Agronomic Practices for Reducing Soil Erosion in Hillside Vineyards under Atlantic Climatic Conditions (Galicia, Spain)

Article Agronomic Practices for Reducing Soil Erosion in Hillside Vineyards under Atlantic Climatic Conditions (Galicia, Spain) JoséM. Mirás-Avalos 1,2,* , Juan M. Ramírez-Cuesta 1, María Fandiño 3, Javier J. Cancela 3 and Diego S. Intrigliolo 1 1Centro de Edafología y Biología Aplicada del Segura (CEBAS), Consejo Superior de Investigaciones Científicas (CSIC), Espinardo, 30100 Murcia, Spain; [email protected] (J.M.R.-C.); [email protected] (D.S.I.) 2 Unidad de Suelos y Riegos (asociada EEAD-CSIC), Centro de Investigaci ó n y Tecnolog í a Agroalimentaria de Aragón (CITA), Montañana, 50059 Zaragoza, Spain 3 GI-1716. Departamento de Ingenier í a Agroforestal. Escola Polit é cnica Superior de Enxeñar í a, Universidade de Santiago de Compostela, Campus de Lugo, 27002 Lugo, Spain; [email protected] (M.F.); [email protected] (J.J.C.) *Correspondence: [email protected]; Tel.: +34-976-713832 Received: 13 February 2020; Accepted: 31 March 2020; Published: 2 April 2020   Abstract: Water erosion is a severe threat to soil resources, especially on cultivated lands, such as vineyards, which are extremely susceptible to soil losses. In this context, management practices aiming at reducing erosion risks must be favored. This current study aimed at estimating soil losses in two vineyards under Atlantic climatic conditions (Galicia, North West Spain). The capacity of two management practices for reducing soil erosion was tested and compared with tilled soil in the inter-rows: (i) application of mulching, and (ii) maintaining native vegetation. Soil losses were assessed using erosion pins and micro-plots. In addition, the improved stock unearthing method (ISUM) was employed in one of the vineyards to estimate soil remobilization since plantation. Soil loss rates in one of the vineyards were lower when soil was managed under mulching (0.36 Mg ha −1 ) and native vegetation (0.42 Mg ha −1 ), compared to tilled soil (0.84 Mg ha −1 ). Sediment losses measured in the second vineyard ranged between 0.21 and 0.69 Mg ha −1 , depending on the treatment, but no clear conclusions could be drawn. Long-term soil loss, as estimated by ISUM, was of the same order of magnitude than that obtained by erosion pins and micro-plots. In both vineyards, soil loss rates were lower than those registered in Mediterranean vineyards, and were below the limit for sustainable erosion in Europe. Nevertheless, soil management practices alternative to tillage in the inter-row might reduce erosion risks under Atlantic climate conditions. Keywords: cover crops; mulching; runoff; soil erosion assessment; sustainability; temperate climate; vineyards; Vitis vinifera 1. Introduction Water erosion on cultivated lands represents a severe threat to soil resources, affecting 12% of the emerged lands in Europe [ 1 ], and causing substantial environmental and economic losses [ 2 ]. In this context, the concept of tolerable soil erosion has been defined as “any actual erosion rate at which a deterioration or loss of one or more soil functions does not occur” [2]. For Europe, actual soil loss rates for cultivated lands are unsustainable, being 3 to 40 times greater than the upper threshold (1.4 Mg ha −1 year −1 ) [ 2 ]. Among cultivated lands, vineyards, especially in Soil Syst. 2020,4, 19; doi:10.3390/soilsystems4020019 www.mdpi.com/journal/soilsystems Soil Syst. 2020,4, 19 2 of 15 sloping terrains, are highly susceptible to erosion, with substantial soil losses as compared to other agricultural lands [3–6]. A great research effort has been devoted to quantify soil loss rates in vineyards from all over Europe, revealing a large variability depending on the soil type, climate conditions and management practices [ 7 ]. For instance, erosion rates ranged from 0.04 to 11.5 Mg ha −1 year −1 in Catalonia, Spain [ 8 ], and they were, approximately, 7.9 Mg ha −1 year −1 in Madrid, Spain [ 9 ], 11.2 Mg ha −1 year −1 in Piedmont, Italy [ 10 ], 23.0 Mg ha −1 year −1 in Burgundy, France [ 4 ], and up to 62.5 Mg ha −1 year −1 in Rhineland-Palatinate, Germany [ 11 ], or 88.7 Mg ha −1 year −1 in Sicily, Italy [ 12 ]. Most of these studies have been conducted in Mediterranean areas, being vineyards under Atlantic or continental climates less surveyed due to the perception of these vineyards suffering from lower