Growth conditions, plant material, crop management, and experimental design
The trial was conducted in a greenhouse located at Paraje Águilas Bajas, Santa María del Águila, Almería, Spain (36°47ʹ39″N 2°46ʹ32″W). The greenhouse was composed of polycarbonate walls and a roof made of trilaminated low-density polyethylene (LDPE) film (200 μm thickness) with ~60% spectral transmittance in the photosynthetically active radiation (PAR) region. The greenhouse was unheated and passively ventilated with lateral side panels and flap roof windows. It had an east-west orientation and a north-south crop row alignment. The air temperature and relative humidity (RH) inside the greenhouse were in the ranges of 12–32 °C and 50–70%, respectively. Transplants of the sweet pepper (Capsicum annuum L.) hybrid ‘SV1204PB’ (Seminis, Montornés del Vallés, Barcelona, Spain) at the 4–5 true-leaf stage were planted in “Enarenado” sandy soil commonly used in greenhouse production in Almería. The planting date was 19 July 2017 and the planting density was 2.0 m-2. The soil composition was 13.5% (w/w) clay, 72.8% (w/w) sand, and 13.7% (w/w) silt. Its pH was 7.52, its organic matter content was 0.71%, and its total nitrogen, available phosphorus, and exchangeable potassium were 690 mg kg−1, 51.4 mg kg−1, and 321 mg kg−1, respectively. Aerial drip irrigation was used. The in-line emitters were positioned at 0.30-m intervals and the emitter flow rate was 3.4 L h-1. Preplant fertiliser was broadcast at 90 kg·ha-1 P, 120 kg·ha-1 K, and 15 kg·ha-1 Mg and incorporated into the soil. Additional fertiliser in the form of K2SO4 (80 kg·ha-1 K) was applied through the drip irrigation system. Nitrogen was applied via fertigation in the form of 27% NH4NO3soluble fertiliser starting 10 d after transplanting until day 83. The total N supply was split into ten weekly dressings. Powdery mildew caused by Leveillula taurica was controlled by three foliar applications of penconazole (Topas 10EC; Syngenta, Madrid, Spain) at the label-recommended rate. Aphids and spider mites were controlled by one foliar application each of imidacloprid (Confidor 200 SL; Bayer Crop Science, Valencia, Spain) and fenpyroximate (Miro; Bayer Crop Science, Valencia, Spain), respectively. Bumblebees promoted flower pollination. Weeds were controlled by hand hoeing. The control and microbial-based biostimulant treatments were compared in a randomised block design with four replicates for a total of eight experimental plots. The microbial-based biostimulants were applied through a drip irrigation system. The first application was made at 15 d after transplanting (DAT) (3 August 2017) at the rates of 1×10-6 spores ha-1 Rhizoglomus irregularis BEG72 and 1×10-6 spores ha-1Funneliformis mosseae BEG234 in the form of 2.0 kg ha-1 Team Horticola (Agrotecnologías Naturales, S.L., Tarragona, Spain) plus 1×1012 CFU ha-1 Trichoderma koningii TK7 in the form of 1.0 kg ha-1 Condor Shield (Agrotecnologías Naturales, S.L., Tarragona, Spain). The second treatment was applied 43 DAT (31 August 2017) at the rate of 5×1011 CFU ha-1 Trichoderma koningii TK7 as 0.5 kg Condor Shield (Agrotecnologías Naturales, S.L., Tarragona, Spain). Each experimental plot was 30 m2 and contained 60 plants.