · 13 ± 1 · 23 0 · 43 ± 0 · 26 0 · 09 ± 0 · 05 38 · 23 ± 19 · 82 32 · 2 ± 5 · 2 12 · 4 ± 1 · 9 Pine 5 · 7 ± 0 · 3 0 · 00 ± 0 · 00 10 · 02 ± 0 · 36 5 · 92 ± 3 · 02 2 · 055 ± 1 · 10 0 · 42 ± 0 · 24 0 · 09 ± 0 · 05 30 · 33 ± 15 · 58 33 · 7 ± 5 · 4 12 · 9 ± 1 · 6 All 5 · 9 ± 0 · 5 0 · 03 ± 0 · 17 9 · 39 ± 1 · 52 6 · 17 ± 3 · 07 2 · 21 ± 1 · 24 0 · 45 ± 0 · 26 0 · 09 ± 0 · 06 26 · 79 ± 16 · 78 31 · 1 ± 6 · 3 12 · 6 ± 1 · 9 320 N. ALANÍS ET AL. Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) Relation Between Intensity and Persistence of Soil Water Repellency Spearman’s rank correlation coefficients between EPT and WDPT were significant (p<0·0001) and positive for soil points below firs (r s =0·7240), pines (r s =0·6832) and cedars (r s =0·7246). In contrast, correlation for soil points below oaks was weak (r s =0·4607; p<0·0001). EPT from soils below pines and firs was not significantly different (p=0·9719). In contrast, EPT classes below cedars were significantly lower than below firs (p=0·0002) and pines (p=0·0001). Soils below oaks showed lower EPT values than soils below the rest of species (p= 0·0000). Concerning WDPT, we found significant differences between median WDPT values in all cases except for soils below cedars and pines (p=0·1780). Soils below firs showed higher WDPT values than soils below pines and cedars. As in the previous case, soils below oaks showed lower WDPT values than soils below the rest of species. Table II. Intensity of soil water repellency (EPT classes: median and range between parentheses) per tree species, orientation (U: upslope; D: downslope) and distance from trunk. EPT classes: 1: very wettable; 2: wettable; 3: slightly water repellent; 4: moderately water repellent; 5: strongly water repellent; 6: very strongly water repellent; 7: extremely water repellent. S–W, p:p-value of the Shapiro–Wilk test for normality of EPT data distribution below each species and at different orientations; K–W, p:p-value for the Kruskal–Wallis test of differences between EPT values at different distances from the trunk below each species and at different orientations; M–W, p:p-value for the Mann–Whitney U test for comparison of median EPTs below upslope and downslope points below tree species. For each species and distance from the trunk, significant differences (p>0 · 05) between the intensity of soil water repellency of upslope and downslope areas are marked with an asterisk (*). The number of samples for each tree species, orientation and distance is 20 Distance (m) Fir-U Fir-D Pine-U Pine-D Cedar-U Cedar-D Oak-U Oak-D 1 7 (6, 7) 6 (5, 7) * 5· 5 (5, 6) 7 (5, 7) * 6 (5, 7) 7 (5, 7) 3 (2, 3) 4 (2, 5) * 2 6 (5, 7) 6 (5, 7) 5 (4, 6) 6 (4, 7) * 6 (4, 7) 7 (5, 7) * 2 (1, 3) 3 (1, 4) * 3 6 (5, 6) 6 (5, 6) 4· 5 (4, 5) 6 (4, 7) * 5 (4, 6) 6 (4, 7) * 2 (1, 3) 3 (1, 5) * 4 5 (4, 6) 5 (4, 6) 4· 5 (3, 5) 5 (3, 7) * 4 (3, 5) 5 (4, 7) * 1 (1, 2) 2 (1, 5) * 5 5 (4, 5) 5 (4, 6) 5 (4, 5) 6 (5, 7) * 4 (3, 4) 5 (3, 6) * 1 (1, 2) 2 (1, 4) * 6 4 (3, 5) 5 (4, 6) * 4 (3, 5) 5 (3, 7) * 3 (2, 4) 4 (3, 6) * 1 (1, 2) 3 (1, 4) * 7 3 (3, 4) 4 (4, 4) * 3 (3, 4) 5 (3, 6) * 2 (2, 3) 3 (2, 4) * 1 (1, 1) 2 (1, 3) * 8 4 (3, 4) 4 (3, 4) 3 (3, 4) 4 (3, 6) * 2 (2, 3) 3· 5 (2, 4) * 1 (1, 1) 2 (1, 3) * 9 3 (3, 3) 3 (3, 4) * 3 (3, 3) 4 (3, 5) * 2 (2, 3) 3 (2, 5) * 1 (1, 1) 2 (1, 3) * 10 3 (2, 3) 2 (1, 3) 2 (2, 3) 4 (2, 5) * 2· 5 (2, 3) 3 · 5 (2, 5) * 1 (1, 1) 2 (1, 3) * All cases 4 (2, 7) 5 (1, 7) 4 (2, 6) 5 (2, 7) 3 (2, 7) 4 (2, 7) 1 (1, 3) 2 (1, 5) S–W, p<0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 K–W, p<0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 M–W, p>0·05 >0·05 >0·05 >0·05 Table III. Persistence of soil water repellency (WDPT classes: median and range between parentheses) per tree species, orientation (U: upslope; D: downslope) and distance from trunk. WDTP classes: 1: wettable (0–5s); 2: slightly water repellent (5–60s); 3: strongly water repellent (60–600 s); 4: severely water repellent (600 