erosion rates [ 13 ]. In order to reduce these unsustainable rates, the employment of cover crops in the vineyard inter-rows has been proposed as a means for soil protection [14,15]. Therefore, the aims of the current study were two-fold: first, to quantify the soil erosion rates in two vineyards from an Atlantic area (North West Spain), and second, to assess the efficiency of two management techniques (mulching and native vegetation) for reducing the risk of soil erosion. We hypothesized that erosion rates in these two vineyards would be lower than those reported for Mediterranean sites, and that employing cover crops or mulching would reduce soil loss rates when compared with tilled soil. 2. Materials and Methods 2.1. Description of the Experimental Sites The experiments were conducted in two commercial vineyards (Figure 1), one of them planted on terraces within the Ribeira Sacra Designation of Origin (DO), and the other one planted without terraces, and following the slope direction in the Rías Baixas DO, both located in North West Spain. Soil Syst. 2020, 4, x FOR PEER REVIEW 2 of 16 especially in sloping terrains, are highly susceptible to erosion, with substantial soil losses as compared to other agricultural lands [3–6]. A great research effort has been devoted to quantify soil loss rates in vineyards from all over Europe, revealing a large variability depending on the soil type, climate conditions and management practices [7]. For instance, erosion rates ranged from 0.04 to 11.5 Mg ha−1 year−1 in Catalonia, Spain [8], and they were, approximately, 7.9 Mg ha−1 year−1 in Madrid, Spain [9], 11.2 Mg ha−1 year−1 in Piedmont, Italy [10], 23.0 Mg ha−1 year−1 in Burgundy, France [4], and up to 62.5 Mg ha−1 year−1 in Rhineland-Palatinate, Germany [11], or 88.7 Mg ha−1 year−1 in Sicily, Italy [12]. Most of these studies have been conducted in Mediterranean areas, being vineyards under Atlantic or continental climates less surveyed due to the perception of these vineyards suffering from lower erosion rates [13]. In order to reduce these unsustainable rates, the employment of cover crops in the vineyard inter-rows has been proposed as a means for soil protection [14,15]. Therefore, the aims of the current study were two-fold: first, to quantify the soil erosion rates in two vineyards from an Atlantic area (North West Spain), and second, to assess the efficiency of two management techniques (mulching and native vegetation) for reducing the risk of soil erosion. We hypothesized that erosion rates in these two vineyards would be lower than those reported for Mediterranean sites, and that employing cover crops or mulching would reduce soil loss rates when compared with tilled soil. 2. Materials and Methods 2.1. Description of the Experimental Sites The experiments were conducted in two commercial vineyards (Figure 1), one of them planted on terraces within the Ribeira Sacra Designation of Origin (DO), and the other one planted without terraces, and following the slope direction in the Rías Baixas DO, both located in North West Spain. Figure 1. Location of the experimental vineyards within Galicia (North West Spain). The vineyard from Ribeira Sacra is located in Doade, a municipality in the South of the Lugo Province (42°24′36.9′’N, 7°28′26.38′’W, 500 m above sea level). The vines (Vitis vinifera L. cv. Mencía) are, approximately, 7 years old, and are established in terraces, with two vine rows per terrace, Figure 1. Location of the experimental vineyards within Galicia (North West Spain). The vineyard from Ribeira Sacra is located in Doade, a municipality in the South of the Lugo Province (42◦24036.90’N, 7◦28026.380’W, 500 m above sea level). The vines (Vitis vinifera L. cv. Mencía) Soil Syst. 2020,4, 19 3 of 15 are, approximately, 7 years old, and are established in terraces, with two vine rows per terrace, spaced 2 m between rows and 1 m between the vines. In general, this area presents an average annual rainfall around 800 mm, and the soil at this site is developed on schists, and it is classified as an Inceptisol [ 16 ]. Three treatments were established in this vineyard: a) No tillage (native vegetation grows in the inter-row), b) Tillage in the inter-row and c) Mulching in the inter-row (a mixture of straw and residues from Ulex europaeus L.). Each treatment had three replications, each at a different altitude (one replicate per terrace). The vineyard