European Journal of Agronomy 140 (2022) 126597 Contents lists available at ScienceDirect European Journal of Agronomy journal homepage: www.elsevier.com/locate/eja Effects of herbaceous covers and mineral fertilizers on the nutrient stocks and fluxes in a Mediterranean olive grove S. Marañón-Jiménez a, b, *, P. Serrano-Ortíz c, J. Peñuelas b, d, A. Meijide e, S. Chamizo f, g, A. López-Ballesteros h, J.L. Vicente-Vicente i, E. Fernández-Ondoño j a Autonomous University of Barcelona, Department of Animal Biology, Vegetal Biology and Ecology, 08193, Bellaterra, Spain b CREAF, 08193, Bellaterra, Spain c University of Granada, Department of Ecology, Av. Fuentenueva s/n, 18071 Granada, Spain d CSIC, Global Ecology Unit CREAF-CSIC-UAB, 08193 Bellaterra, Spain e University of Göttingen, Department of Crop Sciences, Division Agronomy, Von Siebold Str. 8, 37075 Göttingen, Germany f University of Almeria, Department of Agronomy, La Cañada de San Urbano, 04120 Almeria, Spain g Centre for Scientific Collections from the University of Almería (CECOUAL), La Cañada de San Urbano, 04120 Almeria, Spain h Department of Agricultural and Forest Systems and the Environment, Agrifood Research and Technology Centre of Aragon (CITA). Avda. Montañana 930, 50059 Zaragoza, Spain i Leibniz Centre for Agricultural Landscape Research (ZALF), Eberswalder Straße 84, 15374 Müncheberg, Germany j University of Granada, Department of Soil Science and Agricultural Chemistry, Av. Fuentenueva s/n, 18071 Granada, Spain A R T I C L E I N F O A B S T R A C T Keywords: The preservation of nutrient capital, soil fertility, and carbon (C) sequestration capacity in Mediterranean olive Agroecosystem groves requires evaluation of agricultural practices beyond short-term productivity. We aim to contribute with a Litter decomposition mechanistic understanding on the effects that the preservation of herbaceous cover and the use of chemical Microclimate amelioration fertilizers have on the performance of olive trees and on the biogeochemical cycles of the agroecosystem. We Nutritional status compared nutrient fluxes and aboveground leafy stocks in an olive grove that had been organically managed for Olive production Weed cover more than 60 years, in a treatment in which the annual spontaneous herbaceous cover was maintained (H), and Sustainable agriculture after two years of shift to conventional management treatments in which the growth of herbaceous vegetation was avoided by the use of herbicides (NH), and where exclusion of the herbaceous cover is also combined with the supply of mineral fertilizers (NHF). Maintenance of herbaceous vegetation in H contributed to the retention of a high aboveground capital of C and nutrients, particularly nitrogen, (N), phosphorus (P) and potassium (K) that were about 2.9, 3.9 and 7.4 times greater than in NH, respectively. The permanence of herbaceous cover stimulated olive tree leaf litter decomposition rates by about 86 % and increased nutrient release. However, the H treatment led to a 37 % decrease in olive yield and lowered olive foliar N and P content as negative short-term effects. The addition of fertilizers (N, P, K, and Mg) in mineral and solid form in NHF resulted inefficient to improve olive tree nutritional status and olive production, and decelerated olive tree litter decomposition rates by 21 % and nutrient release. The nutrient retention in organic forms in the fast-growing species of herbaceous covers and the progressive nutrient release as litter decomposes may contribute to regulate and better adapt nutrient availability to the nutrient requirements of olive trees. 1. Introduction 76 % of world production, and of this, 75 % derives from Spain (FAO­ STAT, 2022). However, the implementation of inadequate conventional Olive trees (Olea europaea L.) cover around 5.1 Mha in Europe, which soil-management practices in olive groves has led to several environ­ accounts for 49 % of the global area of olive cultivation (FAOSTAT, mental problems such as high runoff and erosion rates, soil C and 2022), with a large proportion cultivated on poor or rocky soils of the fertility losses (Gó mez et al., 2009a, 2009b). The implementation of Mediterranean basin. European olive oil production represents around sustainable agricultural practices in olive groves may potentially derive * Corresponding author at: Autonomous University of Barcelona, Department of Animal Biology, Vegetal Biology and Ecology, 08193, Bellaterra, Spain. E-mail address: s.maranon@creaf.uab.cat (S. Marañón-Jiménez). https://doi.org/10.1016/j.eja.2022.126597 Received 26 January 2022; Received in revised form 26 July 2022; Accepted 27 July 2022 Available online 5 August 2022 1161-0301/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 large environmental, social, and economic benefits for Europe (Camarsa understanding of factors that enhance adaptive capacity and resilience et al., 2010). For this reason, the European Union has prioritized agri­ of ecosystems (Angeler et al., 2019). For example, ecosystems charac­ cultural policies designed to mitigate effects of global environmental terized by frequent disturbances and low complexity ecological in­ change in olive groves, such as high-level agreement to reform the teractions select for r-strategists, such as the ruderal weed species that Common Agricultural Policy (CAP) and link 30 % of direct agricultural farmers tend to control in agricultural systems. K-strategists may play an payments to environmentally-friendly farming practices. The assess­ important role in buffering effects of disturbances in agroecosystems, ment of sustainable olive grove management practices is therefore such as during crop harvest, since most of their energy is allocated to essential for the design of policies that guarantee conservation and de­ vegetative plant parts. A wide spectrum of life-history strategies in livery of ecosystem services in these agricultural systems, including soil ecosystems may thus increase functional diversity and spatio-temporal retention, carbon (C) sequestration and biodiversity (FAO, 2004). heterogeneity (Gliessman, 2007), allowing to take advantage of oppor­ The maintenance of spontaneous resident vegetation cover (here­ tunities created by perturbations and to prevent major losses in after referred to as herbaceous cover) has become an effective conser­ ecosystem functions and services (Wood et al., 2015). However, an vation practice in olive groves if managed adequately (Nieto et al., ecological framework for olive groves that accounts for biogeochemical 2013). The presence of herbaceous cover during autumn and spring functioning and nutritional strategies of olive trees and herbaceous reduces soil erosion (Gómez et al., 2009b), ameliorates the soil micro­ vegetation remains lacking. climate (Kairis et al., 2013), improves soil physicochemical properties In this study, we aim to contribute with a mechanistic understanding (Sastre et al., 2018), and increases interception and storage of rainfall on the individualized effects that the preservation of the herbaceous water and water availability in deep soil layers (Palese et al., 2014). cover and the use of chemical fertilizers have on the performance of Greater organic C inputs to the soil associated with herbaceous cover olive trees and on the biogeochemical cycles of the agroecosystem. For also translate into increases in soil C content (Vicente-Vicente et al., that, we compare C and nutrient stocks and fluxes in aboveground leafy 2016; Morugán-Coronado et al., 2020) and greater levels of ecosystem C litter (herbaceous biomass and olive leaf litter) of olive grove plots with sequestration (Aguilera et al., 2015; Chamizo et al., 2017). The the presence of herbaceous cover (H), with inputs of mineral fertilizers increasing availability of organic substrates also enhances microbial (NHF) and without any herbaceous cover or fertilizers inputs (NH) from activity and biogeochemical cycling, with the consequent improvement April 2015 to October 2017. This allows to 1) assess, for the first time, of soil structure and fertility (Soriano et al., 2014; Herencia, 2015). the direct and indirect management effects of these individualized Despite their contribution to soil fertility, there is no consensus practices on litter decomposition and on the rates of release or immo­ regarding the effects of vegetation covers on the nutritional status of bilization of C and nutrients, 2) determine the magnitude and relative olive groves (Ferreira et al., 2013; Zipori et al., 2020). Nonetheless, relevance of C and nutrient pools in the aboveground leafy litter in each spontaneous vegetation, and leguminous in particular, may contribute of these management treatments, and 3) evaluate the effects of these to reduce dependency of crops on external nutrient inputs by enhancing organic versus conventional management practices on the nutritional soil microbial communities involved in the soil N cycle, leading to status and yield of olive trees. We predict lower nutrient stocks but faster improvement of leaf and fruit nutritional status (Rodrigues et al., 2015; nutrient fluxes in the aboveground herbaceous vegetation compared to Lombardo et al., 2021). Moreover, this practice reduces considerably the the olive leaf litter. As a result, we hypothesize that the coexistence of management cost and environmental damage related to the repeated use herbaceous vegetation and olive trees would lead to tighter nutrient on herbicides, tillage or land-clearing using heavy machinery (Guzmán cycles in olive grove agroecosystem due to their contrasting and com­ and Foraster, 2011) and also contributes to higher plant and animal plementary ecological strategies for the acquisition of limiting re­ biodiversity (Rey et al., 2019), increasing the resilience of olive groves sources, despite the competition for these resources in the short-term. By to pests (Paredes et al., 2019). contrast, we predict a lower efficiency in the use of nutrients by olive However, the presence of herbaceous covers may also have some trees when these are provided in the form of mineral fertilizers. disadvantages. For instance, the interference with harvesting and other cultural operations and greater damage from spring frosts (Saavedra, 2. Methods 2007a) and the increased risk of fires are some arguments against the presence of herbaceous covers that have been used by olive farmers 2.1. Study site (Koutsias et al., 2012). Herbaceous covers may also compete with young olive trees for water, resulting into reductions