Relationship of the rose varieties infestation level by spider mite with the bush structural elements under the Phytoseiulus persimilis application in greenhouses

Varieties of roses grown for cutting differ in the degree of costs for protection against pests, primarily from the two-spotted spider mite Tetranychus urticae Koch. To control this pest, from 6-8 to 25 or more treatments with acaricides are required. The predatory mite phytoseiulus Phytoseiulus persimilis A.-H. can be used as an alternative or addition to chemical treatments. Here, we report on a long-term monitoring of the spider mite abundance in commercial rose greenhouses. The observation allows us, for the first time, to assess a relationship between two Rosa hybrida variety-specific morphometric parameters, the area of a compound leaf segment and the total leaf area per bush, and an abundance of the spider mite in a triotrophic system, i.e., rose plant-spider mite-predatory mite. From this data, we obtained the equations to predict the development of the pest and determined the predatory mite number effective on a certain variety. This work aimed i) to assess the two-spotted spider mite infestation in a set of rose varieties, ii) to establish the relationship of the spider mite infestation level with the bush structure elements, and iii) to choose mathematical models for prediction of the pest infestation levels and the number of the predatory mite phytoseiulus necessary to use for the control of the pest. Observations on the two-spotted spider mite development were carried out in a block glass greenhouse of ZAO Agroleader (Vyborgsky District, Leningrad Province) on rose plants ( Rosa sp., hybrid tea group) of 18 varieties. The area of the greenhouse was 45,000 m2. A scoring system was used to assess the infestation levels of roses by spider mites. The greenhouse was divided into plots. Each plot was a 3.95 m long (8.02 m2 in area) segment of a double row of rose bushes. The survey consisted of a visual inspection of plants and assignment of the infestation level score from 1 to 5. Surveys were carried out twice a month, the total number of counts per year was at least 24. The dynamics of rose plant infestation by spider mites was assessed over 8 years (2011-2018). Since 2011, on particular varieties, and since 2012, on the entire area of the rose greenhouse, the predatory mite Ph. persimilis , introduced continuously or into the infestation foci, was used to control the two-spotted spider mite. Continuous application from 3 to 10 individuals/m2 over the entire area of the greenhouse was carried out 1-1.5 times a month; from 10 to 60 individuals per bush were introduced into foci weekly until new significant foci of the pest continued to appear. Acaricides were used only in cases where the T. urticae infestation level exceeded 2.5 points. Seven days after the first treatment the second treatment was carried out. We determined the number of stems in the upper part of the bush (crown) and on the whole bush, the productive stem length, the number of lobes of the complicated leaf, the number of leaves on the entire stem and on 10 centimeters of the stem, the number of leaves in the bush crown and on the entire bush, the surface areas of the lobule and the entire leaf, the area of the leaves surface in the crown and in the entire bush. Correlation analysis was used to assess the relationship between the occupancy of individual varieties of roses and the structural elements of their bushes, and regression analysis was used to describe it mathematically (rectilinear regression equations). To establish the relationship between the parameters of individual elements of the structure of rose bushes and the infestation level of spider mites, a two-factor ANOVA was used. When comparing the parameters of regression models built from sample data, the least squares method was used. In the most contrasting varieties Brazil and Aqua, the average long-term level of infection differed by 17.8 times. The remaining varieties could be divided into 6-8 groups, of which the most contrasting ones differed by 5.0