from R í as Baixas is located in the municipality of Meis, in the Northwest of the Pontevedra Province (42 ◦ 34 0 03 0 ’N, 8 ◦ 45 0 22 0 ’W, 110 m above sea level). The vines (Vitis vinifera L. cv. Albariño) are, approximately, 14 years old, and are spaced 3 m between the rows and 2.5 m between the vines. In general, this area presents an average annual rainfall around 1500 mm, and the soil at this site is developed upon granite, and is classified as an Inceptisol [ 16 ]. Four treatments were established in this vineyard: a) Rain-fed without soil tillage, and rows oriented against the terrain slope (R-Against); b) Irrigated to cover 30% of the reference evapotranspiration (ET o ) during July and August, without soil tillage (I-NT); c) Irrigated to cover 30% of ET o during July and August, with soil tillage and herbicide application, so as to maintain soil without the protection of a vegetal cover (I-ST); and d) Rain-fed without soil tillage, and rows oriented in favor of the terrain slope (R-Slope). Treatments were laid out following a randomized block design with three replications. 2.2. Soil and Climate Characterization of the Experimental Vineyards At the beginning of the experiment, three samples from the soil surface layer (0–40 cm depth) were collected per experimental unit to characterize the soil in both vineyards. Its physical and chemical properties were determined following standard procedures [17]. For characterizing climate in both vineyards, data were collected from the nearest weather stations managed by MeteoGalicia (www.meteogalicia.gal) (‘Tremoedo’ in the case of R í as Baixas and ‘Ponte da Boga’ in the case of Ribeira Sacra). Data available from 2003 to 2017 were used for climate description at both sites. Apart from the description of the main weather variables, ET o and bioclimatic indices with relevance to viticulture were computed, including Huglin, cool night and dryness indices [ 18 , 19 ]. 2.3. Soil Loss Quantification 2.3.1. Erosion Pins In order to determine erosion rates in both vineyards, seven erosion pins per experimental unit (300 mm length, 7 mm diameter) were used as erosion markers, since they are the simplest and most effective method for monitoring the minute changes in the altitude of the ground surface which are due to erosion and deposition [ 20 ]. During May 2018, after two months since the last tilling practice, in both vineyards, pins were installed in the vine rows at regular intervals (5 m) following the slope of the terrain. The installation and measurement recommendations by Haigh [ 20 ] were taken into account. After installation, the initial height of the erosion pins (h i ) was recorded using a flex measurement tape. For each erosion pin, the difference (h) between the over-ground height at a given time (h f ) and the initial value (h i ) indicated the topographical change since pin installation, which is the soil erosion or deposition. The soil erosion volume was equal to the area multiplied by the plot length. For each vineyard, the soil bulk density was measured at 10 cm depth in one spot per experimental unit, using the core method [ 17 ] in order to transform the calculated volume into weight (Mg ha −1 ). Due to soil redistribution, patches of eroded and accumulated material were found. An area index (I) was calculated between two consecutive erosion pins using the following equation [12]: I= hf,1 −hf,1+n hi,1 +hi,1+n 2 (1) Soil Syst. 2020,4, 19 4 of 15 where h f,1 is the over-ground height of an erosion pin (cm) at the present time, and h f1+n is the height of the next erosion pin (cm) at lower elevation; the subscript iindicates the initial measurements of the above-ground height for each pin. Positive values for Irepresent soil erosion, while negative values indicate soil sedimentation [21]. 2.3.2. Erosion Plots To monitor the water and sediment yield over short time periods, two erosion plots (0.98 m long and 0.46 m wide: approximately, 0.5 m 2 ) per treatment were installed in the inter-rows of the R í as Baixas vineyard. Installation was completed in early January 2019, when a couple of erosion plots were substituted due to the damage caused by machinery trafficking. These micro-plots consisted of three metal sheets on the sides, and a V-shaped metal structure on the lower side for sample collection [ 22 ], which was performed using 8 L plastic containers connected to the erosion plot by a tube. This kind of system has been proven useful for assessing runoffand sediment yields under several cover crops in steep vineyards in Spain [ 9 ]. Runoffsamples were collected in 100 mL plastic jars to determine water and soil losses using a gravimetric method with oven drying [23]. 