in olive yield (Ferreira The study site, an irrigated orchard of olive trees var. ‘Arbequina’, et al., 2013). Nonetheless, the net effect on soil water availability is was located at Cortijo Guadiana, near Jaén, Spain (37◦ 55’10.13"N, controversial (Ruíz-Colmenero et al., 2011; Palese et al., 2014). It is, as 3◦ 14’24.62"W, Fig. 1a). The site is situated at 370 m above sea level. The well, not clear whether herbaceous vegetation also competes with olive climate is classified as Mediterranean, with hot dry summers and mild trees for nutrients, whether yield reductions are extensible to mature winters. The highest air temperatures occur in July (mean maximum: trees (Palese et al., 2014; Soriano et al., 2014), and whether this 37.4 ◦ C), the lowest air temperatures in January (mean minimum: competition also translates into reductions in olive oil quality or pro­ 0.8 ◦ C), and the mean annual temperature for the area is 16.1 ◦ C. Mean duction (García-González et al., 2019). Understanding these knowledge annual precipitation is 466 mm, with the majority occurring in spring gaps may allow optimization of the potential benefits of herbaceous and autumn (data obtained from 15-year records at the Agroclimatic cover maintenance for olive productivity and sustainability. Station of Úbeda, located at 7 km from our study site, Red de Olive groves, as agroecosystems, may be framed within the r/K se­ Información Agroclimática de Andalucía, 2022, http://www.junt lection theory (Margalef, 1959), where herbaceous vegetation and olive adeandalucia.es/agriculturaypesca/ifapa/ria/servlet/FrontController). trees have contrasting life-history strategies. Herbaceous vegetation The olive trees, which were about 20 years old, planted at 12 m comprises opportunistic with rapid growth rates, short generation times, average distance among olive trees and at a density of 69.6 trees ha-1, small sizes and efficient dispersal and colonization strategies useful for were located in a flat area that had been managed using organic agri­ unstable, ruderal environments, typical of r-strategists. Olive trees are, cultural practices for more than 60 years. The study area in the olive in contrast, specialized stress-tolerators adapted to more stable envi­ plantation was selected due to its representative and well-established, ronments, have slow growth rates, longer generation times, and larger non-perennial herbaceous cover that generally senesced at the start of sizes, typical of K-strategists (Gunderson and Holling, 2001). Although the summer dry period (May) and re-emerged in autumn following the limitations of this dichotomous r/K selection concept have been high­ first post-summer rains (Rey et al., 2019; Tarifa et al., 2021). The most lighted in empirical studies (Roff, 1993), positioning species along an abundant species in the herbaceous cover of the olive grove were Med­ r/K continuum (Jones, 1976) may, nevertheless, be useful to improve icago polymorpha (61 % frequency of occurrence), Torilis arvensis (61 %), 2 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 1. A. Study site location in Jaén province on the Iberian Peninsula, B. Aerial view of the study plots. An olive tree was initially missing from the grid in the selected area, so the area was extended to include an additional tree. Scorpiurus muricatus (56 %), Bromus madritensis (47 %), Anacyclus clav­ only fertilization in the study area was applied as foliar fertilizer. An atus (44 %), Galium aparine (39 %) and Hordeum murinum (36 %). The aqueous solution containing 60 g tree-1 of NPK fertilizer in a ratio management in the selected study area consisted in the maintenance of 14:5:25 was sprayed to the leaves with an atomizer twice per year, once the spontaneous herbaceous cover from autumn to spring. In spring, in spring and once in autumn. In order to determine the effects of the herbaceous vegetation was mowed mechanically and left on the surface treatments on the nutritional status and yield of olive trees, foliar to avoid competition for water with trees. The olive groves were irri­ fertilization was not further applied over the course of the experiment. gated using pressure compensating drippers (Netafim, Tel Aviv, Israel) Soils were classified as Calcaric Regosols (Table 1). once per week from May to October, at an average rate of 34 m3 ha-1 week-1 using water from an irrigation pond. The irrigation water is 2.2. Baseline soil physicochemical properties characterized by low sodium hazard, high salinity hazard and moderate levels of chloride (Zaman et al., 2018) as indicated in Appendix 1. The We analyzed pre-treatment soil physicochemical properties from a 3 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Table 1 exchange capacity was obtained after saturation of the soil exchange Initial soil variables at two soil depths before establishment of the management complex with Na+ cations, by adding sodium acetate, and then displaced treatments. TC: total carbon; TN: total nitrogen; SOM: soil organic matter; SOC: Na+ cations with ammonium acetate by atomic absorption (Soil Con­ soil organic carbon; DOC: dissolved organic carbon; CEC: cation exchange ca­ servation Service, 1972). Soil pH was determined by stirring and settling pacity. Values are means ± standard errors from 18 soil samples, collected at in distilled water with a pH meter (Crison micropH-2001, Barcelona, three distances from six randomly selected olive trees. Spain), according to (Pansu and Gautheyrou, 2006). Soil texture was Soil depth 0–5 cm 5–15 cm determined using the standard pipette methods of Robinson-Köhn or Texture (%): Clay loam Andreasen (Pansu and Gautheyrou, 2006), and the mineralogy of a Sand (0.05–2 mm) 25.65 ± 0.95 24.90 ± 1.19 subsample of soil that was milled to powder was determined by X-ray Coarse loam (0.02–0.05 mm) 10.97 ± 0.52 12.05 ± 0.34 diffraction (Whittig and Allardice, 1986) using an X-ray diffractometer Fine loam (0.002–0.02 mm) 29.66 ± 0.61 27.41 ± 0.92 (Bruker D8 Advance, Madrid, Spain). Clay (<0.002 mm) 33.76 ± 0.81 35.63 ± 0.99 Mineralogy (%): Quartz 17 ± 2.31 20.33 ± 3.84 2.3. Experimental design Calcite 32.67 ± 3.53 26 ± 1.53 Muscovite 24 ± 1.53 28.33 ± 2.60 In April 2015, three agricultural management treatments, with Paragonite 4.67 ± 0.33 5.33 ± 1.45 Plagioclase 6±1 6 ± 1.73 contrasting herbaceous cover and use of mineral fertilizers, were Dolomite 11.67 ± 1.86 9.67 ± 1.45 randomly applied to three plots of nine replicated trees. Treatments Filosilicate 4 ± 0.58 4.33 ± 0.67 comprised the maintenance of herbaceous vegetation without mowing Soil fertility parameters: during all the study period (H), use of herbicide to avoid the growth of pH 8.46 ± 0.02 8.58 ± 0.02 TC (%) 6.09 ± 0.04 5.67 ± 0.04 herbaceous vegetation (NH), and use of herbicide combined with the TN (%) 0.20 ± 0.00 0.15 ± 0.01 increase of nutrient inputs by the addition of solid mineral fertilizers SOM (g kg-1) 49.65 ± 2.49 33.68 ± 2.33 (NHF). Since herbaceous cover at the study site was well established, the SOC (%) 2.88 ± 0.15 1.96 ± 0.14 H treatment did not require any intervention. The NH and NHF treat­ SOC:TN 14.19 ± 0.62 12.78 ± 0.72 ments were maintained free of herbaceous cover along the complete DOC (mg kg-1) 223.80 ± 15.29 118.51 ± 16.85 CaCO3 (% eq.) 41.46 ± 0.42 41.85 ± 0.38 duration of this study (April 2015 to October 2017). For that, herba­ CEC (cmol+ kg-1) 17.35 ± 0.57 16.49 ± 0.62 ceous vegetation was first cleared using a weeding machine, and then an Base saturation (%) saturated saturated aqueous mixture of herbicides (24 % oxyfluorfen at 27.8 L ha-1 and 36 % Ca (cmol+ kg-1) 11.49 ± 0.64 11.87 ± 0.56 glyphosate at 66.7 L ha-1), similar to that used in conventional man­ Mg (cmol+ kg-1) 2.86 ± 0.09 3.23 ± 0.13 agement of olive groves (Saavedra, 2007b), was applied evenly over the K (cmol+ kg-1) 2.85 ± 0.30 1.38 ± 0.13 Na (cmol+ kg-1) 0.41 ± 0.08 0.31 ± 0.03 soil surface using air-carrier sprayers annually in spring. An additional Pinorg (mg kg-1) 5.13 ± 0.47 2.17 ± 0.45 application of preemergence herbicide (oxyfluorfen 24 %, 27.8 L ha-1) NO-2 (mg kg-1) 11.53 ± 3.73 4.26 ± 1.34 was also applied annually in autumn in the same way as described NO-3 (mg kg-1) 73.10 ± 550 57.20 ± 3.62 above. Annually in spring (2015− 2017), we evenly applied 50 kg of PO-3 4 (mg kg ) -1 11.22 ± 4.24 9.45 ± 4.51 solid mineral fertilizer to the 1292 m2 of the NHF treatment, which comprised 55 kg ha-1 y-1 of N (2.9 % NO3, 10.9 % NH4, 6.2 % NH3), delimited study area of about 5170 m2 containing 36 olive trees 8.2 kg ha-1 y-1 of P (10 % P2O5), 16 kg ha-1 y-1 of K (10 % K2O), and (Fig. 1b). Prior to establishment of the treatments, soil samples were 4.7 kg ha-1 y-1 of Mg (2 % MgO) (García, 2009), to soils between rows of taken at 1, 3, and 5 m from six randomly selected olive trees, where olive trees and under the tree canopies. This allowed to standardize the three to four 15-cm deep soil cores were collected using a gouge auger mineral fertilizer application and to disentangle the effect of the fertil­ (2.5 cm diameter). Cores were divided into 0–5 and 5–15-cm soil depths ization from the effects of other factors of covariation on soil variables and then mixed to create single composite samples by depth and sam­ while keeping a feasible number of soil sampling points. Fertilization pling point (n = 6 trees x 3 distances x 2 depths = 36 soil samples). Soil was synchronized with rainfall events when possible. samples were air dried, sieved to 2-mm gauge, and stored in dry con­ ditions prior to elemental analyses. 