CC BY
13ОЦ1НКА ВКЛАДУ ЗЕЛЕНИХ НАСАДЖЕНЬ Р1ЗНОГО ФУНКЦ1ОНАЛЬНОГО ПРИЗНАЧЕННЯ У ЗНИЖЕННЯ ТРАНСПОРТНОГО ШУМУ ВЗДОВЖ АВТОМОБ1ЛЬНИХ ДОР1Г НА
УРБАН1ЗОВАНИХ ТЕРИТОР1ЯХ
Внукова Н.В.
зав1дувач кафедри екологИ] доктор техтчних наук, професор Харювський нацюнальний автомобшьно-дорожнш утверситет
Решетченко А.1.
завгдувач навчальног лабораторИ кафедри тженерноЧ екологИ Micm Харювський нацюнальний унiверситет мюького господарства
iменi О.М. Бекетова Вергелес Ю.1.
старший викладач кафедри тженерно'1 екологИ Miст Харювський нацюнальний унiверситет мюького господарства
iменi О.М. Бекетова
ASSESSMENT OF THE CONTRIBUTION OF PLANTINGS OF DIFFERENT FUNCTIONAL PURPOSE IN REDUCING TRANSPORT NOISE ALONG MOTORWAY ROADS IN URBANIZED
AREAS
VnukovaN.
Head of the Department of Ecology, Doctor of Technical Sciences, Professor
Kharkiv National Highway University Reshetchenko A.
Head of Educational Laboratory of the Department of Engineering Ecology of Cities
O.M. Beketov Kharkiv National University of Urban
Vergeles Yu.
Senior Lecturer of the Department of Engineering Ecology of Cities O.M. Beketov Kharkiv National University of Urban
Анотащя
Стаття присвячена актуальнш проблематищ еколопчно! безпеки в обласп акустичного забруднення урбашзованих територш. Автотранспортний шум розглядаеться як фактор техногенного забруднення ото-чуючого середовища.
Проведено низку польових дослвджень i3 вимiрами рiвнiв шуму на рiзних територiях мюта, розрахо-вано екивалентш рiвнi шуму, що використовувались у подальшш статистичнш обробщ. Також на обраних експериментальних донках враховувались наявнють i тип насаджень, що слугують у якостi противошумового фшьтру, описано 1х видовий склад за ярусами. Використовувались так статистичш апарати: проста статистика, аналiз головних компонент, критерiй Колмогорова-Смiрнова. Отриманий масив даних було розподiлено на чотири групи за класифiкючою ознакою «тип насаджень»: насадження вiдсутнi, люопарк, парки, iншi насадження (вуличнi та внутршньо-квартально). За допомогою програмного засобу MS Excel ® побудовано лшшш регресiйнi моделi абсолютного зниження е^валентного рiвня транспортного шуму залежно ввд вiдстанi до джерел шуму насадженнями рiзних типiв. Пвдтверджено, що найб№ше зниження шуму досягаеться на вщсташ 20 м. вiд пропно! частини, а найефектившшим типом насаджень у якосп противошумового фiльтру слугують насадження паркового та люопаркового типiв.
Abstract
The article is devoted to current issues of environmental safety in the field of acoustic pollution of urbanized areas. Road transport noise is considered as a factor of technological pollution of the environment.
Series of field studies were conducted with measurements of noise levels in various areas of the city, and equivalent noise levels were calculated, which were used in further statistical processing. Also, in the selected experimental sites, the presence and type of plantings that serve as an anti-noise filter were taken into account, their species composition in tiers was described. The following statistical devices were used: simple statistics, principal components analysis, Kolmogorov-Smirnov criterion. The resulting data array was divided into four groups according to the "type of plantings" classification criterion: there are no plantings, forest park, parks, and other plantings (street and inland). Using MS Excel ® software, linear regression models have been built to absolutely reduce the equivalent level of traffic noise depending on the distance to noise sources by various types of plantings. It is confirmed that the greatest reduction of noise is achieved at the distance of 20 m from the roadway, and the most effective type of plantings as a noise filter are plantings of park and forest park types.
Ключов1 слова: еколопчна безпека, зелеш насадження, автомобшьна дорога, транспортний шум, ур-башзована територiя, лшшна регреая,статистичний аналiз.