2.3.3. Improved Stock Unearthing Method In order to assess long-term soil losses in the R í as Baixas vineyard, we used a variant of the stock unearthing method (SUM). The SUM is based on the measurements of the vertical distance between frontal marks on the graft union (visible on grape vines) and the actual topsoil level [ 4 ]. At planting, all graft unions are located at 2 cm above the levelled surface of the soil to avoid the detrimental effects of soil moisture, freezing, or fungal infections [ 24 ]. The distance of the graft union in relation to the soil surface does not vary as a result of plant growth, and it is considered a fixed position [ 4 ]. Therefore, changes in the distance between the grafting point and the soil surface are assumed to be consequence of micro-topographical changes, namely soil depletion or the accumulation of sediments [ 25 ]. The main limitation of SUM is the assumption that the topsoil surface between the vine rows remains constantly planar, without any measuring of the uncertainty due to roughness by rills, footpaths and wheel tracks [ 26 ]. In order to overcome this limitation, we employed the improved stock unearthing method (ISUM), which includes measures of topsoil level on three spots in the inter-row between two facing vine stocks [26]. In order to estimate soil losses which have occurred from vineyard plantation, by the end of the experiment, on 10th October 2019, the ISUM was applied to the vineyard located in R í as Baixas. From the vine stock graft union, an aluminum bar with one level attached to each side was put at a given height above the grafting point. This allowed for measuring both buried vine stocks (positive measures) and uncovered ones (negative measures). Two points at a distance of 0.75 m from the vines, and one directly in the middle (1.5 m) of the inter-row area, were also measured [24]. The measurements were always conducted by the same researcher in order to avoid any bias. When the surface had little steps or grass cover, we carefully levelled the surface with the nearby current topsoil level. After measuring the graft union, three more points were also measured. Therefore, for every paired vine, we obtained five different points. The survey was conducted on 13 paired rows distributed all over the vineyard, including those in which the treatments were imposed, in order to account for the spatial variability of this soil erosion. Depending upon the length of the vine rows, ISUM measurements were performed on 10 to 12 sections per row. In total, 670 points were measured. For final estimations, an addition of 2 cm, corresponding to the initial graft union distance to the soil, was applied to all measurements. The total soil loss (Mg ha −1 yr −1 ) was estimated using the volume differences (m 3 ) of soil. The volumes were computed by creating imaginary polygons, which were delimited as the distance between each graft union (3 m) and the average of point measures. Soil Syst. 2020,4, 19 5 of 15 The height of this polygon corresponded to the distance between the botanic marks and the measured point within the inter-row [ 26 ]. Total soil loss was estimated from the erosion–deposition (ER) equation [27]: ER =Vol ×BD St ×Av (2) where the volume (Vol), the total area of the field (St), the age of the vines (Av; 15 years) and the bulk density (BD; 1.13 Mg m−3) were used as inputs. 2.4. Statistical Analysis An analysis of variance considering the treatment imposed in the field as factor was used for assessing the differences among management practices in each vineyard. When necessary, the means were separated using the Tukey Honest Significant Difference test. Statistical tests were conducted within the R Statistical Environment v.3.6.1 [28]. 