RU2728159C1 - Method for taxation of plantations - Google Patents

Abstract

FIELD: measurement.

SUBSTANCE: method involves remote sounding of a taxied territory using space survey data or aerial photography and ground measurements at control points, using air laser scanning data when constructing a digital relief model and a digital model of plantation. As aerial photography or aerial photography data, ultrahigh resolution shots from 0.1 m/pixel are used. Networks of calibration sample areas are constructed on the basis of classification of synthesized images of aerial photography, space survey and air-laser scanning by age groups, groups of types of forest-and-vegetation conditions and prevailing rocks. Based on the results of ground-based measurements, calibration models of the dependence of the taxation characteristics on the aerial photographic variables are plotted on calibration sample areas, space survey and air-laser scanning and based on the determined relationships, calculating taxation factors for cells of calculated matrix 16×16 meters, covering the entire territory of performance of inventory works. Parameters for forest-affected areas are determined with formation of a data base on the basis of which taxation description of plantations is made.

EFFECT: invention provides higher objectivity and minimization of labor costs when determining forest taxation indicators.

1 cl, 4 dwg

Description

The invention relates to forestry, in particular to remote methods of forest inventory over large areas.

There are known methods for assessing the state of plantations during land (natural) forest taxation: eye; eye-measuring; decryption; updating; [Forest management instruction [Electronic resource] (approved by order of the Ministry of Natural Resources of the Russian Federation dated March 29, 2018, No. 122). Access mode: http://docs.cntd.ru/document/542621790/, free. (Date of treatment: 01.12.2019)

The main disadvantage of eye and eye measuring methods is their high labor intensity, since these methods involve a significant amount of ground surveys throughout the taxed area. The disadvantage of existing decryption methods is the lack of automation of image processing processes, high requirements for the qualifications of decoder engineers and limitations in the possibility of measuring the height of dense stands. The disadvantage of the actualization method is the low accuracy of the data obtained, which often does not reflect the real changes in the inventory indicators of stands, since the assessment of these changes is based on the models of forest stands development. In addition, for eye, eye-measuring and decoding methods, the disadvantage is that they are based on the subjective opinion of the tax engineer.

There is a known method of natural taxation of plantations by individual enumeration of trees on test plots, according to which the forest area is divided into a mosaic of plots, and measurements of taxation parameters are carried out on a trial (key) site (see, for example, Anuchin N.P. Forest taxation, 5- e edition. - M .: Forestry, 1962, pp. 248-250).

The disadvantage of this method is the lack of the algorithm for selecting test (key areas), as a result of which the sample is not controlled and the accuracy of the result is not ensured.

A known method for assessing the state of forests, which includes the choice of an integral criterion of state, remote measurement from the orbital station of the spectral brightness coefficients of the probed array in three parts of the visible spectrum with a discreteness of 2 nm, quantization of signal amplitudes with a step of 1/256, calculation of chromatic coefficients of vitality and damage, calculation of the mutual regression function and its calibration in the entire range of changes of the integral criterion, continuous tracking of signs, their recording in the coordinate of the scoring on the on-board printer and dumping the results via a facsimile radio link, (patent RU No. 2038001, IPC A01G 23/00. publ. 27.06. 95 g.).

The disadvantages of this method are:

- low spatial resolution of spectrometric means (from hundreds of meters to units of km), which does not allow detecting affected areas within one hectare;

- static instability of the results, since the assessment is carried out according to the only measurable parameter - the spectral brightness coefficient. The problem solved by a known method is to quantitatively measure the signs of forest pathology that exist simultaneously in several spectral ranges of the solar radiation flux reflected from the tree canopy, and to identify the regression relationship between the measured traits and the standard integral ecological indicator of the degree of weakening of forest stands.

- the method is aimed at assessing the state of forests, without determining the taxation characteristics of plantations.

