Effective methods and means of protection grain crops from Eurygaster integriceps put. in the republic of Uzbekistan

. The most effective methods of protection are described and the results of long-term observations of the dynamics of the number and harmfulness of the Eurygaster integriceps Put on industrial wheat crops in the conditions of Fergana Valley of the Republic of Uzbekistan are presented. A reliable positive correlation dependence of wheat grain damage from the number of Eurygaster integriceps larvae in crops was established. The ways and principles of choosing essential factors in the development of mathematical models for predicting the development and distribution of Eurygaster integriceps, as well as the issues of zoning territories according to their weather and environmental characteristics are given.


Introduction
In recent years, a complicated phytosanitary situation has developed in the Republic of Uzbekistan, associated with an increase in the sown area of grain crops in irrigated agriculture and changes in the system of managing the cultivation of agricultural crops (mainly cotton and grain).
The most important task of agriculture in Uzbekistan is to increase the production of highquality grain.However, due to a number of objective and subjective reasons, the yield and gross grain harvest fluctuate significantly over the years.In some years, on the decrease in the gross harvest of high-quality wheat grain negatively affected mass reproduction of various types of pests.Year after year, the area of distribution of such dangerous pests as Eurygaster integriceps, Empoasca meridiana, Haplothrips tricici, Schizaphis gramina, Sitobion avenae, and harmful short-horned grasshoppers (Acrididae) on pastures is expanding.
Morphological and biological features of development, distribution area, harmfulness, scientifically based forecasting and optimal timing of protective measures to control the abovementioned pests in the conditions of the republic have not been studied enough.The system of effective and environmentally friendly measures for the protection of these crops has not been developed to a sufficient extent.
Despite the successful pest control of grain crops, many problems of organizing pests monitoring on the territory of the republic, the development of scientifically based forecasts of the emergence, development and spread of especially dangerous pests, diseases and weeds, and the introduction of modern methods of pests monitoring remain unresolved.
Based on the abovementioned, the purpose of this article is to develop systems for the protection of grain crops from Eurygaster integriceps Put., which is one of the economically dangerous insect species that require constant monitoring of their populations condition due to their high abundance, harmfulness and wide distribution.

Materials and methods
Observations on the number of Eurygaster integriceps (Sunn pest) were carried out on production crops of wheat in farms of Andijan, Namangan and Fergana regions of Fergana Valley of the Republic of Uzbekistan in 2019-2022.
The number of Sunn pest was estimated on wheat crops by examining 20 accounting plots (0.25 m2) for every 80 ha of the surveyed area.Diagnosis and determination of the degree of grain damage by Sunn pest was carried out on a 5-point scale [1][2][3].Grains damage by Sunn pest was divided into 3 groups: weakly (up to 10%), moderately (from 10 to 20%), and severely (more than 20%) damaged [2,3,15].

Results and discussion
The harmfulness of Eurygaster integriceps is determined by the effect of their digestive enzymes, introduced by this insect into the endosperm of the grain during nutrition, on the main biopolymers of plants, which leads to a sharp deterioration in the technological, baking and other properties of the grain damaged by it [1-5, 10, 11, 19, 20].Therefore, when 12% of the grain is damaged, complete degradation of gluten occurs by 6-10%, a decrease in germination by 22-25% and germination energy by 18-21%.In the presence of 10-20% of grain, damaged by Eurygaster integriceps in the grain mass, food wheat is transferred to the fodder category [3,8,11,15].Potential yield losses from Eurygaster integriceps, depending on the soil-climatic zone, cultivation technologies and culture, can reach 20-50%, and in the absence of protective measures and the use of susceptible varieties, grain losses will be more significant [3,19,20].
Fergana Valley regions of the Republic (Andijan, Namangan, and Fergana) are located in the zone of constant high abundance and significant harmfulness of Eurygaster integriceps.This is due to the instability of the phytosanitary situation, which provokes periodic outbreaks of pests mass reproduction.The cyclical nature of mass reproduction of pest reveals a connection with changes in solar activity, climatic conditions, and anthropogenic impacts [12][13][14][15].In this regard, in order to predict wheat crop losses from this pest and plan sciencebased plant protection measures, it is necessary to carry out phytosanitary monitoring in a timely manner, which allows assessing the number and condition of this species population, which requires simple, effective and affordable methods for predicting the appearance terms, accounting the number and harmfulness of Eurygaster integriceps.
Forecasting the timing of Eurygaster integriceps Put.appearance and the organization of protective measures was carried out using the developed app "Eurygaster integriceps" for smartphones running on the Android system [18].This app is registered with the Intellectual Property Agency of the Republic of Uzbekistan and received a certificate under the number GDU 05283, which allows you to determine the timing of Sunn Pest appearance 10 days in advance.It was introduced in 2019 and 2021 in the practical activities of farms in Ulugnar, Oltinkul, Pakhtaabad, Zhalakuduk and Kurgantepa districts of Andijan region.The block diagram for determining the timing of appearance (figure) and the results obtained are shown in table 1.Similar results were obtained for Namangan and Fergana regions.
Monitoring studies conducted in the regions of Fergana Valley in 2019-2022 showed the number (density) and degree of crops infection with Eurygaster integriceps ranged from 2.2 to 3.8 sp.(and pest larvae from 14 to 22.2) per 1 m2 and 2-4 points out of a 5 point scale [2], respectively.In this case, the average score of the degree of infection can be calculated by the formula: where Бср is the average score; i -points of infection; Ai -the number of grains to the corresponding score; N -the total number of grains in the samples.The monitoring results showed that there is a high correlation between the degree of infection of grain crops and the number (density) of Sunn Pest.The mathematical model of this process is described by the equation: y = 4.99 + 1.76 x (r = 0.8) in this formula, y -the degree of infection; x -the number of larvae of Sunn Pest; r -the correlation coefficient.An analysis of the regression equation shows that an increase of Sunn Pest larvae by 1, increases the infection degree of grains by 1.76%.In other words, it can be argued, that this regression equation makes it possible to obtain an estimate of infection degree of grain crops in the conditions of regions of Fergana Valley and plan protective measures to control them.