3. Results 3.1. Climate and Soil Characterization of the Experimental Sites The station (“Ponte da Boga”) located close to the Ribeira Sacra vineyard recorded an annual mean temperature of 13.0 ◦ C for the period from 2003 to 2017. Annual rainfall and ET o amounted 786 and 1055 mm, respectively. If referred to the growing season (April to September), mean temperature, total rainfall and ET o were 17.1 ◦ C, 239 mm and 806 mm, respectively. Huglin and cool night indices for this station are, respectively, 2028 ◦ C and 11.9 ◦ C. The dryness index is − 20.96 mm, suggesting the existence of a drought period over the average growing season. The station (“Tremoedo”) located close to the R í as Baixas vineyard recorded an annual mean temperature of 14.3 ◦ C for the period from 2003 to 2017. Annual rainfall and ET o amounted 1433 and 948 mm, respectively. If referred to the growing season (April to September), mean temperature, total rainfall and ET o were 17.4 ◦ C, 417 mm and 696 mm, respectively. Huglin and cool night indices for this station are, respectively, 1880 ◦ C and 13.5 ◦ C. The dryness index is 135.3 mm, indicating the absence of a drought period over the average growing season. Concerning the soils from the two experimental vineyards, Table 1summarizes their main physical and chemical properties. Table 1. Main physical and chemical properties from the surface layer of the soils from the two experimental vineyards studied (averages ±standard errors). Soil Property Units Rías Baixas Ribeira Sacra Bulk density g cm−31.13 ±0.02 1.23 ±0.08 Fine fraction % 59.8 ±0.3 58.3 ±1.8 Coarse fraction 40.2 ±0.9 41.7 ±1.8 Sand 54.2 ±0.7 41.3 ±2.9 Silt 22.3 ±0.7 32.9 ±2.0 Clay 23.5 ±0.3 25.8 ±1.7 Organic matter 4.06 ±0.14 2.94 ±0.27 pH (H2O) 5.84 ±0.08 5.19 ±0.19 pH (KCl) 5.27 ±0.08 4.39 ±0.20 Effective cation exchange capacity cmol(+) kg−18.50 ±0.46 4.97 ±0.72 Electrical conductivity mS cm−1149.1 ±8.8 75.8 ±12.0 The soil from the R í as Baixas vineyard has a sandy clay loam texture, whereas that from Ribeira Sacra is loamy. This difference in texture between the soils in the two experimental vineyards reflects in the bulk density, which was lower in R í as Baixas than in Ribeira Sacra. Both soils have a pH in Soil Syst. 2020,4, 19 6 of 15 water ranging from 4.9 to 6.0 (Table 1), so they are acid soils with no base saturation, and a very good solubility for iron. The soils from both vineyards are mineral with a medium content in organic matter and low salinity (Table 1). The cation exchange capacity is low in the soil from R í as Baixas, and very low in that from Ribeira Sacra. 3.2. Weather Conditions over the Study Period Over the study period, May 2018 to August 2019, 1004 and 1384 mm rainfall were registered in Ribeira Sacra and R í as Baixas, respectively. The distribution of this rainfall amounts over the study period differed between vineyards (Figure 2). In Ribeira Sacra, June and July 2018 were rainy, with daily events of more than 50 mm by mid-July. However, the magnitude of rainfall events was reduced, but their frequency increased between October 2018 and May 2019 (Figure 2a). In R í as Baixas, June and July 2018 were dry, with sporadic rainfall events. In contrast, the frequency and magnitude of rainfall events increased from October 2018 to May 2019, with several daily events surpassing 40 mm (Figure 2b). Evapotranspiration rates followed a similar pattern in both vineyards, although slightly higher values were observed in Ribeira Sacra (Figure 2). Soil Syst. 2020, 4, x FOR PEER REVIEW 6 of 16 3.2. Weather Conditions over the Study Period Over the study period, May 2018 to August 2019, 1004 and 1384 mm rainfall were registered in Ribeira Sacra and Rías Baixas, respectively. The distribution of this rainfall amounts over the study period differed between vineyards (Figure 2). In Ribeira Sacra, June and July 2018 were rainy, with daily events of more than 50 mm by mid-July. However, the magnitude of rainfall events was reduced, but their frequency increased between October 2018 and May 2019 (Figure 2a). In Rías Baixas, June and July 2018 were dry, with sporadic rainfall events. In contrast, the frequency and magnitude of rainfall events increased from October 2018 to M (a) (b) Figure 2. Dynamics of daily rainfall and reference evapotranspiration (ETo) over the study period (May 2018–August 2019) for (a) “Ponte da Boga” in Ribeira Sacra; (b) “Tremoedo” in Rías Baixas weather stations. 