2.4. Soil microclimate variables Total C and total nitrogen (N) concentration were determined by combustion at 850 ºC (Leco Tru Spec autoanalyzer, St Joseph, MI, USA). Soil temperature (at 5-cm depth) and moisture (at 25-cm depth) were Soil organic C concentration was determined from reduced potassium monitored every 30 min, from February to October 2017, at 1-m dis­ dichromate (Cr2O7K2) with Mohr salt (Fe(NH4)2(SO4)2 6 H2O), tance from six randomly selected olive trees in each of the H and NH following oxidation under excess Cr2O7K2. Soil organic matter concen­ treatments using sensors (S-TMB-M002 and 10HS smart sensor S-SMD- tration was calculated as the product of the soil organic C concentration M005, for soil temperature and moisture, respectively; Onset Computer and the factor of 1.724 (Tyurin, 1951). Dissolved organic C was Corporation, Bourne, USA). extracted using the Ghani method (Ghani et al., 2003) and concentration was determined using the Tyurin method (Tyurin, 1951). Concentration 2.5. Litter decomposition of soil calcium carbonate (CaCO3) was determined as the amount of carbon dioxide emitted during the reaction of CaCO3 with hydrochloric In April 2015, six nylon mesh bags (2 mm in diameter; hereafter acid (HCl) (Bascomb, 1961). Available inorganic phosphorous (P) con­ referred to as litter bags) containing about 15 g of senescent olive leaves centration was determined in sodium bicarbonate extracts using the collected from olive trees were placed 1 m from the base of each olive Olsen method (Watanabe and Olsen, 1965) and measured using a Perkin tree in all three treatments (n = 9 replicates x 6 sampling times x 3 Elmer 2400 spectrophotometer (Waltham, MA, USA). Concentration of treatments = 162 olive litter bags). Similarly, four litter bags containing nitrite, nitrate, and phosphate anions was determined using ionic about 15 g of fresh aboveground herbaceous plant material were placed chromatography (DX-120, Dionex, Sunnyvale, USA) with an anionic under each olive tree in H (n = 18 replicates x 4 sampling times = 72 suppresser (ASRS-II, Dionex, Sunnyvale, USA). Concentrations of cal­ herbaceous litter bags). We included more litter bag replicates for her­ cium (Ca), magnesium (Mg), potassium (K), and sodium (Na) cations baceous litter since they were composed by a more heterogeneous ma­ were determined from displacement with ammonium acetate using the terial. Litter bags were sampled more frequently (every 5 weeks) in the ammonium acetate method (Soil Conservation Service, 1972). Cation early stages of litter decomposition than in the later stages (every 3− 6 months) as decomposition slows down at later decomposition stages. 4 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 The decomposition period for herbaceous litter was one year, whereas 2.10. Chemical analysis of plant material for the more recalcitrant olive tree leaves was 1.5 years (Rodríguez-­ Pleguezuelo et al., 2009). On each sampling occasion, a single olive leaf Samples of olive and herbaceous litter, olive tree litterfall, and fresh and herbaceous litter bag was carefully placed inside a paper bag and olive tree leaf material were digested in a microwave oven and then transported to the laboratory; then, the leaf and litter material were extracted using 2 % HCl. We determined Ca, Mg, K, manganese (Mn), carefully removed from the bags and dirt was removed using a small iron (Fe), copper (Cu), and zinc (Zn) concentrations using atomic- brush. The leaf and litter were dried to a constant weight at 60 ºC and absorption spectrophotometry. We analysed P concentration using the then ground using a ball mill prior to elemental analysis. V/UV spectrophotometry nitro-molybdovanadate method. Concentra­ tions of N and C were quantified using an elemental analyzer (LECO TruSpec CN 2.4., St Joseph, MI, USA). Reported nutrient concentrations 2.6. Herbaceous biomass are based on dry weight. Herbaceous biomass was sampled annually during peak growing 2.11. Calculation of nutrient stocks and nutrient release season in spring 2015, 2016, and 2017 from single 100 × 25 cm quad­ rats placed at 1, 3, and 5 m from the base of olive trees to account for Calculation of annual olive production and harvested litterfall per spatial variability in herbaceous vegetation cover in the H treatment unit area was based on the olive yield and litterfall per tree in a (n = 9 replicates x 3 distances x 3 years = 81 herbaceous biomass 12 × 12 m area that equates to the average distance between olive trees samples). Aboveground vegetation was non-existent or negligible in the in a row. Crown area of olive trees did not vary with management NH and NHF treatments, so was not sampled. All aboveground vegeta­ treatment, so there was no need to use this as a correction factor in the tion in the quadrats was clipped and taken to the laboratory, where it calculation of olive production. Calculation of annual herbaceous was dried to a constant weight at 60 ºC, cleaned with a small brush, biomass production was based on the dry weight of herbaceous biomass weighed, and then ground using a ball mill prior to elemental analysis. per unit area. In order to account for the spatial variability of the her­ baceous vegetation cover, samples collected at 1, 3, and 5 m from each olive tree were considered representative of concentric circular areas 2.7. Olive tree litterfall representing the 13 %, 31 % and 56 % of the 12 × 12 m area, respec­ tively. Weighted values of herbaceous biomass per unit area were then Two litter traps, comprising two 22.5 cm diameter plastic pots, were summed to obtain the total herbaceous biomass in the 12 × 12 m area. hung in different branches of each of the nine olive trees per treatment Annual olive tree litterfall by crown area was calculated with the annual from June 2016 to June 2017. Pots were positioned at randomized cumulative amount of litterfall per tree. Annual nutrient stocks were horizontal distances from the tree trunk and hanged below the olive tree calculated as the product of biomass stock and the elemental concen­ canopy. Pots were also leveled horizontally with the soil surface. The trations of the corresponding plant material. The annual release of C and base of the litter traps was perforated to prevent water accumulation. On nutrients from aboveground herbaceous vegetation and olive tree leaf each sampling occasion, collected litter was emptied into a paper bag litter was calculated based on the annual nutrient stocks and the rates of that was then carefully transported to the laboratory. Sampling fre­ C and nutrient released (100 - % element remaining after one year of quency (every 3 months during summer and autumn and every 6 months decomposition) in the correspondent decomposing litter (i.e. above­ during winter and spring) reflected seasonal litterfall rates. The olive ground herbaceous vegetation or olive tree litter). tree litterfall was then dried to a constant weight at 60 ºC, cleaned with a small brush, weighed, and then ground using a ball mill prior to 2.12. Statistical analysis elemental analyses. Olives, if present, were discarded prior to weighing and grinding. Main effects of management on tree crown area, olive production (yield), olive tree litterfall, harvested olive tree litterfall, elemental 2.8. Olive tree crown area, olive yield and harvested litterfall composition of harvested olive tree litterfall (Appendix 2), amount of C and nutrient stocks, and C and nutrient released by olive tree litter, and Olive tree crown area was measured at satellite images of high- final concentrations of single elements in olive tree leaf litter and fresh resolution using Google Earth pro (Google LLC., Mountain View, leaves were tested using one-way ANOVAs, with management as a fixed EEUU). Harvesting of olives was performed as usual, using a combina­ factor. When P-values where lower than 0.05, within-treatment effects tion of manual stirring and specialized machinery, and annual olive were analyzed using post-hoc tests, with Tukey correction for multiple yield was determined in November 2016 and 2017 as the fresh weight of testing. The effect of management on the averages of daily soil tem­ biomass harvested per tree (olive fruit plus leaves), minus leaf litter that perature ranges, daily minimum, maximum and daily mean soil tem­ fell during harvesting operations (hereafter referred to as harvested peratures across days during the senescent and growing periods for litterfall). Fresh weight of harvested litterfall per olive tree was also herbaceous covers was similarly tested using one-way ANOVAs, with recorded in 2016 and a subsample was collected and processed, as management as a fixed factor. The effect of treatment on olive produc­ described for olive tree litterfall, to determine moisture content and for tion and its variation along the time was also explored by repeated elemental analysis. The determination of nutrient stocks and fluxes of measures ANOVA split-plot design, with treatment as main fixed factor tree pruning residues would have required a monitoring period that between subjects and time and its interaction with treatment as factors exceeds the duration of this project as well as our work capacity and within subjects. Differences in elemental composition between herba­ economic budget. ceous and olive tree leaf litter were tested using one-way ANOVAs, with litter type as a fixed factor. The proportion of litter dry weight remaining after the period of 2.9. Nutritional status of olive trees decomposition ( %) by litter type and treatment was fitted to an expo­ nential model, with the form: At the beginning of the experiment (spring 2015) and 2 years after Xt = X0 e-kt (1). the establishment of the treatments (spring 2017), fresh olive tree leaves where Xt is the proportion of litter dry weight remaining at time t, X0 were sampled from the uppermost shoot tip that had elongated during is the initial proportion of litter dry weight (100 %), and k is the the preceding year. Leaves were processed, as described above, prior to instantaneous decay rate. Effects of management treatment on the decay elemental analysis. rates of olive tree leaf litter were also tested for individual k values for 5 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 each treatment and replicate using a one-way ANOVA, with manage­ September (Fig. 2b). The averages of daily mean, minimum and ment as a fixed factor. maximum soil temperatures across days did not differ significantly. Principal component analyses (PCA) were performed for the Nonetheless, the means of daily ranges of soil temperatures were larger elemental composition of herbaceous and olive tree leaf litter at the in NH compared to H both during the senescent period (P = 0.0168) and beginning and end of the decomposition periods, using standardized during the vegetative growing period for herbaceous plants variables. Differences between initial and final stages of decomposition, (P = 0.0221). This was caused mainly by higher values (although not and among management treatments for olive tree leaf litter, according to significant) of minimum daily soil temperatures in H during the senes­ the first two component scores of the PCAs were tested using one-way cent period (Fig. 2c) and by lower values (although not significant) of ANOVAs, as well as Tukey HSD post-hoc tests for management effects maximum daily soil temperatures in H during the vegetative growth (Jolliffe, 2002; Quinn and Keough, 2009). The nutritional status of olive period (Fig. 2d). trees was explored using a PCA of