The closest in technical essence and the achieved result to the proposed technical solution is a method of forest inventory of the territory based on multi-temporal datasets of remote sensing of the Earth of high and medium resolution (2-40 m / pixel) on 3-10 spectral channels in the range of 0.45-90 microns (patent RU No. 2565280, IPC A01G 23/00, publ. 20.10.15).). The resulting datasets are classified and the number of control points is determined for each selected class, which is then placed inside the classes. At the control points, forest inventory field descriptions are produced, which are then interpolated to the rest of the territory not covered by ground-based observations.

Disadvantages of the prototype method: data of average resolution do not allow to accurately establish the boundaries of forest inventory allotments, the list of forest inventory indicators determined using this technology is limited only by the stock of individual species and the average height of stands. Another disadvantage of this method is the low accuracy of the data obtained, which does not meet the requirements of the methods established by the forest management instructions.

The task of creating the proposed invention is to develop a new method for taxation of plantations, free from the disadvantages of the prototype method.

The problem is solved using the features specified in the claims in common with the prototype, such as the method of forest inventory of the territory, including remote sensing of the taxed area using space survey data (CS) or aerial photography (APS) and ground measurements at control points, and distinctive essential features , such as using airborne laser scanning (VL) data when constructing a digital elevation model and a digital model of plantation canopy, and as space imagery (CS) or AFS data use ultra-high resolution images from 0.1 m / pixel, build networks of calibration test plots based on the classification of synthesized images of APS, KS and VLS according to age groups, groups of types of forest growing conditions and predominant species, statistical models of the dependence of taxation indicators on the variables These dependencies are used to calculate taxation indicators for the cells of the 16 × 16 meter design matrix covering the entire territory of taxation works, then the parameters for forest taxation allotments are determined with the formation of a taxation database, on the basis of which a taxation description of plantations is made.

One of the main tasks solved by the proposed method is the technical ability to obtain taxation characteristics in the context of forest-forming species with an accuracy that meets the requirements of Russian forest legislation.

In contrast to the prototype, which makes it possible to compose the spatial distributions of individual taxation indicators in the form of maps, in the proposed technology, in addition to such distributions, a database is formed, on the basis of which taxation descriptions can be compiled.

The above set of essential features allows you to obtain the following technical result - increasing objectivity and minimizing labor costs in determining forest inventory indicators, which are understood as stock, average height, average diameter, average age, absolute completeness (sum of cross-sectional areas) in the context of forest-forming species, as well as the number , age, height and species composition of the undergrowth, as well as the possibility of obtaining high-precision data that meets the requirements of the methods established by the forest management instructions.

The invention is illustrated in FIG. 1. A schematic diagram of an airborne laser scan, FIG. 2. The layout of the trial plots in the cluster, FIG. 3 Scheme of the calibration trial plot, where A is the undergrowth measurement area, B is the tree measurement area with more than 10 trees, C is the tree measurement area with less than 10 trees in area B, and FIG. 4. Calculated matrix 16x16 m, combined with AFS.

Below is an example of implementation of the invention.

The forest inventory method is based on the complex use of Earth remote sensing data and sample measurements in kind. A feature of this method is that, in contrast to the existing taxation methods, it is proposed to comprehensively use airborne laser scanning (VLS) data and aerial photography (AFS) or space imagery (CS) data with ground-based measurements as the initial information for obtaining taxation characteristics. In this case, the digital model of the canopy of woody vegetation, obtained from the VLS data (see Fig. 1), is used as a source of variable values of the height and density of the stand, and the data of the APS or KS are used to determine the species composition of plantations (the share of the stock of each forest-forming species in the total stock stand). Ultra-high-resolution images from 0.1 m / pixel are used as space imagery data or AFS.