Keywords: ecological safety, plantings, highway, transport noise, urbanized territory, linear regression, statistical analysis.
Introduction. In modern cities the level of anthropogenic impact on the environment is constantly increasing. The issue of environmental safety in the conditions of intensive operation of vehicles in urban areas is relevant in terms of various factors of environmental and human impact. It is well known that traffic noise accounts for about 80% of the total pollution of the acoustic environment. The main reason for the unfavorable acoustic environment, which is caused by the operation of vehicles, is the increase in the number of land vehicles and the low capacity of the roads, which are not designed for modern vehicles, and the road surface needs major overhaul. The problem of the location of the road network near residential buildings and the density of urban development, which is the cause of insufficient recreation area between quarters, as well as of active noise and transport activity, are extremely acute [1].
One of the measures to protect residential areas from increased noise levels occurring on roads is the use of plantings. The contribution of plantings to noise reduction has not yet been properly investigated. Recommendations on the location of plantings, which are given in DSTU [2] cannot be applied within the rural areas in connection with the established architectural and planning ensemble of the city territory.
In the paper, the traffic noise is considered as a factor of anthropogenic impact on the urban environment. Road transport noise is a dangerous parametric pollution of the environment and has a negative impact on public health. According to the World Health Organization, thousands of people worldwide die prematurely due to heart disorders caused by prolonged exposure to high noise levels, in particular from vehicles [3]. Analysis of studies of domestic and foreign scientists confirmed the negative effect of noise on human health and the emergence of diseases such as disorders of the cardiovascular system, hearing and nervous systems, can be a cause of poor sleep and cause irritation [4-9].
Goal of the research: assessment of the contribution of plantings of different functional purpose in reducing transport noise along motorway road in urbanized areas.
To achieve this goal, it was necessary to solve the problem of conducting a series of researches to determine the levels of noise pollution and statistical processing of the data obtained, which will allow to evaluate the contribution of plantings to the reduction of urban transport noise.
Field observations and measurements were carried out for two years, in the spring-summer and autumn-winter periods on the territory of the city of Kharkiv. 10 profiles were laid down for the experiment:
No. 1. Bursatskyi Uzviz; No. 2. Kravtsova Str.; No. 3. Park named after T.H. Shevchenko; No. 4. Botanical Garden of Karazin Kharkiv National University (site on Klochkivska Str.); No. 5. Klochkivskyi Uzviz; No. 6. Kosmichna Str.; No. 7. Sosnova Hirka Park; No. 8. Novhorodska Str.; No.9. Forest Park (Pavlove Pole District); No.10. Forest Park (Pomirky District).
To measure noise levels from vehicles along each of the profiles (except profile No. 8), 5 points were laid
at distances of 0 m, 10 m, 20 m, 50 m and 100 m from the roads. The segments along profile No. 8, but included points located at distances of 0 m, 10 m and 20 m from the road. In most cases, the first row of plantings is located 5-7 m from the edge of the road.
In selected areas full-scale measurements of noise levels were carried out, the main source of which is vehicles. Measurements were performed using DT 8852 sound recorder according to the method specified in ISO 1996-1:2016 [10]. According to this method, measurements should be made at the distance of 7.5 m from the carriageway. But, based on the goal of the research of assessing the contribution of urban plantings to reducing traffic noise, five points were selected on each profile under study at the distance of 10 m, 20 m, 50 m and 100 m. The other conditions given in the methodology were unchanged. Field studies were conducted in dry, windless weather. Instantaneous noise levels were recorded during such period of time that at least 200 vehicles were traveling in all directions. The noise characteristic calculated to evaluate the condition of the acoustic environment in the territories of the residential areas is an equivalent noise level calculated in accordance with the procedure given in Appendix 1 of GOST 23337-78 [11].