the elemental composition of olive Olive tree crown area, annual litterfall, and annual litterfall during tree leaves sampled in spring 2017 (two years after the establishment of harvest did not differ among treatments. However, olive production the management treatments), using standardized variables. The effect of (olive fresh weight) was affected by management particularly during the the agricultural management on the nutritional status of olive trees was second year, where production was lower in H than NH and NHF then tested for the first two component scores of the PCAs using one-way (Table 2). Overall, the year of harvest did not affect olive production, ANOVAs and Tukey HSD post-hoc tests. The assumptions of normality although the production tended to decrease more in H from 2016 to and homoscedasticity were tested before ANOVAs and data were 2017 (P = 0.0015 for the interaction of year*treatment). transformed when required to achieve these assumptions (Quinn and Keough, 2009). Statistical analyses and models construction were per­ 3.2. Nutrient stocks formed using JMP®, Version 16.0 (2022). Aboveground herbaceous biomass represented an important pool 3. Results and potential annual input of C and nutrients for the olive grove soil, and accounted for the greatest organic input of C (52 %), N (71 %), P (81 %), 3.1. Microclimate and stand variables K (89 %), Ca (46 %), Mn (54 %), Fe (58 %), Mg (57 %), and Zn (58 %) in the olive grove in the H treatment (Fig. 3). Moreover, herbaceous Although overall mean daily soil temperature in the upper 5 cm of biomass contained 136 %, 110 %, 96 %, and 6 % of the annual amount of soil did not differ visually between areas with (H) or without (NH) Mg, P, N, and K applied in inorganic fertilizers, respectively. Olive tree herbaceous cover, they were seemingly higher in H from the beginning litterfall also represented a large potential input of C and nutrients, and of September to mid-October (Fig. 2a). In general, soil moisture did not accounted for the largest organic input of Cu (79 %) in the H treatment. show substantial differences between H and NH visually, but lower Stocks of N, K, Mn, Fe, and Cu in olive tree litter that fell during har­ levels of moisture were recorded in H after rain events from July to vesting operations were greatest in NHF. As a result, total C, P, K, Ca, Fig. 2. Soil microclimatic variables in treatments with (H) and without herbaceous cover (NH) from February to October 2017. A. Mean daily soil temperature (5 cm depth) and air temperature. B. Mean daily soil moisture (25 cm depth) and precipitation. C. Diurnal cycles of soil temperature during herbaceous cover senescence periods. D. Diurnal cycles of soil temperature during herbaceous cover vegetative growth periods. Diurnal cycles are the averaged soil temperatures measured every 30 min for each treatment and period. Lines represent the mean of n = 6 sensors per treatment. Grey areas indicate the senescence period. Color shaded areas around lines represent standard deviation. 6 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Table 2 Effects of management treatments on main stand variables in the olive grove. Values are means ± standard errors. Different letters above means indicate significant differences among treatments according to Tukey post-hoc test, following one-way ANOVA. F: Value of the statistic; P: P-value. H: presence of herbaceous cover; NH: absence of herbaceous cover; NHF: absence of herbaceous cover and addition of solid mineral fertilizer. Stand variable Treatment F P H NH NHF Tree crown area (m2) 27.2 ± 2.2 26.0 ± 2.44 25.1 ± 2.5 0.19 0.8252 Olive production, 1st year (kg ha-2 y-1) 3463 ± 571 4151 ± 313 3744 ± 311 0.69 0.5120 Olive production, 2nd year (kg ha-2 y-1) 2019 ± 521a 4649 ± 554b 5060 ± 684b 7.34 0.0034 Herbaceous biomass production (kg ha-1 y-1) 2983 ± 244 N.A. N.A. N.A. N.A. Olive tree litterfall (kg ha-1 y-1) 1969 ± 229 2045 ± 277 1935 ± 274 0.05 0.9537 Harvested olive tree litterfall (kg ha-1 y-1) 322 ± 51 388 ± 32 460 ± 30 3.14 0.0613 Mn, Fe, Mg, Cu, and Zn pools in aboveground vegetation inputs in the H herbaceous litter were particularly important, and represented 70 %, 79 treatment exceeded the total inputs of these elements in mineral %, 575 %, and 61 % of the nutrients provided by fertilizers in NHF. In fertilizer-treated NHF plots. contrast, there was a net retention of Ca, Mn, Fe, Cu and Zn in decom­ posing litter. In general, C and nutrient releases from olive tree leaf litter (harvested litter and litterfall) were greatest in H and lowest in NHF. 3.3. Nutrient release Specifically, immobilized Ca and Mn in H was 42 % and 65 % that in NHF, while immobilization of Cu in H was 3.4 times higher than in NHF. There were clear differences in initial elemental composition be­ tween herbaceous litter and olive tree leaf litter, where olive tree litter 3.4. Nutritional status of olive trees contained greater concentrations of C, Ca, Cu, Mn, and C:N, and N:P and C:P ratios, lower concentrations of N, P, Mg, K, and Zn (P < 0.0001), and After two years, the olive grove management affected the nutritional greater Fe concentrations (P < 0.05) (Table 3). status of olive trees (P = 0.0004, Fig. 7), according to the second Litter type and quality had clear effects on decomposition rates and component of the PCA. Foliar N concentrations in olive trees were nutrient release (Table 4, Fig. 4), where there were faster rates of lowest in the presence of herbaceous vegetation (P < 0.0001, Appendix decomposition and release of C, N, P, K, and Mg, and lower retention of 3). Phosphorous concentrations followed the same trend, but effects Zn from herbaceous litter after the first year of decomposition. In were on the borderline of significance (P = 0.048). In contrast, there contrast, herbaceous litter retained more Cu than olive tree litter. The were no differences on olive tree nutritional status between NH and NHF agricultural management also affected olive tree litter decomposition treatments (Fig. 7). rates (P < 0.01). The addition of solid mineral fertilizers (NHF) decel­ erated decomposition rates of olive tree litter to a half-life of 6.31 years (Table 4, Fig. 4a), whereas the presence of herbaceous vegetation (H) 4. Discussion accelerated the process (half-life = 2.64 years). The effects of manage­ ment were also reflected in the release and retention rates of most nu­ Allowing spontaneous growth of vegetation in olive groves repre­ sents a common agricultural practice that prevents soil erosion and trients, where there were greater rates of release and lower rates of retention in the presence of herbaceous vegetation (Fig. 4). improves soil fertility. This study reveals, for first time, both direct and indirect effects of this practice on litter decomposition. Results here In general, decomposing litter released progressively C, P and K (Fig. 4), and herbaceous leaf litter was also a source of N and Mg to the presented also allow the evaluation of the biogeochemical relevance of the herbaceous cover in terms of magnitude of released macro- and soil. In contrast, other meso- and micronutrients, such as Ca, Mn, Fe, Cu, Zn, as well as N and Mg in the case of olive tree litter, were retained in micronutrients relative to those provided by fertilizers and to other aboveground plant reservoirs in the agroecosystem. Moreover, this the litter during the decomposition process. Peaks in nutrient retention coincided with periods of greater water availability in early spring and study allows comparison of effects of mineral fertilizer inputs with those of the presence of herbaceous cover on olive yield, nutritional status of autumn at the end of the decomposition period. Consequently, elemental composition of litter differed between the beginning and end olive trees and biogeochemical functioning of an olive grove agro­ ecosystem. Leaving the herbaceous vegetation in situ (H treatment) points of the measured decomposition process (after 18 and 12 months for olive tree and herbaceous litter, respectively), according to the first contributed to the retention of a high nutrient capital in organic forms component of the PCAs (P < 0.0001, Fig. 5). Concentrations of C and K (Fig. 3) and to faster litter decomposition rates (Fig. 4) than where and C:N ratios in olive tree leaf litter decreased (P < 0.0001), concen­ herbaceous vegetation had been excluded (NH and NHF), indicating a trations of Mg and P and ratios of C:P remained stable, and concentra­ stimulation of microbial activity in H. However, diminished olive pro­ tions of the remaining nutrients and N:P ratios increased (P < 0.01, duction and poorer nutritional status of olive trees in H compared to the Fig. 5a, b) with decomposition. Regarding herbaceous litter, concen­ other treatments were observed as negative short-term effects. In trations of N, P, and K decreased (P < 0.0001), concentration of C and N: contrast, the addition of mineral fertilizers in solid form (NHF) resulted P ratios remained stable, and concentrations of the remaining meso- and inefficient to improve olive tree nutritional status (Fig. 7) and olive micronutrients and C:N and C:P ratios increased (P < 0.01, Fig. 5c, d). production (Table 2) and decelerated rates of litter decomposition and Final nutrient concentrations in olive leaf litter also differed among the nutrient release (Table 4; Figs. 4 and 5). The relative magnitudes of olive grove management treatments (P = 0.02, according to PCs 1 and 2, aboveground C and nutrient stocks and fluxes in leafy litter provided Fig. 5a), where concentrations of P, K, Mn, and Fe were greater in NHF evidence for two contrasting ecological resource acquisition strategies in than in H (P < 0.01), but concentrations of C, and ratios of C:N, C:P, and herbaceous vegetation and olive trees. N:P were lower (P < 0.01). Total C and K released by aboveground leafy plant biomass in the H 4.1. Nutrient stocks treatment was about 6.6 and 3.7 times greater, respectively, than the total amount of these elements released in the mineral fertilizer-treated Aboveground herbaceous biomass represented the largest potential NHF plots (Fig. 6). Releases of N, P, K, and Mg from aboveground input of C and nutrients in organic form to the soil of the olive grove, 7 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 3. Effects of management treatments on total annual aboveground capital of carbon and nutrients in litterfall (olive tree litterfall), herbaceous litter (litter from herbaceous vegetation), in mineral fertilizer and in