Introduction

In the temperate-semiarid region of Aguascalientes, Mexico, most rural population depends on rainfed agriculture related to the family milk production system. However, increasing soil degradation and few crop options in that region, with greater adaptation to climate change manifested by increasingly longer dry periods (^{Núñez-López et al., 2007}), reduce yield and profitability of forage-milk production systems. Moreover, sowing traditional forage crops, such as maize and sorghum, in conventional furrows (76 to 80 cm) decreases dry matter accumulation due to the number of plants per m² (^{Bolaños-Aguilar & Emile, 2013}; ^{Reta-Sánchez et al., 2007}). This forces us to look for other forage production options, using innovative agronomic practices; that is, to implement technical and cultural changes in current processes to make water and soil resources more efficient, and to take advantage of solar energy to maintain or increase yield and quality of the forage produced (^{Cuevas-Reyes et al., 2013}; ^{Häubi-Segura & Gutiérrez-Lozano, 2015}; ^{Johannessen et al., 2001}; ^{Osuna-Ceja et al., 2015}).

Sustainable development of agriculture in the rainfed area of Aguascalientes refers to the need to minimize agricultural land degradation and decrease climate change effects, while maximizing forage production. This considers a set of agronomic practices, such as soil, water, nutrition, crop management and biodiversity conservation (^{Altieri & Nicholls, 2013}). Therefore, this system seeks sustainable production of rainfed forage for dairy cattle feeding (^{Häubi-Segura & Gutiérrez-Lozano, 2015}).

Forage crop production systems are based on intensive tillage, conventional furrow planting (76 or 80 cm) in favor of the slope, supply of external products as a strategy to increase soil and crop yields. However, intensive tillage practices generate soil degradation and compaction, loss of biodiversity, and increased runoff and erosion (^{Navarro-Bravo et al., 2008}). On the contrary, agricultural practices that conserve soil, water and nutrients, such as conservation tillage, topological arrangement of plants, rainwater harvesting, organic soil conservation and management, nutrient supply boosters in seed (biofertilizers) and those natural minerals such as zeolite with high attraction for ammonium ion, help to improve soil structure and fertility, which can enhance forage and grain production of rainfed crops (^{Osuna-Ceja et al., 2012}). These technologies can contribute to the solution of agroecological and environmental problems, and crop yield (^{Bolaños-Aguilar & Emile, 2013}; ^{Figueroa-Sandoval & Talavera-Magaña, 2012}; ^{Obregón-Portocarrero et al., 2016}).

Improved management practices in rainfed forage crops, such as sowing in seedbeds at 160 cm with four or six rows, contribute significantly to forage and grain yields (^{Osuna-Ceja et al., 2015}; ^{Osuna-Ceja & Martínez-Gamiño, 2017}). In this case, topological arrangement explores a better spacing between rows and plants, which supports better crop development and adequate management of competition for nutrients and solar radiation. Moreover, land surface is efficiently used with this system, and plant density per unit area is increased (^{Osuna-Ceja & Martínez-Gamiño, 2017}; ^{Reta-Sánchez et al., 2007}). Soil conservation practices, and rainwater harvesting and fertilization, are crucial for growth and development of rainfed crops, especially in degraded soils with low nutrient and organic matter content, limiting factors for crop growth and production efficiency (^{Arellano-Arciniega et al., 2015}).

The combined application of these agronomic practices improves rainwater use and increases biomass and some quality parameters of rainfed forages (^{Osuna-Ceja et al., 2015}). However, in the case of maize, as plant density increases, net lactation energy per kilogram of dry matter decreases, due to the reduction in digestibility as a result of lower grain content and higher fiber content of plants (^{Peña-Ramos et al., 2010}). Therefore, it is necessary to evaluate yield potential of maize forage and other alternative rainfed forage crops using integrated agronomic practices. Regarding the above, the objective of this study was to evaluate chemical, organic and biological fertilization practices, as well as their relationship with yield and forage quality of rainfed maize, sunflower and sorghum planted in four- and six-row seedbeds with minimum tillage and in situ water harvesting.