s–1 h); 5: extremely water repellent (1–3 h); 6: extremely water repellent (3–6 h); 7: extremely water repellent (>6 h). S–W, p:p-value of the Shapiro–Wilk test for normality of WDPT data distribution below each species and at different orientations; K–W, p:p-value for the Kruskal–Wallis test of differences between WDPT values at different distances from the trunk below each species and at different orientations; M–W, p:p-value for the Mann–Whitney Utest for comparison of median WDPTs below upslope and downslope points below tree species. For each species and distance from the trunk, significant differences (p>0 · 05) between the intensity of soil water repellency of upslope and downslope areas are marked with an asterisk (*). The number of samples for each tree species, orientation and distance is 20 Distance (m) Fir-U Fir-D Pine-U Pine-D Cedar-U Cedar-D Oak-U Oak-D 1 7 (5, 7) 7 (5, 7) 6 (4, 7) 6 (4, 7) 7 (4, 7) 7 (4, 7) 2 (1, 5) 3 (2, 5) * 2 7 (5, 7) 7 (5, 7) 5 (3, 7) 5 · 5 (3, 7) 6 (4, 7) 6 · 5 (5, 7) 3 (1, 5) 3 (2, 5) 3 6 (4, 7) 6 · 5 (4, 7) 5 (3, 6) 5 (3, 6) 6 (4, 7) 7 (4, 7) 2 · 5 (1, 5) 3 (1, 6) 4 6 (4, 7) 6 · 5 (5, 7) 4 (3, 6) 5 (3, 7) 5 (3, 7) 5 (3, 7) 2 (1, 4) 3 (1, 5) 5 5 · 5 (3, 7) 6 (4, 7) 4 (3, 6) 5 (3, 7) 4 (2, 6) 5 (2, 7) 2 (1, 4) 3 (1, 4) 6 5 (2, 7) 5 · 5 (2, 7) 4 (2, 6) 4 (3, 6) 3 · 5 (2, 5) 4 (3, 6) 2 (1, 4) 2 (1, 5) 7 4 (2, 5) 5 (2, 6) 4 (2, 5) 4 · 5 (2, 6) 3· 5 (1, 5) 4 (1, 5) 2 (1, 3) 2 (1, 4) 8 4 (2, 6) 5 (3, 6) 4 (2, 5) 4 (2, 6) * 3 (1, 5) 4 (2, 5) 1 (1, 2) 2 (1, 3) 9 3 (2, 5) 3 (2, 6) 3 (2, 4) 3 (2, 5) 3 (1, 5) 4 (1, 6) 1 (1, 2) 2 (1, 3) * 10 4 (2, 5) 4 (2, 6) 2 (1, 4) 3 (1, 5) 3 (1, 5) 4 (2, 6) 1 (1, 2) 2 (1, 3) * All cases 5 (2, 7) 6 (2, 7) 4 (1, 7) 5 (1, 7) 4 (1, 7) 5 (1, 7) 2 (1, 5) 2 (1, 6) S–W, p<0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0· 001 <0 · 001 K–W, p<0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0 · 001 <0·001 <0 · 001 M–W, p>0·05 >0·05 >0·05 >0·05 321SPATIAL GRADIENTS OF SOIL WATER REPELLENCY IN MEXICAN FORESTS Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) Relation Between Intensity of Soil Water Repellency and Plant Cover On average, tree cover varied between 75 ·6% in fir-pine- oak forest and 92·5% in fir forest (Table IV). On the other hand, shrub cover was higher in fir-pine-oak forest (83·8%) and slightly lower under fir and pine-oak forests (78·8 and 78·1%, respectively). Herbaceous plant cover varied in a narrow range between 46·9 (fir and fir-pine- oak forests) and 50·6% (pine-oak forest). The proportion of bare soil was below 2·5% in most cases (only 11 bare soil points were observed for 480 observations). Median EPT value from soil points below the canopy of trees (median ETP=5, strongly water repellent) was significantly higher (Mann–Whitney Utest p=0·0000) than below shrubs or shrubs and herbaceous plants (median EPT=2, wettable). Generally, soil WR was more intense in tree-covered points from fir and fir-pine-oak forests than from pine-oak forest (Figure 1). For both forest types, shrub-covered points (S and S +H) were very wettable to slightly water repellent. No extreme soil water-repellent points were observed in pine-oak forest, with very wettable to very strongly water repellent conditions below trees (T, T+S and T+S+H points) and very wettable to slightly water repellent below shrubs (S and S+H). In all cases, bare soil showed wettable conditions (EPT values 1 and 2). Relation Between Intensity of Soil Water Repellency and Soil Properties The intensity of soil WR (EPT) did not show significant correlations with most soil variables, except for OC content (r s =0·578, p≤0·01), pH (r s =0·186, p≤0·05) and CEC (r s =0·294, p≤0· 01). These coefficients were significant and negative, but, generally, indicated moderate (EPT/OC) or weak correlations (EPT/pH and EPT/CEC). For soil points below different tree species, the maximum Spearman correlation coefficient observed was 0·623 (p≤0·01) below firs, while in other