Introduction

The use of vegetable coverts is an agroecological practice that consists of using dead or living plant material, where in the case of dead material it is spread as dry grass, leaf litter or crop residues on ridges and streets, in order to protect and improve nutritionally soil (Pound, 1999).

In the case of living material, plant species considered as weeds or cultivated species of high nutritional value can be established; if weeds are used, they can be selected by applying selective herbicides, which allow the development of the cover between rows of trees in the center of the streets throughout the year (^{Norton and Storey, 1970}; ^{Martínez, 2010}; ^{Rebolledo- Martínez et al., 2011}).

Among the benefits that have been attributed to plant coverts are reducing soil losses due to erosion, increasing water infiltration, reducing evaporation, reducing the density of competitive weeds, increasing biodiversity, conserving mesofauna and allow access to the field with moist soil (Pound, 1999; ^{Tarango, 2010}; ^{Arenas et al., 2015}).

Regarding the effect of vegetation cover on the infiltration and availability of water in the soil, several authors have shown the advantage of these in the soil profile, in comparison with traditional tillage (Castro et al., 1992; ^{Gomez-Aparisi et al., 1993}).

The use of coverings has been documented in peach orchards (^{Aibar et al., 1990}), mango (^{Rebolledo-Martínez et al., 2011}), almond (^{Arquero et al., 2015}), avocado (Reyes et al., 2016), citrus fruits (^{Arenas et al., 2015}), pear (^{Gómez-Aparisi et al., 1993}) and vid (^{Klik et al, 1998}). In perennial crops that form a dense shade after five years, the cover crop is necessary only during the establishment phase, but in fruit trees of open plantations, such as citrus, mango or walnut, the coverts can be established for longer periods, using the native weeds or with the sowing of some crop species (^{Teasdale et al., 1991}; ^{Arenas et al., 2015}; ^{Tarango, 2010}).

In the walnut orchards of the coast of Hermosillo, Sonora, they develop several species of annual weeds such as chinita (Sonchus spp.), pamita (Sysimbrium irio), mustard (Brasssica spp.), pinto grass (Echinochloa spp.), huachapore (Cenchrus spp.) and salted grass (Leptochloa spp.) and perennials as corm (Convolvulus arvensis), Johnson grass (Sorghum halepense), stafiate (Ambrosia confertifolia) and grama grass (Cynodon dactylon) (^{Martínez-Díaz, 2010}). These species, if not controlled with opportunity and efficiency cause damage to the development of trees, both by the consumption of nutrients and water or by encouraging the increase of harmful biological agents such as pests and diseases; in addition, they obstruct the implementation of management practices (^{Norton and Storey, 1970}).

Live plant coverts consume water and nutrients, inputs that must be provided in addition to the needs of the crop. The mixture of the species Lolium perenne, Festuca arundianacea and Trifolium repens, was a combination that reported a low water requirement (^{Klik et al, 1998}).

Previous studies on the sowing of some species of crops in walnut orchards in the region of the coast of Hermosillo, showed that several legumes could not be established within the area of the garden, however, Bermuda grass for golf courses (Cynodon dactylon) was established permanently (^{Martínez, 2010}). Based on the aforementioned experience, the present work was conducted whose objective was to determine the effect of the weed established as a living vegetation covert in the fertility of the soil and in the performance of pecanal walnut.

Materials and methods

The present work was developed from 2007 to 2016, in the walnut orchard field “Los Fumicos”, which is located in the region of the Coast of Hermosillo (Latitude of 28° 49’ 32’’ North latitude 111° 33’ 1.44’’ West longitude, 51 masl). The climate is BW(h’) very dry and very warm (^{García, 1988}) with maximum annual average temperature of 31.8 and minimum annual average of 13°C, with extreme maximum of 48 °C in July and extreme minimum of -6.5 °C in January, relative humidity of 20 to 90% throughout the year and average annual rainfall of 175 mm. The soil is of alluvial origin with a sandy clay loamy texture, good superficial and internal drainage and slightly alkaline pH ^{Ruiz et al. (2005)}. In the orchard where the experiment was carried out, there are Western and Wichita varieties, but the work was carried out in rows of Western cv. Which have a planting frame of 10 x 10 m between rows and between plants, and irrigated with two Hose lines buried 30 cm deep. At the time of the establishment of the experiment the trees were 32 years old.

The treatments evaluated were: 1) vegetable covering with native weeds; and 2) clean control, which were distributed in a design of paired plots with four repetitions. The experimental unit consisted of a band four meters wide by 30 meters long which included three trees. Covering treatment had as dominant species water grass (Leptochloa sanguinalis) and pinto grass (Echinochloa crus-galli), which covered 80% of the surface of the established band, and other species such as Johnson grass (Sorghum halepense) and Bermuda grass (Cynodon dactylon). In the treatment with the covering the growth of the weed was allowed up to a height of 40 cm and then its pruning was done at 10 cm with a mechanical trimmer coupled to the tractor; in total a total of six cuts were required in the growing season; after the harvest of the trees, weed growth was allowed until the summer of the following year in this treatment. The glyphosate herbicide was applied to the clean control in a dose of 1 kg ha^-1 and the mixture of diuron + pendimethalin 0.4 + 2 kg ha^-1 or oxyfluorfen + pendimethalin 0.5 + 2 kg ha^-1, to eliminate the weed.

In order to measure the effect of the treatments on the physical and chemical characteristics of the soil, soil samples were taken in each of the treatments at 30, 60 and 90 cm depth at 10 years (2016) after the start of the treatment. experiment. The salinity of the samples was analyzed in saturated soil paste considering pH, electrical conductivity (C.E), sodium adsorption ratio (RAS) and interchangeable sodium percent (PSI). On the other hand, for the fertility analysis the solution extracted from the soil sample was used and the organic matter and levels of the macroelements in ppm were measured as N-NO₃, P-PO₄, K, Ca, Mg and S and microelements which include Zn⁺⁺, Fe⁺⁺, Mn⁺⁺, Cu⁺⁺, Na⁺. As regards tree, its yield was measured annually. The data of each one of the measured variables were analyzed by means of the Anova test and the comparison of means with Tukey at 0.05.

Results and discussion

Conclusions

The live vegetation covering using the weeds present in the walnut orchard, consisting mainly of annual grasses, increased the content of organic matter and most of the nutrients required by the tree, especially in the first 30 cm of the soil.

The vegetation cover did not cause interference problems with the tree since the yield was similar to the clean control.