(SET -sum of effective temperatures)
Effective management of crop production requires the timely collection and processing of extensive agrotechnical, agroeconomic, agrometeorological and agroecological information.The collection, storage and appropriate processing of the abovementioned information is necessary for making optimal decisions.The collection of such diverse and extensive information and the solution of prognostic problems require huge labor costs.In this regard, the tasks of developing mechanized data collection systems, improving forecasting methods with a focus on more accessible types of initial information and using modern computer and microprocessor technology for processing, storing information and creating a database (DB) arise.
Moreover, mathematical modeling of the process of pest population dynamics is a complicated multifactorial complex, where, along with essential factors, less significant and insignificant ones act.Very often, the mathematical model and the empirical curve do not coincide due to the fact that the presence of insignificant factors obscures the main aspects of the process under study, and some significant factors that give this process a certain character may not be taken into account.In addition, numerous factors contribute to a sharp increase in the amount of work, associated with the collection and processing of information.In view of these circumstances, when modeling the dynamics of agricultural pest populations, it is advisable to divide the entire factorial process into areas of significant, less significant and insignificant factors.The solution of this problem will be much easier, if we introduce a certain quantitative measure to assess the significance of a particular factor.
With this in mind, arises the problem of developing methods for selecting significant factors in predicting the dynamics of Sunn Pest population and phytosanitary zoning of territories using pattern recognition algorithms, which are the main ones for calculating estimates [6,7].However, methods in which the main stage is the calculation and consideration of the «information weight» of features are becoming more and more widespread.The informational weights of features are calculated by the frequency of occurrence, the analysis of dead-end tests, and algorithms for calculating estimates.Let us consider some general principles on which the algorithms for calculating estimates are built.
Let the table Tnm be given.In it, each line is Sj (j=1, 2… m) the essence of the object that can be represented as where Xji (i=1, 2…, n) are elements of a set of numerical or qualitative values that identify an object.Usually these elements are called signs, and their values are obtained, as a rule, as a result of experiments.Let now a collection of sets qi be given, which we will consider as some alphabet of features.As an alphabet can be: a) a set of two elements {0 , 1}, where "1" means the presence of some property, 0 means its absence; b) a finite set of integers {1,2,…,P }.The meaningful value of each such attribute reflects the degree of expression of the corresponding property.
An object Sj is called accessible if Xji ∩ qi.Note that the concept of a valid object is equivalent to the concept of a valid row.
The set of m valid objects, each of which is characterized by a set of features, can be summarized to a table Tnm, which we will call an admissible table.Now suppose that there is a partition of all admissible rows into classes k1, k2, …, kl , and Ku ∩ Kj ≠ 0 for ≠ j.This partition induces a partition of the rows of the table into l non-overlapping (disjointing) classes.

K1, K2, …, Kl
We assume that the last partition is given.Let us denote the number of elements of class Ke as mu-mu-1 and represent the division of rows by class as ………………….
The introduced concepts and definitions allow us to proceed to the consideration of the principle caused by the principle of calculating some quantitative estimates, obtained by comparing admissible objects or parts and representing the basis of algorithms for calculating estimates.
Let two admissible objects be given: where xjn (j=1,2) takes the values of one of the above alphabets.Let further it is required to establish a degree of "affinity".We will understand this term on an intuitive level, comparing it with the concepts of similarity, sameness, etc.The affinity of S1 and S2 can be established, for example, based on the analysis of the identifying sets or their parts.This approach is associated with obtaining some quantitative estimation using certain transformations of the sets.Based on this assessment, affinity is established by a certain criterion.
Thus, if the identifying sets of objects are denoted by x1 and x2, and the estimate by B, then for the latter one can write