cases r s was always close to zero. The PCA divided the soil samples in three groups (Figure 2). Group A includes soil points below the canopy of firs and cedars, without any clear separation between them. Group B includes soil points below pines and group C soil points below the canopy of oaks, with group B in an intermediate position between groups A and C. Regarding the ordination of soil variables, factor 1 shows a high correlation with CEC and exchangeable cations, while factor 2 corresponds to soil acidity (pH), intensity of soil WR and organic C content. DISCUSSION Different severities of intensity and persistence of soil WR have been observed in soils under different types of forests and tree species in the study area. Although strong relations between soil WR and soil properties have been reported in many areas (Mataix-Solera et al., 2008; Zavala et al., 2009b; Mataix-Solera et al., 2013; Mataix-Solera et al., 2014; Zavala et al., 2014), no relevant correlations have been observed in this research at the working scale. On the other hand, the spatial distribution of soil WR in the study area is strongly conditioned by vegetation, in agreement with results reported by other authors, who observed that, in areas with homogeneous soil properties, vegetation is the main factor controlling soil WR (Doerr et al., 1996; Cerdà et al., 1998; Buczko et al., 2005; Cerdà & Doerr, 2007; Jordán et al., 2008; Jordán et al., 2009; Bodí et al., 2011; Granged et al., 2011b; Bodí et al., 2012; Zavala et al., 2014; Jiménez-Pinilla et al., 2015). Soils below fir and fir-pine-oak forest were extremely water-repellent under the canopy of trees (T, T+S and T +S+H plots). Although extreme values were not observed in tree-covered soils under pine-oak forest, T, T+S and T +S+H also showed the highest intensity of soil WR (very strong WR). Generally, the proportion of water-repellent soil points increased following the sequence bare soil <shrubs and herbaceous plants (S+H)<shrubs (S) <trees (T, T+S and T+S+H). This observation is in agreement with most of previous research, which has reported a limited contribution of herbaceous plants to soil WR (Zavala et al., 2009a; Schnabel et al., 2013; Zavala et al., 2014). Although differences between OC content were found (Kruskal–Wallis p=0·000), these are relatively small. The Spearman correlation coefficient between OC content and WR suggests that there is no direct relation between both properties in the study area. Therefore, each type of tree species, vegetation and associated microbiota may contribute to the development of different hydrophobicity severities. The highest severity of soil WR has been observed below conifers (firs, cedars and pines), whereas soils below oaks have been found to be much more wettable. These results are in agreement with findings from other authors, who have reported that conifers are much more prone than oaks to induce soil WR, because of the chemical composition of their tissues, which include resins, waxes and other substances able to cause hydrophobicity in soils (Conde et al., 1998; Ito et al., 2002; Mataix-Solera & Doerr, 2004; Lozano et al., 2013; Mao et al., 2015). Different authors have Table IV. Fir, pine, cedar, oak, total tree, shrub and herbaceous cover (%) for fir: fir-pine-oak and pine-oak forests in the study area. *Tree cover is calculated as the sum of all points below the canopy of one or more trees (firs, pines, cedars and/or oaks) Vegetation type Fir Pine Cedar Oak Total tree* Shrubs Herbaceous plants Bare soil Fir 87 · 5 0 · 0 31 · 9 0 · 0 92 · 5 78 · 8 46 · 9 1 · 9 Fir-pine-oak 51 · 9 34 · 4 0 · 0 21 · 3 75 · 6 83 · 8 46 · 9 2 · 5 Pine-oak 0 · 0 64 · 4 0 · 0 42 · 5 80 · 6 78 · 1 50 · 6 2 · 5 322 N. ALANÍS ET AL. Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) reported soil WR in oaklands from Mediterranean areas (Cerdà et al., 1998; Mataix-Solera et al., 2007; Jordán et al., 2008; Zavala et al., 2009b; Zavala et al., 2014). Although our results show that severity of soil WR below oaks is lower than in soils below conifers, as shown by Zavala et al. (2009b) and Zavala et al. (2014), preeminence Figure 1. Number of observations of intensity of soil water repellency classes (EPT) from fir, fir-pine-oak and pine-oak forest plots for different cover types, bare soil (Bare), shrub cover (S), shrub and herbaceous cover (S + H), tree cover (T), tree and shrub cover (T + S) and tree, shrub and herbaceous cover (T + S + H). N= 160 for each forest type. No observations below tree and herbaceous cover (T + H) or below herbaceous cover (H) only were recorded. Figure 2. Principal component analysis of intensity of soil WR (LN(EPT)) and soil chemical and physical variables, pH, organic C content (OC), exchangeable cations (Ca 2+ ,Mg 2+ ,K + and Na + ), cation exchange capacity (CEC) and sand and clay contents for soil plots under cedars, firs, oaks and pines. (A) cases; (B) variables. 323SPATIAL GRADIENTS OF SOIL WATER REPELLENCY IN MEXICAN FORESTS Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) of conifers versus oaks, however, is not always clear. Jiménez-Morillo et al. (2014) studied soil WR under different vegetation types in an area in southwestern Spain and found more severe WR in soils below oaks than below pines or other species. They attributed this to higher organic matter contents and a high degree of evolution of organic matter, mainly in organic films coating the finer particles. Recurrent wildfires in the region may contribute also to the development of hydrophobic soil conditions (Ramírez Ramírez et al., 2005; Garduño Bernal, 2011; Jordán et al., 2011). Tree cover conditioned the spatial variation of soil WR. Soil WR decreased with increasing distance from individual trees. Although results between firs, cedars and pines were different for intensity (EPT) and persistence (WDPT) of soil WR, both properties decreased with distance from the trunk and showed strong correlations. In the case of oaks, the severity of soil WR was limited but also decreased with distance (according to median values, intensity of WR decreased from very wettable/wettable to slightly/moderately water repellent and persistence decreased from wettable/slightly water repellent to slightly/strongly water repellent). The scattered disposition of trees contributes to a patchy pattern of soil WR, which may contribute to form a runoff generation/infiltration pattern. Very few authors have reported data about the spatial distribution of soil WR regarding the relative distance to tree trunks or canopy boundaries. In burned soils under junipers and pines, Madsen et al. (2010) found a correlation between the spatial distribution of soil WR and the canopy boundary. They observed that about 66% of the canopy region was water repellent, and persistence of soil WR decreased with distance from the canopy boundary, with wettable conditions in the inter-canopy areas (Madsen et al., 2008; Madsen et al., 2010). Our results show a similar decreasing trend with distance, but, in many cases, hydrophobic conditions were observed at relatively long distances from the tree canopy. Keizer et al. (2005) studied the variation of soil WR with distance from eucalypt trees, and found that distance only is not enough to explain hydrophobicity patterns. They concluded that local variability in topsoil WR may be caused by differences in tree characteristics and soil properties related to potential sources of hydrophobic compounds. Our results show significant differences between the intensity of soil WR from upslope and downslope points at all distances from the trunk below the canopy of pines and oaks, most of points below cedars (2–10m) and some points below firs (1, 6, 7 and 9 m). In most cases, these differences were only one EPT class. These differences may be explained by different soil microclimate conditions. Although soil WR is known to vary irregularly