After obtaining the initial variables for height, density and species composition of plantings, a network of calibration sample plots is designed. The territory of the object of work is divided according to the prevailing species, groups of forest growing conditions and age groups. For each allocated class, from 30 to 50 trial plots are designed, depending on the degree of prevalence of the class, where 30 is the minimum number of samples for the least common classes and 50 is the minimum value for significantly common classes. Sample plots are arranged in clusters in the form of squares 3 by 3 samples with a distance of 200 meters between them (see Fig. 2)

Each sample is a circular area of constant radius with a radius of 9 meters (see Fig. 3). The radius of the sample increases to 12.62 meters if less than 10 trees fall within the original 9 meters. On the sample, the growing trees are measured with the definition of the species, diameter, technical suitability category. After the enumeration, for each species that entered the site, three model trees are selected, for which the height and age were additionally measured. In addition, within a radius of 5 meters, undergrowth was counted with the determination of its average height, quantity by species, and average age.

The calculation of the taxation parameters was carried out on the basis of the revealed relationships between the characteristics of the stand measured on the test plots, the parameters of the digital model of the canopy, and the spectral characteristics of space and / or aerial photographs.

The calculation of the taxation indicators of the stands was carried out initially for each cell of a regular grid (for example, with dimensions of 16 by 16 meters), which completely covers the object of plantation taxation (see Fig. 4). After obtaining the boundaries of forest inventory units, the taxation indicators of such units are determined through the area-weighted average values of the cells falling within the boundaries of each unit.

The obtained values of indicators for each section are formed in the form of a database, on the basis of which a standard taxation description is constructed (Certificate of the Russian Federation No. 20176600455 for computer programs).

The proposed technology has been tested in the forestry of the Arkhangelsk region. The approbation was carried out on a forest plot with a total area of 43,567 hectares. The results were checked at 36 forest inventory divisions in four quarters of Krasnoborskoye uchastkovoye lesnichestvo of Krasnoborskoye lesnichestvo. The allocation of the allotments was carried out by eye-measuring method using normative and reference materials used in the territory of the Arkhangelsk region. When carrying out the taxation, the requirements established by the current forest management instructions were fully taken into account. When checking, special attention was paid to determining the average stock and completeness of plantings.

The prevailing breed and economy is determined correctly in all areas. The composition coefficient of the dominant breed was determined correctly in all areas, but a systematic error with a value of -3.5% was revealed.

In terms of age, no deviations exceeding the permissible values were found, the systematic error was -0.6%.

A deviation exceeding the permissible value was found in height in one section. The systematic error was -3.5%.

There were no deviations or systematic errors in diameter.

By relative completeness, all errors with a tolerance. A systematic overestimation of + 5% was revealed.

In terms of margin, all deviations are within acceptable limits. The systematic error was + 9%.

In terms of bonitet class, a discrepancy was found in five sections. According to the class of marketability, the coincidence was found in all divisions. In terms of the amount of undergrowth, a discrepancy was found in three sections, and in four sections, a discrepancy in the type of forest was revealed.

Thus, the check showed that, in general, the taxation materials obtained on the basis of the proposed method are satisfactory. According to the accuracy of determining taxation indicators, all indicators are determined without exceeding the permissible deviations established for the eye-measuring method of taxation. In terms of systematic errors, the deviations are also within the values permissible for the eye-measuring method.

This description is considered as material illustrating the invention, the essence of which and the scope of patent claims are defined in the following claims, a set of essential features and their equivalents.

Claims (1)

A method of planting taxation, including remote sensing of the taxed area using space survey data or aerial photography and conducting ground measurements at control points, characterized in that it uses airborne laser scanning data when constructing a digital elevation model and a digital model of the plantation canopy, and as space data surveys or aerial photography use ultra-high resolution images from 0.1 m / pixel, build networks of calibration sample plots based on the classification of synthesized images of aerial photography, space imagery and airborne laser scanning by age groups, groups of types of forest conditions and prevailing species, based on the results of ground measurements on calibration test plots, statistical models of the dependence of taxation indicators on the variables of aerial photography, space imagery and airborne laser scanning are built and on the basis of the revealed dependencies, the rate is calculated parameters for the cells of the calculation matrix 16 × 16 meters, covering the entire territory of the taxation works, then the parameters for the forest taxation allotments are determined with the formation of a taxation database, on the basis of which the taxation description of the plantations is made.

Images