Materials and methods

Three experiments with maize, sunflower and sorghum for rainfed forage were established in 2018 in San Luís de Letras, municipality of Pabellón de Arteaga, Aguascalientes, Mexico (22° 13’ 82” N and 103° 30’ 75” W, 1 960 m a. s. l.). The varieties sown were V-209 maize from the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Victoria sunflower (also from INIFAP) and Triunfo brown rib forage sorghum, all with good yield potential. V-209 and Victoria are early-cycle and intermediate-height, while sorghum is intermediate-cycle. During the 120-day growing cycle (July to mid-October), an average of 210 mm of precipitation and an average temperature of 16.7 °C are recorded (^{Osuna-Ceja et al., 2015}). The soil in the area has a sandy loam texture with a pH of 7.9 and less than 1 % organic matter.

The experiment was established in 160 cm wide seedbeds, four rows with 30 cm spacing for maize and sunflower, and six rows with 20 cm spacing for sorghum. A density of 11 plants·m^-2 was used for maize and sunflower, and 25.2 plants·m^-2 for sorghum. The experimental unit consisted of three 10 m long seedbeds, and the same bed width was considered for all three crops. Within each experimental unit, a central bed of 8 m length was used as a useful plot for soil and plant data collection.

Five fertilization treatments were evaluated in each crop: T1) absolute control (no fertilizer), T2) chemical fertilization (40-40-00 kg·ha^-1 of N-P-K), T3) nitrogen, phosphorus and zeolite mixture (28-40-00 kg·ha^-1 of N-P-K and 26 kg·ha^-1 of zeolite), T4) organic fertilizer (5 t·ha^-1 of composted bovine manure) and T5) mycorrhiza (seed inoculation with 350 g·ha^-1 of mycorrhizal substrate). As a source of N and P, urea and triple calcium superphosphate were used, respectively. Fertilization doses were defined based on INIFAP recommendations for this region, which were applied in the three experiments. The experimental design used was a randomized complete block design with four replicates.

The field was tilled with vertical tillage using a biomimetic integral subsoiler at a depth of 0.20 m before sowing (^{Osuna-Ceja et al., 2019}). Sowing was performed under rainfed conditions and in moist soil on July 3, 2018. To establish seedbeds, a machine designed for sowing in 160 cm wide seedbeds with integrated pile-driving system was used (^{Garibaldi-Márquez et al., 2020}). Crop sowing was done by hand, as well as the application of fertilization treatments. After sowing, spaces were established for in situ rainwater harvesting by means of a corrugation system on soil surface with "Aqueel" (a method creating microreservoirs on the surface of the seedbed in a homogeneous manner for water harvesting) and "pileteo" (a practice that consists of raising 20 cm high mounds of soil on the sides of seedbeds at regular distances to store water and reduce soil erosion).

Agronomic management was intended to achieve high yield. All crops, at the time of sowing, received 100 % of N and 100 % of P for the two fertilization treatments. The dose of cow manure was applied over the entire surface of the experimental unit before tillage of the soil. For T5, seeds were inoculated with mycorrhizal substrate one day before sowing. Mechanical weeding was carried out in maize and sunflower crops 25 days after sowing (das) and hand-weeding 40 days after sowing. Three hand-weeding operations were carried out for sorghum. To control fall armyworm (Spodoptera frugiperda) in maize and sorghum, the insecticide Palgus™ (active ingredient: Spinetoram J + Spinetoram L) was applied at a dose of 0.075 L·ha^-1. No insecticides were applied in sunflower.