during drying (González-Peñaloza et al., 2013), it is generally destroyed above certain critical soil moisture content (Doerr & Thomas, 2000; Dekker et al., 2001; Zavala et al., 2010). Plant cover and shading may affect soil microclimate both in the canopy and intercanopy regions through interception of sunlight, and this effect is much more marked in the case of woody plants (Lebron et al., 2007; Madsen et al., 2008). Soils from upslope transects, preferentially shaded by trees in a southfacing slope in the study area, are expected to show relatively higher moisture contents through the year and, consequently, relatively lower severity of soil WR is expected. As soil WR is induced by organic compounds released by plant residues and soil microbiota (Doerr et al., 2000), it may be suggested that hydrophobicity is induced not only below the canopy of trees, but also in neighbor inter-canopy areas where litter and hydrophobic soil particles are redistributed by wind or surface water flow. It can be concluded that, in the area, soil WR distributes following a heterogeneous spatial pattern conditioned by canopy and inter-canopy areas. Especially in soils under oaks, intercanopy areas are wettable or show subcritical WR. CONCLUSIONS The spatial distribution of soil WR in the San Pedro River catchment (Angangueo, Mich., Mexico) is strongly conditioned by vegetation. Generally, soils below conifers showed the strongest WR. The highest intensity and persistence of soil WR were observed below firs/pines and firs, respectively. In both cases, soils below oaks showed the weakest values. Persistence and intensity of soil WR inside the 1-m radius area near the tree trunk of coniferous species were very strong to extreme and decreased progressively with distance, with wettable inter-canopy areas at 10-m distance. Although soils under oaks showed a wide range of values, soil WR was generally weaker, with subcritical WR or wettable conditions at shorter distances. It can be suggested that the position of trees in the study area contributes to a patchy pattern of soil WR, which, in turn, may contribute to a patchy distribution of runoff and infiltration areas. The irregularity of WR at the soil surface may have significant influence on soil hydrological and geomorphologic processes. For all studied forest types (fir, fir-pine-oak and pine-oak forest), the intensity of soil WR increased according to the sequence bare soil (very wettable to wettable conditions), shrubs (very wettable to slight water repellency) and trees (very wettable to extreme water repellency). Although treecovered areas include also shrubs and herbaceous vegetation, our results show that residues from tree vegetation are the main source of soil hydrophobic substances. This study alone cannot explain the effects of soil WR on runoff and sediment connectivity patterns in the study area. Further studies should aim at mapping the influence of soil WR in connectivity pathways, identifying runoff generation and infiltration areas and sediment sources and sinks through the slope. ACKNOWLEDGEMENT(S) The first author received funding for an international stay during her MSc studies by the Mexican Council for Science 324 N. ALANÍS ET AL. Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017) and Technology (CONACYT). This study was funded by the Spanish Ministry of Economy and Competitiveness through the research projects POSTFIRE (CGL2013- 47862-C2-1-R) and GEOFIRE (CGL2012-38655-C04-01). Authors acknowledge Joan and the Cerdocarpa Team for their help during the fieldwork. Two anonymous reviewers have contributed to improve the quality of the original manuscript. REFERENCES Adams RH, Guzmán Osorio FJ, Zavala Cruz J. 2008. Water repellency in oil contaminated sandy and clayey soils. Source of the Document International Journal of Environmental Science and Technology 5:445–454. DOI: 10.1007/BF03326040. Álvarez-Icaza P, Cervera G, Garibay C, Gutierrez P, Rosete