Assessment of the spatial distribution of the sunflower tumbling beetle population

Abstract

We have evaluated the distribution of the sunflower tumbling beetle (STB) to optimize the methods of its monitoring and to establish the spatial localization of the phytophage concentration in order to increase the efficiency of measures to regulate its abundance. We counted the STB after harvest at least in 20 points of the field; we collected stems and their root parts in plots 1x1 m, opened them and counted the number of larvae in each stem; we also counted broken plants at a height of 50-60 cm. Svedberg and Lloyd models used to analyze the distribution (uniform, random, contagious) of the STB population on the crops of F1 hybrid "Sonyachny Nastriy". The data obtained as a result of solid - by field squares (grid) and W-like route surveys processed according to their parameters. The last ones was process both with application of statistical correction of plots according to the number of the STB (larvae / plant) - by changing the index of abundance by the size of an accounting plot, which is different from the average area, and by results of direct counts carried out in one or another point of the field. The distribution can be spatial - within the entire agrocenosis, and linear - within the vector (strip of registration plots). It can also be based on complete (grid data) or incomplete (en-route counts) information. Lloyd's model assumes a higher probability of random values of distribution. When estimating the distribution based on incomplete information, the values of the Swedberg and Lloyd coefficients for the spatial distribution are 1.708 and 1.604 for W-route with area-adjusted abundance and 1.692 and 1.596 without adjustment respectively, which gives values of contagious and random distribution. Applying the linear version of the estimate gives a contagious distribution for the Swedberg and Lloyd models for the 1st vector and an even distribution for the 2nd and 3rd vectors. When estimating the distribution based on complete information (the "grid" variant), the spatial distribution of Lloyd - 0.657 and Swedberg - 0.957 is uniform, but for the latter it is very close to the random distribution. The linear distribution by Swedberg varies from contagious - 1.287 to uniform - 0.916 and 0.447, and by Lloyd - from random - 1.237 to uniform - 0.499 and 0.014. Given incomplete information, the polynomial trends allow us to determine the magnitude of the uncertainty in the distribution as the geometric distance between the points of intersection of the trend lines with the 45o optimization line and the Swedberg contagion and Lloyd's randomness levels. Provided the information is complete, linear trends allow us to determine the magnitude of the uncertainty in the distribution as the geometric distance between the points of intersection of the trend lines with the Swedberg-Lloyd uniformity levels. "Grid" of these counts allows revealing marginal effects in spatial distribution of phytophage population, the number of which is maximal in the marginal strip bordering the forest belt, and minimal in the center of agrocenosis. Under the condition of incomplete information it was found that according to Swedberg criterion the spatial distribution of the STB is contagious and Lloyd's is random, the linear distribution for both models changes from contagious to uniform; under the condition of full information it is found that on the basis of Swedberg-Lloyd models the spatial distribution of the stonefly is uniform, linear distribution for the Swedberg model changes from contagious to uniform, Lloyd - from random to uniform; estimation of the spatial distribution of the population on the basis of full information is the most optimal because it allows to determine the marginal effects.