harvested olive tree litter as an additional potential input measured from April 2015 to October 2017. Different letters indicate treatment differences according to Tukey post-hoc tests, following one-way ANOVA. Values are means ± standard errors. Table 3 Table 4 Initial nutrient concentrations in olive tree and herbaceous leaf litter. Values are Effect of management treatments on the decay rate (k) of litter dry weight (olive means ± standard errors. F: Value of the statistic; P: P-value. tree leaves and aboveground herbaceous biomass), half-life (0.693/k) and Nutrient Herbaceous leaf Olive tree leaf F P turnover time (1/k). Estimates ( ± standard errors) were obtained by nonlinear concentration litter litter regression with Equation 1. Model SE and adjusted R2 values are reported as measures of goodness of fit. H: presence of herbaceous cover; NH: absence of C (%) 42.08 ± 0.09 48.15 ± 0.06 3391.82 < 0.0001 herbaceous cover; NHF: absence of herbaceous cover and addition of solid N (%) 2.73 ± 0.1 1.50 ± 0.01 139.72 < 0.0001 mineral fertilizer. C:N 15.62 ± 0.58 32.12 ± 0.19 740.67 < 0.0001 N:P 7.74 ± 0.32 9.99 ± 0.14 40.76 < 0.0001 Treatment k (year- Half-life Turnover time Model Adjusted P (mg kg-1) 3547.39 ± 66.50 1503.47 863.53 < 0.0001 1 ) (year) (1/k, year) SE R2 ± 20.38 Olive tree litter Ca (mg kg-1)† 10420.32 22021.90 234.33 < 0.0001 ± 519.36 ± 375.95 H 0.26 2.64 3.81 ± 0.38 9.00 0.64 Mg (mg kg-1) 2432.54 ± 91.27 1788.61 42.23 < 0.0001 ± 0.03 ± 0.26 ± 38.58 NH 0.14 4.84 6.98 ± 1.06 7.33 0.45 K (mg kg-1)† 24765.82 6258.90 1503.74 < 0.0001 ± 0.02 ± 0.73 ± 816.07 ± 79.30 NHF 0.11 6.31 9.11 ± 1.70 8.15 0.31 Fe (mg kg-1) 110.75 ± 9.86 134.16 ± 5.32 4.36 0.048 ± 0.02 ± 1.18 Mn (mg kg-1) 38.64 ± 1.93 53.12 ± 0.77 48.69 < 0.0001 Herbaceous litter Zn (mg kg-1) 21.22 ± 0.76 12.74 ± 0.31 107.14 < 0.0001 H 1.00 0.69 1.00 ± 0.08 14.26 0.70 Cu (mg kg-1) 13.28 ± 0.82 130.55 ± 4.54 644.72 < 0.0001 ± 0.08 ± 0.06 †Log-transformed data prior to ANOVA testing 8 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 4. Effects of management treatments on temporal changes in remaining dry matter and nutrient content in decomposing leaf litter from olive trees and litter from herbaceous vegetation from April 2015 to October 2016. Herbaceous vegetation was only present in H, whereas potential inputs from olive tree leaf litter to the soil were present in all three treatments (H, NH, and NHF). Lines in the figure of remaining litter dry weight represent the fitted exponential models for each litter type and treatment (see text for more details). Values are means ± standard errors (n = 18 and 9 for herbaceous and olive tree leaf litter, respectively). except for the case of Cu (Fig. 3). Moreover, this pool contained 136 %, deficiency represents the major nutritional disorder in olives growing 110 %, 96 %, and 6 % of the annual amount of Mg, P, N, and K applied in both in drylands and on calcareous soils, due to its interaction with inorganic fertilizers. This is particularly relevant for the case of N, P and water shortage and calcium, respectively (Parra et al., 2003; Restre­ K, which are essential nutrients for plant nutrition and are frequently ́ et al., 2008). Moreover, the high mobility of K in soils with low po-Diaz limiting in agriculture due to the removal of these nutrients in the clay and organic matter content and of NO-3 render them highly sus­ harvest (Ferná ndez-Escobar, 2019). Herbaceous covers composed by ceptible to leaching (Sardans and Peñuelas, 2015). Phosphorous can also leguminous N fixing species, as was the case in this study, may precipitate with Ca and be sorbed by Fe- and Al-compounds of clay contribute particularly to increase the soil N inputs and trees nutritional minerals (Addiscott and Thomas, 2000), and low levels of precipitation status (Rodrigues et al., 2015; Lombardo et al., 2021). Potassium limit microbial and plant ability to uptake P in Mediterranean and 9 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 5. Principal Component Analyses (PCA) of the effects of management treatments on nutrient concentration in olive leaf litter (A and B) and herbaceous litter (C and D) during the decomposition period (April 2015 to October 2016). Symbols in A and C represent the subject scores: initial litter samples (April 2015) are indicated by black circles and final litter samples (October 2016) are indicated by colored triangles. The arrows in B and D represent the variable scores. Percent contribution of the first two principal components to the total variability is shown in parentheses. Herbaceous litter was only present in H. semiarid regions (Belnap, 2011). In Mediterranean ecosystems, the (Almagro et al., 2010). Irrigated olive trees count on less efficient stra­ accessibility of soil nutrients may be also impaired by low moisture tegies to overcome water stress and generally have higher crown volume conditions, while nutrients can be rapidly mobilized in response to rain and leaf area index than rainfed olive trees (Fernández, 2014). Partic­ events (Nielsen and Ball, 2015). The high N, P and K storage capacity in ularly low precipitations during the collection period (282 mm) herbaceous vegetation may contribute to the retention of these nutrients compared to the mean annual precipitation in the area (466 mm) may in the agroecosystem following rapid pulses of high availability. The explain the elevated litterfall values found in this study and the partic­ large nutrient pool retained in herbaceous plant biomass implies that a ularly low yields during that year in the area. On the other hand, litter lower proportion of these nutrients are leached out or adsorbed in soil nutrients exported from the system during harvest operations repre­ minerals. However, more rapid assimilation of these nutrients by her­ sented 3.5–33.3 % of the nutrients contained in annual olive tree lit­ baceous vegetation may result in their reduced availability for olive tree terfall. Exported stocks of N, K, Mn, Fe and Cu were highest in NHF as a nutrition (Lipecki, Berbeć, 1997), at least in the time frame of two years, result of marginally greater amounts of harvested olive tree litter as it is shown in the H treatment in this study in comparison with NH and biomass (Table 2) and higher litter concentrations of K, Mn and Fe NHF treatments. (Appendix 2). Olive tree litter is an effective plant residue mulch that Olive tree litterfall also represented an important potential input of C increases supply of organic matter to the soil (Malamidou et al., 2018), and nutrients to the soil (Fig. 3), particularly of Cu. Annual litterfall improves soil microbial activity and biogeochemical cycling, contributes values in this study (Table 2) about 3.7 times higher than the annual to weed control, and reduces negative impacts associated with herbicide litterfall reported in a rainfed olive grove in drier region (370 mm) use, such as pollution and soil erosion (Saavedra Saavedra, 2007a). 10 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 6. Effects of management treatments on annual releases of carbon and nutrients from litterfall (olive tree litterfall), herbaceous litter (litter from herbaceous vegetation), mineral fertilizer and from harvested olive tree litter measured from April 2015 to October 2017. Negative values indicate a net nutrient retention. Different letters indicate treatment differences according to Tukey post-hoc tests, following one-way ANOVA. Values are means ± standard errors. Therefore, the return of harvested olive tree leaf litter to olive grove retention of elements such as Ca, Mg, Mn, Fe, Cu, and Zn was greater agroecosystems and the maintenance of herbaceous cover may represent during wet periods in spring and autumn, while C, N and the highly appropriate approaches to compensate or reduce, at least partially, the mobile K were released in a continuous and progressive way throughout nutrient withdrawal that occurs during olive harvesting operations. the decomposition period (Fig. 4). These variations may reflect abiotic adsorption of metal elements on humified litter and microbial immo­ bilization of other elements (Rustad and Cronan, 1988). The herbaceous 4.2. Nutrient release litter function on Cu immobilization may be particularly relevant in organic olive groves, due to the feared effects of copper accumulation in Despite the large nutrient stocks contained in herbaceous biomass soil (Vitanovic, 2012). Olive grove management affected the decom­ and olive tree litter, not all elements were released to the soil after one position rates of olive tree leaf litter, where the presence of herbaceous year of decomposition. Litter decay rates of olive tree litter in this vegetation indirectly stimulated decomposition, likely through Mediterranean system were particularly slow (k = 0.11 − 0.26) enhanced microbial activity and ameliorated microclimate conditions compared with those in subtropical agroecosystems (k = 1.18); (Rodrí­ (Fig. 2). This result contrasts with observations in a short-term experi­ guez-Pleguezuelo et al., 2009), probably due to the much drier condi­ ment (<60 d), in which there were increases in microbial biomass and tions at our study site. As reported elsewhere (Gómez-Muñoz et al., respiration following application of glyphosate (Nguyen et al., 2016). 