F. 1993. Los umbrales del deterioro: la dimensión ambiental de un desarrollo desigual en la región Purépecha. Fundación Friedrich Ebert Stiftung—UNAM– PAIR, Pátzcuaro. Arcenegui V, Mataix-Solera J, Guerrero C, Zornoza R, Mataix-Beneyto J, García-Orenes F. 2008. Immediate effects of wildfires on water repellency and aggregate stability in Mediterranean calcareous soils. Catena 74: 219–226. DOI:10.1016/j.catena.2007.12.008. Blackwell PS. 2000. Management of water repellency in Australia, and risks associated with preferential flow, pesticide concentration and leaching. Journal of Hydrology 231-232: 384–395. DOI:10.1016/ S0022-1694(00)00210-9. Bodí M, Mataix-Solera J, Doerr S, Cerdá A. 2011. The wettability of ash from burned vegetation and its relationship to Mediterranean plant species type, burn severity and total organic carbon content. Geoderma 160: 599–607. DOI:10.1016/j.geoderma.2010.11.009. Bodí M, Doerr SH, Cerdà A, Mataix-Solera J. 2012. Hydrological effects of a layer of vegetation ash on underlying wettable and water repellent soil. Geoderma 191:14–23. DOI:10.1016/j.geoderma.2012.01.006. Bouyoucos CJ. 1962. Hydrometer method improved for making particlesize analysis of soils. Agronomy Journal 54: 464–465. Buczko U, Bens O, Hüttl RF. 2005. Variability of soil water repellency in sandy forest soils with different stand structure under Scots pine (Pinus sylvestris) and beech (Fagus sylvatica). Geoderma 126: 317–336. DOI:10.1016/j.geoderma.2004.10.003. Cerdà A. 1998. Effect of climate on surface flow along a climatological gradient in Israel: a field rainfall simulation approach. Journal of Arid Environments 38: 145–159. DOI:10.1006/jare.1997.0342. Cerdà A, Doerr SH. 2007. Soil wettability, runoff and erodibility of major dry-Mediterranean land use types on calcareous soils. Hydrological Processes 21: 2325–2336. DOI:10.1002/hyp.6755. Cerdà A, Schnabel S, Ceballos A, Gomez-Amelia D. 1998. Soil hydrological response under simulated rainfall in the Dehesa land system Extremadura, SW Spain. under drought conditions. Earth Surface Processes and Landforms 23: 195–209. DOI:10.1002/(SICI)1096-9837(199803) 23:3<195::AID-ESP830>3.0.CO;2-I. Conde E, Cadahía E, García-Vallejo MC, Fernández de Simón B. 1998. Polyphenolic composition of Quercus suber cork from different Spanish provenances. Journal of Agricultural and Food Chemistry 46: 3166–3171. DOI:10.1021/jf970863k. De Rooij GH. 2000. Modelling fingered flow of water in soils owing to wetting front instability: a review. Journal of Hydrology 231-232: 277–294. DOI:10.1016/S0022-1694(00)00201-8. Dekker LW, Jungerius PD, Oostindie K. 2000. Extent and significance of water repellency in dunes along the Dutch coast. Journal of Hydrology 231-232: 112–125. DOI:10.1016/S0022-1694(00)00188-8. Dekker LW, Doerr SH, Oostindie K, Ziogas AK, Ritsema CJ. 2001. Water repellency and critical soil water content in a dune sand. Soil Science Society of America Journal 65: 1667–1674. DOI:10.2136/ sssaj2001.1667. Doerr SH. 1998. On standardizing the ‘water drop penetration time’and the ‘molarity of an ethanol droplet’techniques to classify soil hydrophobicity: a case study using medium textured soils. Earth Surface Processes and Landforms 23: 663–668. DOI:10.1002/(SICI)1096-9837(199807) 23:7<663::AID-ESP909>3.0.CO;2-6. Doerr SH, Thomas AD. 2000. The role of soil moisture in controlling water repellency: new evidence from forest soils in Portugal. Journal of Hydrology 231-232: 134–147. DOI:10.1016/S0022-1694(00)00190-6. Doerr SH, Shakesby RA, Walsh RPD. 1996. Soil hydrophobicity variations with depth and particle size fraction in burned and unburnt Eucalyptus globulus and Pinus pinaster forest terrain in the Águeda basin, Portugal. Catena 27:25–47. DOI:10.1016/0341-8162(96)00007-0. Doerr SH, Shakesby RA, Walsh RPD. 2000. Soil water repellency: its causes, characteristics and hydro-geomorphological significance. Earth- Science Reviews 51:33–65. DOI:10.1016/S0012-8252(00)00011-8. Doerr