3.3. Soil Loss Quantification 3.3.1. Erosion Pins The differences between pin height at the installation date and at the end of the experiment for the Ribeira Sacra vineyard are illustrated in Figure 3. Pin height variation was not constant, showing high variability depending on the treatment and the terraces (Figure 3a). The area index was also highly variable, although mulching seemed to favor deposition processes when compared to the Figure 2. Dynamics of daily rainfall and reference evapotranspiration (ET o ) over the study period (May 2018–August 2019) for ( a ) “Ponte da Boga” in Ribeira Sacra; ( b ) “Tremoedo” in R í as Baixas weather stations. Soil Syst. 2020,4, 19 7 of 15 3.3. Soil Loss Quantification 3.3.1. Erosion Pins The differences between pin height at the installation date and at the end of the experiment for the Ribeira Sacra vineyard are illustrated in Figure 3. Pin height variation was not constant, showing high variability depending on the treatment and the terraces (Figure 3a). The area index was also highly variable, although mulching seemed to favor deposition processes when compared to the other treatments (negative values in Figure 3b), whereas tillage seemed to favor erosion (positive values in Figure 3b). Soil Syst. 2020, 4, x FOR PEER REVIEW 7 of 16 other treatments (negative values in Figure 3b), whereas tillage seemed to favor erosion (positive values in Figure 3b). -0,5 -0,4 -0,3 -0,2 -0,1 0,0 0,1 0,2 0,3 0,4 0,5 0 5 10 15 20 25 30 35 Distance (m) Pin height variation (cm) Native vegetation Mulching Tillage (a) -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 0,8 0 5 10 15 20 25 30 35 Distance (m) Area index Native vegetation Mulching Tillage (b) Figure 3. Pin height variation (a) and area index (b) over a year (May 2018 to April 2019) timespan for the treatments imposed in the Ribeira Sacra vineyard. Error bars refer to standard errors. Positive values of the area index indicate erosion, whereas negative values indicate soil deposition. When extrapolated to the whole area surveyed in the Ribeira Sacra vineyard, soil erosion occurred in all the treatments considered, following the rank order expected: tillage (0.84 Mg ha−1), native vegetation (0.42 Mg ha−1) and mulching (0.36 Mg ha−1). However, due to the high variability among replications, no significant differences were detected among treatments. In the Rías Baixas vineyard, the differences between pin heights at the beginning and end of the assessment period are illustrated in Figure 4. Pin height variation was not constant along the slope, but ranged from −0.6 to 0.7 cm, depending on the treatment (Figure 4a). Figure 3. Pin height variation ( a ) and area index ( b ) over a year (May 2018 to April 2019) timespan for the treatments imposed in the Ribeira Sacra vineyard. Error bars refer to standard errors. Positive values of the area index indicate erosion, whereas negative values indicate soil deposition. Soil Syst. 2020,4, 19 8 of 15 When extrapolated to the whole area surveyed in the Ribeira Sacra vineyard, soil erosion occurred in all the treatments considered, following the rank order expected: tillage (0.84 Mg ha −1 ), native vegetation (0.42 Mg ha −1 ) and mulching (0.36 Mg ha −1 ). However, due to the high variability among replications, no significant differences were detected among treatments. In the R í as Baixas vineyard, the differences between pin heights at the beginning and end of the assessment period are illustrated in Figure 4. Pin height variation was not constant along the slope, but ranged from −0.6 to 0.7 cm, depending on the treatment (Figure 4a). Soil Syst. 2020, 4, x FOR PEER REVIEW 8 of 16 -0,8 -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 0,8 0 20 40 60 80 100 120 Distance (m) Pin height variation (cm) I-ST R-Slope I-NT R-Against (a) -0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 0 20 40 60 80 100 120 Distance (m) Area index I-ST R-Slope I-NT R-Against (b) Figure 4. Pin height variation (a) and area index (b) over a year (May 2018 to April 2019) timespan for the treatments imposed in the Rías Baixas vineyard. Positive values of the area index indicate erosion, whereas negative values indicate soil deposition. R-Against: Rain-fed without soil tillage and rows oriented against the terrain slope; I-NT: Irrigated to cover 30% of reference evapotranspiration (ETo) without soil tillage; I-ST: Irrigated to cover 30% of ETo with soil tillage and herbicide application to maintain soil without the protection of a vegetal cover; and R-Slope: Rain-fed without soil tillage and rows oriented in favor of the terrain slope. The area index ranged from −0.53 (deposition area) to 0.45 (erosion area), and was different among treatments (Figure 4b). Those treatments in