2014; Rodríguez-Pleguezuelo et al., 2009), we found that litter quality Improved microclimatic conditions created by herbaceous cover and strongly determined rates of litter decomposition and nutrient release, higher organic matter inputs from vegetation may have masked any where N-rich herbaceous leaf litter, with low C:N ratios, decomposed short-term stimulation in microbial activity following the use of herbi­ more rapidly than sclerotic olive tree leaf litter with high C:N ratios and cides in the NH and NHF in our experiment. By contrast, litter decom­ greater lignin content (Table 4, Fig. 4). Additionally, temporal variations position and nutrient release was slower in absence of herbaceous nutrient release from litter differed among nutrients; for example, 11 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Fig. 7. Principal Component Analyses (PCA) of the effects of management treatments on carbon and nutrient concentrations of olive tree leaves after two years (from spring 2015 to spring 2017). Symbols in A represent the subject scores, where final leaf samples (spring 2017) are indicated by colored triangles. The arrows in B represent the variable scores. Percent contribution of the first two principal components to the total variability is shown in parentheses. vegetation cover and with application of mineral fertilizers (NHF). enhanced nutrient release from olive tree leaf litter, contributing to Greater availability of mineral forms of nutrients in the superficial soil of more rapid releases of C and P and reduced immobilization of N, Ca, Mn, NHF may have reduced the stoichiometric demand of microbes for nu­ Fe, Mg, Cu, and Zn in olive tree leaf litter (Fig. 6). Herbaceous vegetation trients during the first stages of litter decomposition (Soong et al., 2020), is also a key source of C and organic substrates for soil microbes, leading to reductions in the microbial nutrient mining and slowing down contributing to soil mineralization activity and fertility and biogeo­ litter decomposition rates (Ramírez et al., 2012). Inorganic nitrogen also chemical cycling (Herencia, 2015; Morugán-Coronado et al., 2020). inhibits lignin-degrading enzymes and the growth of lignolitic fungi Despite releases of C from decomposing herbaceous litter, the presence (Edwards et al., 2011), explaining the slower degradation of olive leaf of herbaceous vegetation has been shown to increase net C sequestration litter with high lignin content. and reduce the C footprint of olive grove agroecosystems (Palese et al., Overall, herbaceous litter was the largest aboveground plant source 2013; Chamizo et al., 2017). of N, P, K and Mg, and it released 70 %, 79 %, 575 %, and 61 % of the Despite releases of nutrients from herbaceous vegetation and the amount of these nutrients provided annually by mineral fertilizers, indirect stimulation of olive tree leaf decomposition rates, the mainte­ respectively (Fig. 6). Due to its rapid growth rate, herbaceous vegetation nance of herbaceous cover did not improve the nutritional status of olive may be highly efficient in the acquisition of nutrients when they are trees. Instead, olive leaves contained lower levels of N and marginally provided in isolated pulses, such as in response to rain events or to the lower levels of P in H than in NH and NHF (Fig. 7 and Appendix 3), application of mineral nutrients as inorganic fertilizer. This rapid falling behind the sufficiency thresholds of 1.5 % and 0.1 %, respectively growth trait allows herbaceous vegetation to intercept and store highly (Ferná ndez-Escobar, 2019). Similarly, olive yields were lower in H mobile nutrients in its biomass that will then be progressively released at (Table 2). It is possible that competition between herbaceous vegetation rates more in accordance to those for acquisition in olive trees, and olive trees for water and nutrients during spring may explain these contributing to more efficient nutrient cycling. This becomes particu­ negative effects on olive tree nutritional status and yield, particularly in larly relevant in Mediterranean and semiarid regions, where most of this case where foliar N, P and K levels were deficient and where the olive groves are located. These areas are characterized by drought pe­ herbaceous covers were not mowed. Soil moisture actually showed riods, that limit the nutrient accessibility and uptake by plants, and by (although not significantly) lower minimum values during summer in an irregular precipitation distribution with isolated rainfall events, the H treatment (Fig. 2). Moreover, leaf levels of K were below the during which nutrients are rapidly mobilized (Sardans et al., 2020). sufficiency threshold of 0.8 % in all treatments (Ferná ndez-Escobar, Moreover, climate change predictions foresee more irregular pre­ 2019), as is common in dry and calcareous soils due to its interaction cipitations (Vicente-Serrano et al., 2014), exacerbating the irregular with water and calcium (Parra et al., 2003; Restrepo-Diaz ́ et al., 2008). patterns of water and nutrient accessibility for plants in these regions. Potassium plays an important role in the regulation of water status in the Nutrient immobilization in plant biomass therefore may buffer nutrient olive (Arquero et al., 2006; Ferná ndez-Escobar (2019). Potassium defi­ availability, by preventing nutrient leaching and sequestration in less ciency may have increased the vulnerability of olive trees to water available mineral forms during pulses of high-availability, while it fa­ scarcity and made olive trees less competitive with herbaceous plants for cilitates the progressive release of nutrients as organic matter de­ this resource. In addition, K uptake is restricted by both leaf K deficiency composes (Gómez-Muñoz et al., 2014). The presence of micronutrients and water stress (Restrepo-Diaz ́ et al., 2008), which may explain the in organic forms may also improve the capacity for their acquisition in inefficiency of solid mineral fertilization to increase leaf K levels in this olive trees and reduce immobilization in soil minerals (Chatzistathis case (Appendix 3). The preservation of herbaceous covers, however, has et al., 2017). demonstrated to contribute to soil protection and fertility in degraded The slow rates of decomposition and nutrient release from recalci­ soils or in highly vulnerable soils to degradation, such as in areas of trant olive tree leaf litter may limit the use of this litter for short-term medium to steep slopes (Vicente-Vicente et al., 2016; Gómez et al., soil nutrient correction, but its use may be more appropriate to ach­ 2009a; Soriano et al., 2014). The average slope of the olive orchards in ieve long-term increases in soil fertility (Rodríguez-Pleguezuelo et al., Andalusia is 8–16 %, where around 0.5 million hectares are located 2009). Nonetheless, the presence of herbaceous vegetation indirectly under soils > 15 % slope, and 72,000 ha are located in > 30 % of slope 12 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 (Unidad de Prospectiva de la Consejeriá de Agricultura y Pesca de la effects of competition for these resources between herbaceous vegeta­ Junta de Andalucia, ́ 2002). In these areas, the positive effects of the tion and olive trees. The addition of fertilizers should be sufficient to presence of herbaceous covers on soil protection and on the increase of meet the specific nutrient requirements of the olive crop to reduce un­ soil organic matter content may compensate any potential negative ef­ necessary costs and prevent nutrient leaching and eutrophication fects derived from the competition with olive trees for limiting resources problems. To achieve this, periodic soil and plant nutrient status ana­ (Vicente-Vicente et al., 2016). In these situations, the preservation of lyses are highly recommended. Mowing the herbaceous cover crops in herbaceous cover was enough to maintain or even increase the olive spring and leaving the residues on the ground as mulch can also yield without the need of additional fertilization (Palese et al., 2014; contribute to reduce the potential negative effects of the competition for Sastre et al., 2016). However, the benefits of the presence herbaceous water (Palese et al., 2014). covers on soil organic matter inputs and soil erosion might not be so Framing these results within ecological theory and principles allows relevant in highly fertile flat areas (Ferreira et al., 2013), as it is the case understanding and prediction of the outcomes of ecological relations of the olive groves in this study. between tree crops and herbaceous cover crops. These results also On the other hand, the addition of solid mineral fertilizers elicited no highlight the need to consider fertility not just as the amount of nutrient effects on the nutritional status (Fig. 7) or yield (Table 2) of olive trees. that the soil contains, but as the function of the ecosystem to provide Application of the fertilizers in our study occurred just after rain events, nutrients to plants at the right moments, and safely immobilize them which generally favors the incorporation of nutrients to the soil and when plants do not need them. This longer-term perspective on enhances the nutrient uptake by olive trees (García, 2009). Despite this, ecological functioning of agroecosystems, can assist on the decision it is likely that a high proportion of mineral nutrients, which were making of optimal agricultural practices beyond short-term maximiza­ applied as a pulse and in solid mineral form, were volatilized before tion of productivity and contribute to ensure long-term ecological sta­ entering the soil (García, 2009), immobilized by soil microbial biomass bility and sustainability of the olive grove agroecosystem. The adoption or leached out in the soil solution before they could be taken up by olive of an organic olive grove management system, based on enhancement of trees. It is, nonetheless, also possible that these parameters may need ecological interactions among vegetation species, may also contribute to longer time periods to experience appreciable differences in response to agricultural climate change adaptation in a desertification-prone area, management changes. such as the Mediterranean basin. Olive groves with resident vegetation cover can be considered agroecosystems in which two distinct plant life-history strategies Funding coexist. Herbaceous vegetation, which comprises fast growing species with rapid turnover rates, is characterized by high rates of resource This research received funding from the European Union Seventh (nutrient and water) acquisition and nutrient release. Large fluxes in Framework Program, Ministry of Economy, Innovation, Science and energy and elemental turnover may sustain relatively low nutrient pools Employment of the Junta de Andalucía (postdoctoral fellowship of the and biomass (Gunderson and Holling, 2001). In contrast, olive tree Andalucía Talent Hub Program, Marie Skłodowska-Curie actions, growth, resource acquisition and turnover rates are much lower, but COFUND – Grant Agreement nº 291780, to SMJ), European Union Ho­ they represent much larger nutrient and total biomass pools in the rizon 2020 research and innovation program under the Marie Skło­ agroecosystem. These two divergent ecological responses to resource dowska-Curie fellowship “StoiCa”(Grant Agreement no 750252, to availability, and the duration and frequency of periods of resource SMJ), and the Spanish Ministry of Economy and Competitiveness, scarcity will likely determine the outcome of their competition. Pulses of including ERDF funds–“ELEMENTAL”– (CGL2017–83538-C3–1-R) and nutrient and water availability, such as those triggered by fertilization “INTEGRATYON3”– (PID2020–117825GB-C21). SMJ was supported by and precipitation events in