SH, Shakesby RA, MacDonald LH. 2009. Soil water repellency: a key factor in post-fire erosion? In Fire effects on soils and restoration strategies, Cerdà A, Robichaud PR (eds). Science Publishers: Enfield, N.H. Gabarrón-Galeote MA, Martínez-Murillo JF, Quesada MA, Ruiz-Sinoga JD. 2013. Seasonal changes in the soil hydrological and erosive response depending on aspect, vegetation type and soil water repellency in different Mediterranean microenvironments. Solid Earth 4: 497–509. DOI:10.5194/se-4-497-2013. García-Moreno J, Gordillo-Rivero AJ, Zavala LM, Jordán A, Pereira P. 2013. Mulch application in fruit orchards increases the persistence of soil water repellency during a 15-years period. Soil & Tillage Research 130: 62–68. DOI:10.1016/j.still.2013.02.004. Garduño Bernal N. 2011. Reserva de la Biósfera Mariposa Monrca en el Estado de México. Diagnóstico Fitosanitario Forestal. Gobierno del Edo. de México. Toluca. González-Peñaloza FA, Cerdà A, Zavala LM, Jordán A, Giménez-Morera A, Arcenegui V, Ruiz-Gallardo J-R. 2012. Do conservative agriculture practices increase soil water repellency? A case study in citrus-cropped soils. Soil & Tillage Research 124: 233–239. DOI:10.1016/j. still.2012.06.015. González-Peñaloza FA, Zavala LM, Jordán A, Bellinfante N, Bárcenas- Moreno G, Mataix-Solera J, Granged AJP, Granja-Martins FM, Neto- Paixão HM. 2013. Water repellency as conditioned by particle size and drying in hydrophobized sand. Geoderma 209-210:31–40. DOI:10.1016/j.geoderma.2013.05.022. Granged AJP, Zavala LM, Jordán A. 2011a. Post-fire evolution of soil properties and vegetation cover in a Mediterranean heathland after experimental burning: a 3-year study. Geoderma 164:85–94. DOI:10.1016/j. geoderma.2011.05.017. Granged AJP, Jordán A, Zavala LM, Bárcenas G. 2011b. Fire-induced changes in soil water repellency increased fingered flow and runoff rates following the 2004 Huelva wildfire. Hydrological Processes 25: 1614–1629. DOI:10.1002/hyp.7923. Guzmán Osorio FJ, Adams RH. 2015. Mitigation of water repellency in the treatment of contaminated muds using the chemical-biological stabilization process. International Journal of Environmental Science and Technology 12:2071–2078. DOI: 10.1007/s13762-014-0606-z. Hubbert KR, Preisler HK, Wohlgemuth PM, Graham RC, Narog MG. 2006. Prescribed burning effects on soil physical properties and soil water repellency in a steep chaparral watershed, southern California, USA. Geoderma 130: 284–298. DOI:. IBM Corp., 2015. IBM SOSS Statistics for Windows, Version 23 · 0. IBM Corporation. Armonk, NY. Imeson AC, Lavee H, Calvo A, Cerdà A. 1998. The erosional response of calcareous soils along a climatological gradient in Southeast Spain. Geomorphology 24:3–16. DOI:10.1016/S0169-555X(97)00097-4. Ito H, Yamaguchi K, Kim T-H, Khennouf S, Gharzouli K, Yoshida T. 2002. Dimeric and trimeric hydrolizable tannins from Quercus coccifera and Quercus suber.Journal of Natural Products 65: 339–345. Jiménez-Morillo NT, González-Pérez JA, Jordán A, Zavala LM, de la Rosa JM, Jiménez-González MA, González-Vila FJ. 2014. Organic matter fractions controlling soil water repellency in sandy soils from the Doñana National Park (southwestern Spain). Land Degradation and Development. DOI:10.1002/ldr.2314. Jiménez-Pinilla P, Lozano E, Mataix-Solera J, Arcenegui V, Jordán A, Zavala LM. 2015. Temporal changes in soil water repellency after a forest fire in a Mediterranean calcareous soil: influence of ash and different vegetation type. Science of the Total Environment . DOI:10.1016/j. scitotenv.2015.09.121. Jiménez-Pinilla P, Doerr SH, Ahn S, Lozano E, Mataix-Solera J, Jordán A, Zavala LM, Arcenegui V. 2016. Effects of relative humidity on the water repellency of fire-affected soils. Catena 138:68–76. DOI:10.1016/j. catena.2015.11.012. 325SPATIAL GRADIENTS OF SOIL WATER REPELLENCY IN MEXICAN FORESTS Copyright © 2016 John Wiley & Sons, Ltd. LAND DEGRADATION & DEVELOPMENT, 28: 317–327 (2017)