B=B(x1, x2)
The value of B can be intermediate, and, in turn, make it possible to obtain some final estimate characterizing the affinity of the object.This assessment in quantitative expression characterizes the total effect of objects comparison.
Therefore, if the total effect is denoted by Г, then we can write Г=Г (В) =Г (В(x1, x2)) In the theory of recognition algorithms based on the calculation of estimates, the value Г is usually called the number of votes given by objects, for example, for the object S2.It is clear that the number of votes cast by objects S2 for S1 (Г1,2) is equal to the number of votes Г1,2.
Procedures for obtaining the value of Г for comparing objects S1 and S2 are called voting procedures.Voting procedures can be carried out not only to establish the affinity of two objects, but also a group of objects summarized in a specific table.So, if some table Tnm consists of objects, each of which is characterized by a set of n features, then the following voting matrix can be obtained by the voting procedure:

………………….. Гnm
Matrix (1) actually makes it possible to obtain all the necessary values of the votes given by individual objects both for themselves and for other objects of this table.In practical problems, the votes cast for oneself are usually excluded from the matrix (1).
In the theory of evaluation algorithms, such an apparatus has been obtained, with the help of which the number of votes is calculated relatively simply.The works [1][2][3] describe the finding and proofs of the following expressions for counting votes (the value of Г), as well as various modifications of the counting formulas, which differ from each other in the specifics of setting certain stages of algorithms for calculating estimates.Let us write out expressions for counting votes, which will be used later in this paper.
If the table Тnml is given, divided into classes K1, K2, …, Kl, then the number of votes given by row S for class Ku is equal to where S=(x1, x2, …, xn), Sq= (xq1, xq2, … ,xqn) are compared pairs of lines; h is the length of the voting set, i.e. the number of columns by which the rows S and Sq are compared; r(S, Sq) -Hamming distance between the rows S and Sq (if the table is given by binary characters), i.e. the number of columns where rows do not match.
When the table is filled with elements that take a value from an arbitrary alphabet, then instead of a value, you should use r(S, Sq) -the distance between the rows S and Sq, equal to the number of fulfilled inequalities where abs(xl-xqi) ≥ Vi -the proximity threshold of the i-th feature.If h≥ n-r(S, Sq), then expression (2) vanishes.
According to [6,7,9], the value is called the information weight of the i-th feature.Thus, the above method for determining the information weights of signs was used to select the essential factors necessary in the development of mathematical models for predicting the development of harmful objects [17].
The task of optimized zoning of agricultural territories according to weather and environmental characteristics are essential links in the scientific knowledge of the continuous environment.Works in this direction are of great practical importance, since zoning is an essential element in many studies, in particular, when predicting the dynamics of the population of agricultural pests, including E.integriceps, the size of the infested areas by pests, the dates of pests' appearance, the planning of pest control measures, when analyzing the causes of pest outbreaks, etc. [21].
Zoning according to the characteristics of reproduction and the timing of the passage of the main phenological phases of harmful organisms makes it possible to identify zones with different intensity of pests' reproduction.
To solve the problems of zoning in plant protection, the same class of pattern recognition algorithms [6,7,9], which was considered above, is used.In this case, the classification of objects without a standard is used.The problem of spontaneous partitioning of a set of objects is solved using algorithms for calculating estimates using some quantitative measures that characterize the information content of both features and the objects themselves.The informativeness of the object is obtained from expression (3).Computation for all given objects, the information weights are then ordered in descending order.
The division of objects into classes in this case is based on the assumption that when ordering objects by the value of the latter, they will be grouped by rank.The partition obtained in this way is an intermediate stage of spontaneous classification.The final partition is obtained only after the voting confirms that the object belongs to its class.
As an example, table 2 shows the results of zoning the territory of Andijan region in relation to E. integriceps on wheat crops in terms of abundance, the timing of the passage of the main phenological phases, hydrometeorological (the sum of effective temperatures, the sum of precipitation, etc.) indicators.
Thus, as can be seen from table 2, the districts of Andijan region can be divided into classes:  Andijan, Asaka, Buz, Shakhrikhan districts -1st class;  Balykchy, Bulakbashy, Zhalakuduk, Izboskan districts -2nd class;  Kurgantepa, Markhamat, Ulugnar districts -3rd class;  Oltinkul, Pakhtaabad, Khuzhaabad districts -4th class.Therefore, by allocating areas to the corresponding classes, it is possible to apply the results of predicting the timing of Eurygaster integriceps Put.appearance and organizing protective measures to control them in other areas, belonging to the same class.

Conclusion
The pest control system, practiced in the past decades, mainly due to the massive use of chemicals, especially when their use was not sufficiently substantiated, in ecological and economic terms, led to serious problems associated with a negative impact on the environment and the emergence of pest resistance to plant protection means.In part, this even contributed to a direct or indirect increase in the harmfulness of certain types of pests and diseases and the growing dependence of the crop on the effectiveness of measures to control them.
With all this in mind, the ways and principles for choosing essential factors in the development of mathematical models, as well as the issues of zoning territories according to the weather and environmental characteristics of E.integriceps in the conditions of the Republic of Uzbekistan, are given.The optimal allocation of areas to the corresponding classes makes it possible to apply the results of predicting the timing of E.integriceps appearance and organizing protection measures to control them in other areas, belonging to the same class.

Table 1 .
The development of Eurygaster integriceps Put in the districts of Andijan region and the optimal timing of control measures(2019)(2020)(2021)