Forage harvesting was performed when grain had a doughy stage in all three crops. To determine plant height (PH) and green matter yield, for each experimental unit the final height was measured in five plants, the central bed (8 m long) was harvested and the total harvested plants were weighed to estimate green forage yield. Moreover, aboveground biomass (AGB) and root biomass (RB) of three randomly selected plants from each plot were evaluated and sectioned into leaves, stems, fruits and roots, and each fraction was dried in a forced air oven at 60 °C until constant weight. Dry matter (DM) weight was determined from AGB, and total weight of roots from 0 to 15 cm depth was calculated from the root system. Total biomass was the sum of the mass of leaves, stems and fruits. At the end, the ratio between RB and total biomass (AGB + RB) was determined, and a root index (RI): RB/(AGB + RB) was generated, which represented the relative weight of RB in relation to total biomass.

After drying, samples of 0.5 kg of AGB were taken from each crop and each fertilization treatment and were ground in a mill (model TE-650/1, Tecnal®, Brazil) with a 1 mm diameter sieve. Subsequently, samples were sent to the laboratory for bromatological and nutritional analysis to determine forage quality in terms of crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF) and net energy for lactation (NEL).

Roots to determine RB of maize, sunflower and sorghum were obtained by extracting 20 x 20 x 15 cm soil cubes at the plant line. Samples were taken to the work area in closed bags, placed in buckets and left to soak for 24 h according to the methodology described by ^{Barrios et al. (2014}). Subsequently, the material was washed over a 500 µm mesh sieve, and with the help of a very fine brush and tweezers roots were separated from the soil. Roots were placed in an aluminum tray and dried in a forced air oven at 60 °C until constant weight. The results were expressed in kg of roots per m², and root density in kg per m³.

Soil penetration resistance (Pr, kg·cm^-2) was determined with an impact texture analyzer at each selected root sampling point. The number of impacts required to reach each depth indicated was quantified for this variable, and subsequently the following formula was used:

where N is the number of impacts, M is the mass weight (kg), g is the acceleration of gravity (9.81 m·s^-2), SD is the sliding distance (m), PD is the penetration distance (m) and A is the surface area of the cone (m²). The latter was calculated with the following formula:

where r is the cone radius (m) and s is the cone length (m).

Moisture content (ϴs) was measured during growing cycle and at the end of the test with a time domain reflectometer (TDR) in situ per treatment and from 0 to 15 cm depth. This equipment is used to measure soil water content quickly, accurately and in a non-destructive manner.

With the information recorded, some ANOVA’s and comparison of means were carried out using the least significant difference test (LSD, P ≤ 0.05). For this, the statistical package SAS version 9.1.3 (^{SAS Institute Inc., 2013}) was used.

Results and discussion

Precipitation analysis

During the first 31 days of crop development, rainfall was somewhat irregular, accumulating 24 % (56 mm from the first day of sowing to 31 days after sowing) of the rainfall recorded during the entire vegetative cycle in this pre-flowering stage. After August 13, rainfall increased significantly, and its distribution was more uniform (Figure 1). Due to regular rainfall, availability of water was higher in the final stage of the reproductive stage or grain filling of the three crops. Therefore, DM yield and forage quality were not affected. During crop development, temperature ranged from 21 to 31 °C for Tmax, and from 8 to 15 °C for Tmin.

Figure 1 Precipitation and daily maximum (Tmax) and minimum (Tmin) temperature recorded in the growing cycle of rainfed maize, sunflower, and sorghum (San Luís de Letras, Pabellón de Arteaga, Aguascalientes, Mexico, 2018). 

Conclusions

Adding fertilizers, mixing fertilizer with zeolite and bovine manure increased dry matter yield and did not affect forage quality obtained by sowing maize and sunflower in a four-row bed, and sorghum in a six-row bed with vertical tillage and in situ rainwater harvesting. Seed inoculated with mycorrhiza did not significantly increase DM yield or forage quality compared to the other treatments.

The results obtained support the idea that there are no soil limitations or competition for nutrients and water, which can be observed by the positive effects on yield and components of the three crops evaluated. Furthermore, vertical tillage has no effect on the functional balance between root biomass and total biomass for the three crops evaluated.