Mediterranean ecosystems, will favor fast the Serra Hunter fellowship Programme from the regional Government growing r-strategist herbaceous species. In contrast, conditions of scar­ of Catalonia. ALB was supported by a Juan de la Cierva postdoctoral city will favor the maintenance of slow growth rates and survival of the contracts FJC2018–038192-I and IJC2020–045630-I, funded by MCIN/ large K-strategist, competitor olive trees thanks to their evolved adap­ AEI/ 10.13039/501100011033 and European Union NextGener­ tive structures, such as sclerotic leaves and deep roots (Sardans and ationEU/PRTR, and supported by ref. MdM-2017–0714 funded by Peñuelas, 2013). Combining the strengths and advantages of both stra­ MCIN/AEI/10.13039/50110001103. SC was supported by the HIPATIA- tegies in the management of a single crop may increase the resilience UAL postdoctoral fellowship funded by the University of Almería. and stability of the agroecosystem (Gliessman, 2007). CRediT authorship contribution statement 5. Conclusions and implications for management S. Marañón-Jiménez: Funding acquisition, Conceptualization, Data In the short-term, herbaceous vegetation may outcompete olive trees curation, Formal analysis, Methodology, Project administration, Writing for resources when they are available in short pulses, such as after - original draft, Writing - review & editing. P. Serrano-Ortíz: Funding precipitation or fertilization events. Nonetheless, the maintenance of acquisition, Conceptualization, Resources, Writing - review & editing. J. herbaceous covers may be beneficial for the sustainability of the olive Peñuelas: Funding acquisition, Writing - review & editing. A. Meijide: grove agroecosystem in the long term, improving nutrient retention and Data curation, Formal analysis, Writing - review & editing. S. Chamizo: availability, enhancing soil microbial mineralization activity and Conceptualization, Methodology, Writing - review & editing. A. López- biogeochemical cycling, increasing C sequestration and reducing Ballesteros: Data curation, Methodology, Writing - review & editing. J. nutrient losses and erosion (Milgroom et al., 2007; Gómez et al., 2009a; L. Vicente-Vicente: Methodology, Writing - review & editing. E. 2009b). These benefits are particularly significant in olive groves Fernández-Ondoño: Conceptualization, Data curation, Methodology, located in degraded areas or at medium to steep slopes, as it is the case of Resources, Writing - review & editing. most of the olive plantations in Andalusia, where the only the preser­ vation of herbaceous covers may be enough to maintain or even increase Declaration of Competing Interest the olive yield (Palese et al., 2014; Vicente-Vicente et al., 2016; Sastre et al., 2016). In other flat areas, as it is the case of this study, the organic The authors declare that they have no known competing financial management of olive groves may also benefit from a well stablished interests or personal relationships that could have appeared to influence herbaceous vegetation cover, particularly if it is composed by legumi­ the work reported in this paper. nous species, combined with a slow and continual provision of addi­ tional sources of water and organic nutrients that reduces the negative 13 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Acknowledgements Cristina Campos, Andrew Mihm, Jorge Garrido, and Juan Sánchez for their invaluable help in the laboratory and in the field, and Azram Said We are grateful to the Castillo de Canena olive company for for assistance with laboratory chemical analyses. providing the study site and field support. We thank Carmen Unquiles, Appendix 1: Variables related to the quality of the irrigation water. EC: Electric conductivity, SAR: Sodium absorption ratio Variable Value pH 8.12 EC (μS cm-1) 1641 SAR 2.847 HCO-3 (mg l-1) 157.20 CO2-3 (mg l ) -1 0.00 F- (mg l-1) 2.73 Cl- (mg l-1) 200.64 Br- (mg l-1) 0.49 NO-2 (mg l-1) 0.23 NO-3 (mg l-1) 0.89 PO3-4 (mg l ) -1 Non detectable SO2- 4 (mg l-1 ) 329.43 Li+ (mg l-1) Non detectable Na+ (mg l-1) 153.59 -1 NH+ 4 (mg l ) Non detectable K+ (mg l-1) 4.07 Mg2+ (mg l-1) 62.98 Ca2+ (mg l-1) 117.20 Sr2+ (mg l-1) 2.63 Ba2+ (mg l-1) Non detectable Appendix 2: Effects of management treatments on nutrient concentrations in harvested olive litterfall. Values are means ± standard errors. Different letters above means indicate treatment differences according to Tukey post-hoc test, following one-way ANOVA. F: Value of the statistic; P: P-value _ Treatment F P H NH NHF C (%)† 48.9 ± 0.19 48.7 ± 0.12 49.0 ± 0.17 0.8651 0.4337 N (%) 1.32 ± 0.03 1.38 ± 0.03 1.38 ± 0.03 1.3318 0.2836 P (ppm) 1194 ± 38.9a 1454 ± 48.8b 1204.5 ± 44.1a 11.1822 0.0005 C:N 37.06 ± 0.69 35.36 ± 0.65 35.70 ± 0.75 1.6276 0.2182 N:P 10.48 ± 0.67 9.75 ± 0.37 11.46 ± 0.26 3.0116 0.0698 C:P 386.6 ± 21.1a,b 338.0 ± 12.2a 411.3 ± 15.0b 4.9521 0.0163 Ca (ppm) 14769.9 ± 524.2 13051.8 ± 741.4 12883.6 ± 442.4 3.2031 0.0585 Mg (ppm) 1496.5 ± 53.8a 1339.2 ± 39.3b 1381.3 ± 34.9a,b 3.5540 0.0460 K (ppm) 4629.4 ± 210.6a 4808.6 ± 166.8a 5734.7 ± 230.8b 8.3772 0.0020 Fe (ppm) 86.23 ± 6.86a 75.15 ± 7.84a 143.09 ± 5.56b 28.5837 < 0.0001 Mn (ppm) 42.93 ± 2.00a 45.24 ± 1.01a 52.78 ± 2.15b 8.0475 0.0022 Zn (ppm) 32.68 ± 1.02 33.26 ± 1.26 36.43 ± 1.85 2.0367 0.1533 Cu (ppm)‡ 76.06 ± 5.45 74.12 ± 4.03 96.24 ± 10.65 2.1758 0.1354 †Log-transformed data before ANOVA testing ‡Inverse-transformed data before ANOVA testing Appendix 3: Effects of management treatments on the nutritional status of olive leaves of olive tree leaves after two years. Values are means ± standard errors. Different letters above means indicate treatment differences according to Tukey post-hoc test, following one- way ANOVA. F: Value of the statistic; P: P-value Nutrient concentration Treatment F P H NH NHF C (%) 48.2 ± 0.17 48.0 ± 0.16 48.1 ± 0.15 0.2163 0.8070 N (%) 1.42 ± 0.02a 1.68 ± 0.02b 1.76 ± 0.05b 27.3223 < 0.0001 P (ppm) 956.5 ± 55.2 1124.5 ± 46.0 1123.1 ± 53.6 3.4756 0.0480 C:N 34.03 ± 0.53a 28.58 ± 0.50b 27.54 ± 0.75b 33.4573 < 0.0001 N:P 15.23 ± 0.92 14.79 ± 0.87 16.01 ± 1.01 0.4238 0.6596 C:P‡ 516.9 ± 29.4 432.4 ± 18.2 438.6 ± 26.3 3.5635 0.0449 Ca (ppm)† 10631.8 ± 337.4 11014.5 ± 211.3 10385.5 ± 255.1 1.4993 0.2435 Mg (ppm) 1192.2 ± 54.9 1175.0 ± 30.4 1227.4 ± 64.2 0.2661 0.7686 K (ppm) 5865.9 ± 263.2 5883.7 ± 227.3 5864.8 ± 247.1 0.0018 0.9982 (continued on next page) 14 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 (continued ) Fe (ppm) 104.91 ± 3.98 103.57 ± 4.87 96.61 ± 6.46 0.7322 0.4913 Mn (ppm) 43.61 ± 2.42 47.39 ± 1.99 46.19 ± 2.43 0.7089 0.5022 Zn (ppm) 16.35 ± 0.89 18.50 ± 1.22 18.62 ± 0.81 1.6784 0.2078 Cu (ppm)‡ 96.72 ± 6.57 86.09 ± 3.48 91.42 ± 5.19 0.9925 0.3854 †Log-transformed data before ANOVA testing ‡Inverse-transformed data before ANOVA testing References Kairis, O., Karavitis, C., Kounalaki, A., Salvati, L., Kosmas, C., 2013. The effect of land management practices on soil erosion and land desertification in an olive grove. Soil Use Manag. 29, 597–606. Addiscott, T.M., Thomas, D., 2000. Tillage, mineralization and leaching: phosphate. Soil Koutsias, N., Arianoutsou, M., Kallimanis, A.S., Mallinis, G., Halley, J.M., Tillage Res. 53, 255–273. Dimopoulos, P., 2012. Where did the fires burn in Peloponnisos, Greece the summer Aguilera, E., Guzmán, G., Alonso, A., 2015. Greenhouse gas emissions from conventional of 2007? Evidence for a synergy of fuel and weather. Agric. For. Meteorol. 156, and organic cropping systems in Spain. II. Fruit tree orchards. Agron. Sustain. Dev. 41–53. 35, 725–737. Lipecki, J., Berbeć, S., 1997. Soil management in perennial crops: orchards and hop Almagro, M., López, J., Boix-Fayos, C., Albaladejo, J., Martínez-Mena, M., 2010. gardens. Soil Tillage Res. 43, 169–184. Belowground carbon allocation patterns in a dry Mediterranean ecosystem: a Lombardo, S., Restuccia, A., Abbate, C., Anastasi, U., Fontanazza, S., Scavo, A., comparison of two models. Soil Biol. Biochem. 42, 1549–1557. Guarnaccia, P., La Malfa, S., Pandino, G., Mauromicale, G., 2021. Trifolium Angeler, D.G., Fried-Petersen, H.B., Allen, C.R., Garmestani, A., Twidwell, D., subterraneum cover cropping for improving the nutritional status of a Mediterranean Chuang, W., Donovan, V.M., Eason, T., Roberts, C.P., Sundstrom, S.M., Wonkka, C. apricot orchard. J. Sci. Food Agric. 101, 3767–3777. L., 2019. Adaptive capacity in ecosystems. Adv. Ecol. Res. 1–24. Malamidou, T., Nikolaidou, A.E., Mamolos, A.P., Pavlatou-Ve, A.K., Kostopoulou, S.K., Arquero, O., Barranco, D., Benlloch, M., 2006. Potassium starvation increases stomatal Menexes, G.C., Anagnostopoulos, C.D., Kalburtji, K.L., 2018. Litter dynamics of Olea conductance in olive trees. HortScience 41, 433–436. europaea subsp. Europaea residues related to soil properties and microbial N- Bascomb, C.L., 1961. A calcimeter for routine use on soil samples. Chem. Ind. 45, biomass in a Mediterranean agroecosystem. Eur. J. Soil Biol. 84, 11–18. 1826–1827. Margalef, R., 1959. Mode of evolution of species in relation to their places in ecological Belnap, J., 2011. Biological phosphorous cycling in dryland regions. In: Bünemann, E.K., succession. XVth Int. Congr. Zool. Proc. 787–789. et al. (Eds.), Phosphorus in Action, Soil Biology, 26. Springer-Verlag, Berlin, Milgroom, J., Soriano, M., Garrido, J.M., Gómez, J.A., Fereres, E., 2007. The influence of Heidelberg, pp. 371–406. a shift from conventional to organic olive farming on soil management and erosion Camarsa, G., Gardner, S., Jones, W., Eldridge, J., Hudson, T., Thorpe, E., O´Hara, E., risk in southern Spain. Renew. Agric. Food Syst. 22, 1–10. 2010. LIFE among Olives: Good Practice in Improving Environmental Performance Morugán-Coronado, A., Linares, C., Gómez-López, M., Faz, Á., Zornoza, R., 2020. The in the Olive Oil Sector. Office for Official Publications of The European Union, impact of intercropping, tillage and fertilizer type on soil and crop yield in fruit Luxembourg. orchards under Mediterranean conditions: a meta-analysis of field studies. Agric. Chamizo, S., Serrano-Ortiz, P., López-Ballesteros, A., Sánchez-Cañete, E.P., Vicente- Syst. 178, 102736. Vicente, J., Kowalski, A.S., 2017. Net ecosystem CO2 exchange in an irrigated olive Nguyen, D.B., Rose, M.T., Rose, T.J., Morris, S.G., van Zwieten, L., 2016. Impact of orchard of SE Spain: influence of weed cover. Agric., Ecosyst. Environ. 239, 51–64. glyphosate on soil microbial biomass and respiration: a meta-analysis. Soil Biol. Chatzistathis, T., Papaioannou, A., Gasparatos, D., Molassiotis, A., 2017. From which soil Biochem. 