which soil was covered by native vegetation recorded negative erosion rates, I-NT = −0.03 ± 0.03 Mg ha−1 and R-Against = −0.17 ± 0.00 Mg ha−1, except for R-Slope = 0.10 ± 0.02 Mg ha−1. In contrast, the treatment in which soil was tilled registered positive erosion rates: I-ST = 0.10 ± 0.01 Mg ha−1. Figure 4. Pin height variation ( a ) and area index ( b ) over a year (May 2018 to April 2019) timespan for the treatments imposed in the R í as Baixas vineyard. Positive values of the area index indicate erosion, whereas negative values indicate soil deposition. R-Against: Rain-fed without soil tillage and rows oriented against the terrain slope; I-NT: Irrigated to cover 30% of reference evapotranspiration (ET o ) without soil tillage; I-ST: Irrigated to cover 30% of ET o with soil tillage and herbicide application to maintain soil without the protection of a vegetal cover; and R-Slope: Rain-fed without soil tillage and rows oriented in favor of the terrain slope. Soil Syst. 2020,4, 19 9 of 15 The area index ranged from − 0.53 (deposition area) to 0.45 (erosion area), and was different among treatments (Figure 4b). Those treatments in which soil was covered by native vegetation recorded negative erosion rates, I-NT = − 0.03 ± 0.03 Mg ha −1 and R-Against = − 0.17 ± 0.00 Mg ha −1 , except for R-Slope =0.10 ± 0.02 Mg ha −1 . In contrast, the treatment in which soil was tilled registered positive erosion rates: I-ST =0.10 ±0.01 Mg ha−1. 3.3.2. Erosion Plots In those plots installed in May 2018, sediments and runoffsamples were collected from mid-June 2018 to the end of August 2019, whereas in those installed in January 2019, samples were collected from the end of January to the end of August 2019 (Table 2). Therefore, precipitation over the plots differed, as displayed in Table 2, which divides the collections in four periods of, approximately, four months each, and differing in rainfall amounts (Table 2). Precipitation over the first period (12/06/2018 to 17/10/2018) amounted 109.4 mm, which were distributed in 24 days, with two events greater than 20 mm. Precipitation over the second period (17/10/2018 to 21/01/2019) amounted 561 mm, which were distributed in 53 days, with nine events greater than 20 mm. Precipitation over the third period (22/01/2019 to 29/04/2019) amounted 459 mm, which were distributed in 41 days, with eight events greater than 20 mm. Precipitation over the fourth period (30/04/2019 to 29/08/2019) amounted 180 mm, which were distributed in 27 days, with two events greater than 20 mm. Table 2. Rainfall, runoffand sediments mobilized by runofffor the different treatments surveyed in the Rías Baixas vineyard from mid-June 2018 to end August 2019. Period Treatment Rainfall (mm) Runoff(L m−2) Sediment (kg ha−1) First (12/06/2018–16/10/2018) R-Slope 109 4.58 13.3 I-NT 4.41 19.7 I-ST 5.59 22.5 Second (17/10/2018–21/01/2019) R-Slope 561 42.95 79.7 I-NT 34.47 340.4 I-ST 20.56 43.1 Third (22/01/2019–29/04/2019) R-Slope 459 29.05 153.5 I-NT 31.56 298.4 I-ST 32.16 198.2 R-Against 55.19 151.7 Fourth (30/04/2019–29/08/2019) R-Slope 180 7.99 21.8 I-NT 5.49 30.9 I-ST 6.27 34.4 R-Against 7.00 51.9 Total R-Slope 1310 84.57 268.3 I-NT 75.93 689.4 I-ST 64.58 298.2 R-Against 639 62.19 203.6 R-Against: Rain-fed without soil tillage and rows oriented against the terrain slope; I-NT: Irrigated to cover 30% of reference evapotranspiration (ET o ) without soil tillage; I-ST: Irrigated to cover 30% of ET o with soil tillage and herbicide application to maintain soil without the protection of a vegetal cover; and R-Slope: Rain-fed without soil tillage and rows oriented in favor of the terrain slope. In the first period, despite being similar in runoffcollected, R-Slope and I-NT treatments differed in the sediment yield, which was 30% greater in I-NT (13.3 versus 19.7 kg ha −1 , for R-Slope and IN-T, respectively; Table 2). The I-ST treatment showed the highest runoffand sediment yields (5.59 L m −2 and 22.5 kg ha −1 , respectively; Table 2). In the second period, runoffand sediment yield increased in all treatments, especially in I-NT, which reached sediment yield values of 340.4 kg ha −1 (Table 2). In the third period, sediment yields increased, except for the I-NT treatment, where they were of the same order of magnitude as in the former period (298.4 kg ha −1 ). It is noticeable as to the high value of runofffor the R-Against treatment (55.19 L m−2; Table 2).