92, 50–57. metal fractions Fe, Mn, Zn and Cu are taken up by olive trees (Olea europaea L., cv. Nielsen, U.N., Ball, B.A., 2015. Impacts of altered precipitation regimes on soil ‘Chondrolia Chalkidikis’) in organic groves? J. Environ. Manag. 203, 489–499. communities and biogeochemistry in arid and semi-arid ecosystems. Glob. Change Edwards, I.P., Zak, D.R., Kellner, H., Eisenlord, S.D., Pregitzer, K.S., 2011. Simulated Biol. 21, 1407–1421. atmospheric N deposition alters fungal community composition and suppresses Nieto, O.M., Castro, J., Ferná ndez-Ondoño, E., 2013. Conventional tillage versus cover ligninolytic gene expression in a northern hardwood forest. PLOS ONE 6, e20421. crops in relation to carbon fixation in Mediterranean olive cultivation. Plant Soil FAO, 2004. Carbon Sequestration in Dryland Soils. World Soils Resources Reports 102. 365, 321–335. Food and Agriculture Organization of the United Nations, Rome. Palese, A., Pergola, M., Favia, M., Xiloyannis, C., Celano, G., 2013. A sustainable model FAOSTAT, 2022. Data on Crops. http://www.fao.org/faostat/en/#data (Accessed 12 for the management of olive orchards located in semi-arid marginal areas: Some July 2022). remarks and indications for policy makers. Environ. Sci. Policy 27, 81–90. Fernández, J.-E., 2014. Understanding olive adaptation to abiotic stresses as a tool to Palese, A.M., Vignozzi, N., Celano, G., Agnelli, A.E., Pagliai, M., Xiloyannis, C., 2014. increase crop performance. Environ. Exp. Bot. 103, 158–179. Influence of soil management on soil physical characteristics and water storage in a Ferná ndez-Escobar, R., 2019. Olive nutritional status and tolerance to biotic and abiotic mature rainfed olive orchard. Soil Tillage Res. 144, 96–109. stresses. Front. Plant Sci. 10, 1151. Pansu, M., Gautheyrou, J., 2006. Handbook of Soil Analysis. Mineralogical. Organic and Ferreira, I.Q., Arrobas, M., Claro, A.M., Rodrigues, M.A., 2013. Soil management in Inorganic Methods. Springer, Montpellier. rainfed olive orchards may result in conflicting effects on olive production and soil Paredes, D., Karp, D.S., Chaplin-Kramer, R., Benítez, E., Campos, M., 2019. Natural fertility. Span. J. Agric. Res. 11, 472–480. habitat increases natural pest control in olive groves: economic implications. J. Pest García, García C., 2009. Abonado del olivar. In: Camacho García, A. (Ed.), Guía Prá ctica Sci. 92, 1111–1121. de la Fertilizació n Racional de los Cultivos en España. Ministerio de Medio Ambiente Parra, M.A., Ferná ndez-Escobar,R., Navarro,C., and Arquero,O., 2003. Los suelos y la y Medio Rural y Marino, Madrid, pp. 223–233. fertilizació n del olivar cultivado en zonas calcá reas. Mundi-Prensa, Madrid. García-González, I., García-Díaz, A., Sastre, B., Teutscherova, N., Pérez, M., Bienes, R., Quinn, G.P., Keough, M.J., 2009. Experimental Design and Data Analysis for Biologists. Espejo, R., Hontoria, C., 2019. Mycorrhizal, nutritional and virgin olive oil Cambridge University Press,, Cambridge. parameters affected by groundcovers. J. Plant Nutr. Soil Sci. 182, 815–823. Ramírez, K.S., Craine, J.M., Fierer, N., 2012. Consistent effects of nitrogen amendments Ghani, A., Dexter, M., Perrott, K.W., 2003. Hot-water extractable carbon in soils: a on soil microbial communities and processes across biomes. Glob. Change Biol. 18, sensitive measurement for determining impacts of fertilisation, grazing and 1918–1927. cultivation. Soil Biol. Biochem. 35 (9), 1231–1243. Red de Información Agroclimática de Andalucía, Junta de Andalucía, http://www. Gliessman, S.R., 2007. Agroecology: The Ecology of Sustainable Food Systems. CRC juntadeandalucia.es/agriculturaypesca/ifapa/ria/servlet/FrontController, (Accessed press, Boca Raton. 12 July 2022). Gómez, J.A., Guzmán, M.G., Giráldez, J.V., Fereres, E., 2009a. The influence of cover ́ Restrepo-Diaz, H., Benlloch, M., Navarro, C., Ferná ndez-Escobar, R., 2008. Potassium crops and tillage on water and sediment yield, and on nutrient, and organic matter fertilization of rainfed olive orchards. Sci. Hortic. 116, 399–403. losses in an olive orchard on a sandy loam soil. Soil Tillage Res. 106, 137–144. Rey, P.J., Manzaneda, A.J., Valera, F., Alcántara, J.M., Tarifa, R., Isla, J., Molina- Gómez, J.A., Sobrinho, T.A., Giráldez, J.V., Fereres, E., 2009b. Soil management effects Pardo, J.L., Calvo, G., Salido, T., Gutiérrez, J.E., Ruiz, C., 2019. Landscape- on runoff, erosion and soil properties in an olive grove of Southern Spain. Soil Tillage moderated biodiversity effects of ground herb cover in olive groves: Implications for Res. 102, 5–13. regional biodiversity conservation. Agric., Ecosyst. Environ. 277, 61–73. Gómez-Muñoz, B., Hatch, D.J., Bol, R., García-Ruiz, R., 2014. Nutrient dynamics during Rodrigues, Â.M., Dimande, P., Pereira, E.L., Ferreira, I.Q., Freitas, S., Correia, C.M., decomposition of the residues from a sown legume or ruderal plant cover in an olive Moutinho-Pereira, J., Arrobas, M., 2015. Early-maturing annual legumes: an option oil orchard. Agric. Ecosyst. Environ. 184, 115–123. for cover cropping in rainfed olive orchards. Nutr. Cycl. Agroecosyst. 103, 153–166. Gunderson, L.H., Holling, C.S., 2001. Panarchy: Understanding Transformations In Rodríguez-Pleguezuelo, C.R., Zuazo, V.H., Fernández, J.L., Peinado, F.J., Tarifa, F.D., Human And Natural Systems. Island Press, Washington, D.C. 2009. Litter decomposition and nitrogen release in a sloping Mediterranean Guzmán, G., Foraster, L., 2011. El manejo del suelo y las cubiertas vegetales en el olivar subtropical agroecosystem on the coast of Granada (SE, Spain): effects of floristic and ecológico. In: El Olivar Ecológico, Junta de Andalucía, Sevilla, pp. 51–94. topographic alteration on the slope. Agric. Ecosyst. Environ. 134, 79–88. Herencia, J.F., 2015. Enzymatic activities under different cover crop management in a Roff, D.A., 1993. Evolution of Life Histories: Theory and Analysis. Springer, Berlin. Mediterranean olive orchard. Biol. Agric. Hortic. 31, 45–52. Ruíz-Colmenero, M., Bienes, R., Marques, M.J., 2011. Soil and water conservation JMP®, Version 16.0. 2022. SAS Institute Inc., Cary, NC. dilemmas associated with the use of green cover in steep vineyards. Soil Tillage Res. Jolliffe, I.T., 2002. Principal components analysis. Springer Series in Statistics. Springer, 117, 211–223. New York. Jones, J.M., 1976. The r-K-selection continuum. Am. Nat. 110, 320–323. 15 S. Marañón-Jiménez et al. European Journal of Agronomy 140 (2022) 126597 Rustad, L.E., Cronan, C.S., 1988. Element loss and retention during litter decay in a red Tarifa, R., Martínez-Núñez, C., Valera, F., González-Varo, J.P., Salido, T., Rey, P.J., 2021. spruce stand in Maine. Can. J. For. Res. 18, 947–953. Agricultural intensification erodes taxonomic and functional diversity in Saavedra, M.M., 2007a. Soil management in olive orchards. In: Tombesi, A., Tombesi, S., Mediterranean olive groves by filtering out rare species. J. Appl. Ecol. 58, Saavedra, M.M., Fernández-Escobar, R., d’Andria, R., Lavini, A., Jardak, T. (Eds.), 2266–2276. Production Techniques in Olive Growing. International Olive Council, Madrid, Tyurin, I.V., 1951. Analytical procedure for a comparative study of soil humus. Tr. Pochr pp. 85–119. Inst. Dokuchaeva 38, 5–9. Saavedra, M.M., 2007b. Herbicide use. In: Tombesi, A., Tombesi, S., Saavedra, M.M., Unidad de Prospectiva de la Consejeriá de Agricultura y Pesca de la Junta de Andalucia. ́ Fernández-Escobar, R., d’Andria, R., Lavini, A., Jardak, T. (Eds.), Production 2002. El Olivar Andaluz. Junta de Andalucía. Consejería de Agricultura y Pesca, Techniques in Olive Growing. International Olive Council, Madrid, pp. 119–145. Sevilla. Sardans, J., Peñuelas, J., 2013. Plant-soil interactions in Mediterranean forest and Vicente-Serrano, S.M., et al., 2014. Evidence of increasing drought severity caused by shrublands: impacts of climatic change. Plant Soil 365, 1–33. temperature rise in southern Europe. Environ. Res. Lett. 9, 044001. Sardans, J., Peñuelas, J., 2015. Potassium: a neglected nutrient in global change. Glob. ́ Vicente-Vicente, J.L., Garcia-Ruiz, R., Francaviglia, R., Aguilera, E., Smith, P., 2016. Soil Ecol. Biogeogr. 24, 261–275. carbon sequestration rates under Mediterranean woody crops using recommended Sardans, J., Urbina, I., Grau, O., Asensio, D., Ogaya, R., Peñuelas, J., 2020. Long-term management practices: a meta-analysis. Agr. Ecosyst. Environ. 235, 204–214. drought decreases ecosystem C and nutrient storage in a Mediterranean holm oak Vitanovic, E. 2012. Use of Cu fungicides in vineyards and olive groves. INTECH Open forest. Environ. Exp. Bot. 177, 104135. Access Publisher. Sastre, B., Pérez-Jiménez, M.Á., Bienes, R., García-Díaz, A., De Lorenzo, C., 2016. The Watanabe, S., Olsen, R.S., 1965. Test of an ascorbic acid method for determining effect of soil management on olive yield and VOO quality in a rainfed olive grove of phosphorus in water and NaHCO3 extracts from soil. Soil Sci. Soc. Am. J. 29, central Spain. J. Chem. 20–22. 677–678. Sastre, B., Marques, M., García-Díaz, A., Bienes, R., 2018. Three years of management Whittig, L.D., Allardice, W.R., 1986. X-ray diffraction techniques. In: Klute, A. (Ed.), with cover crops protecting sloping olive groves soils, carbon and water effects on Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods. American gypsiferous soil. Catena 171, 115–124. Society of Agronomy, Madison, WI, pp. 331–362. Soil Conservation Service, 1972. Soil survey laboratory methods and procedures for Wood, S.A., Karp, D.S., DeClerck, F., Kremen, C., Naeem, S., Palm, C.A., 2015. Functional collecting soils samples. Soil Surv Report 1. USDA. Washington D.C. traits in agriculture: agrobiodiversity and ecosystem services. Trends Ecol. Evol. 30, Soong, J.L., Fuchslueger, L., Marañon-Jimenez, S., Torn, M.S., Janssens, I.A., 531–539. Peñuelas, J., Richter, A., 2020. Microbial carbon limitation: the need for integrating Zaman, M., Shahid, S.A., Heng, L., 2018. Irrigation Water Quality. Guideline for Salinity microorganisms into our understanding of ecosystem carbon cycling. Glob. Change Assessment, Mitigation and Adaptation Using Nuclear and Related Techniques. Biol. 26, 1953–1961. Springer, Cham. Soriano, M.A., Álvarez, S., Landa, B.B., Gómez, J.A., 2014. Soil properties in organic Zipori, I., Erel, R., Yermiyahu, U., Ben-Gal, A., Dag, A., 2020. Sustainable management of olive orchards following different weed management in a rolling landscape of olive orchard nutrition: a review. Agriculture 10, 11. Andalusia, Spain. Renew. Agr. Food Syst. 29, 83–91. 16