Also, at each point, the presence and type of plantings were taken into account, their species composition in layers was described: the main tree cover (A), juvenile trees (Aj, young trees that have not yet been included in the composition of the main tree cover in height), shrub cover (Fr), and vegetation cover (H). For each cover, the degree of projective coverage was estimated (0 to 100% in 5% increments). The following parameters were measured separately for the main tree cover:
n - is the number of rows of plantings from the road to the measuring point;
B, m - is the width of the plantation stand from the outside relative to the boundary road to the measuring point;
H, m - mean tree height (measured using the angle meter IU-1M);
LCR, fracture units - live crown ratio, that is, the proportion occupied by a continuously crowned crown from the length of the tree trunk (average for a group of trees at a point);
CC h, unit fraction - canopy closure (the same as tree cover), that is, an indicator similar to the projective cover of the tree cover;
CC v, unit fraction - canopy closure (vertical projection), that is, the proportion of the apparent space of the plane perpendicular to the soil surface, which is covered by the leafy crowns of adjacent trees in a row (average for a group of trees at the point);
CT, unit fraction - crown transparency, that is, the total proportion of the total area of the imaginary contour of the living crown of a tree in different projections, which accounts for enlightenment among the leafed shoots;
r, m - average distance between trees in the planting;
a, deg. - surface slope (measured using the angle meter IU-1M).
All ground distances were measured using a tape measure.
The live crown ratio and crown transparency were measured using the forest plant health monitoring methodology adopted in Ukraine [12]. The vertical crown projection for this study was first proposed.
For each series of measurements, MS Excel 2013 software determines the point parameters of normal statistics, as follows:
- mean value (x);
- variance (c2);
- standard deviation (c).
The standard error of the average (Standard error) (s) was calculated by the formula:
* = i , (1)
where N is a sample size [13].
The reliability of the average was checked using Student's criterion for point estimates:
te = f (2)
The obtained value was compared with the standard (tabular) criterion value at the significance level of p>0.95 (a>0.05). The mean value is significant if the Student's criterion calculated is equal to or greater than the standard value [13].
The absolute reduction of the equivalent noise level at the distance R from the roadway was calculated as:
ABS(L(R) - L(0)) = |L(R) - i(0)| (3)
where L (R) is the equivalent noise level at the distance R from the roadway, dBA;
L (0) is the equivalent noise level at the distance of 0 m from the carriageway, dBA.
The relative noise level at the distance R from the roadway was calculated as:
L(R)
REL(L(R),L( 0})=^,
(4)
where L (R) is the equivalent noise level at the distance R from the roadway, dBA;
L (0) is the equivalent noise level at the distance of 0 m from the carriageway, dBA.
Research results. Taking into accounts that the points on different profiles differed in a number of
Table 1
Correlation matrix of features characterizing the absolute decrease in the equivalent noise level from vehicles at
planting parameters, such as the width of the planting stand, the height of the tree cover, the average distance between adjacent trees, the crown density, the ratio of the "living" crown, the crown density in vertical projection, crown transparency, etc., the Principal Component Analysis was conducted, as expected, will help to identify the leading factors associated with the parameters of the vegetation cover, which affect the reduction of noise pollution in both absolute and relative displays bodies. Among the set of values of the eigenvectors of the correlation matrices of all the considered attributes (variables), when selecting the main components, only statistically significant values of the eigenvectors of the variables were taken into account.
In both analyzes, the first two components generally accounted for only 68% of the total variance, which is insufficient to significantly reduce the dimension of the factor space [14] and, thus, to isolate 1-2 "leading" independent factors associated with particular traits of plantations in the studied points that could be used to construct the analytical regression model.
Table 1 shows the correlation matrix of features (Spearman's rank correlation coefficients, rS), which are used to characterize the vegetation cover at the studied points. It is noticeable that the absolute reduction of the equivalent noise level is closely correlated (r s = 0.84) with the distance to the linear noise source, the number of rows of plantings (r s = 0.87) and, slightly weaker, with the width of the planting stand from the edge to the measurement point (r s = 0.78) and the average height of the tree cover at the point (r s = 0.71). In turn, the number of rows of trees in the plantings and the width of the planting stand closely correlate with the distance to the linear noise source (correlation coefficients r s = 0.90 and r s = 0.93, respectively). Thus, these parameters are interdependent. The sign of "crown transparency" also correlates positively with the mean power with such parameters as the distance to the noise source, the number of rows of trees, the width of the planting band, the absolute reduction of the equivalent noise level.
Features R n B H LCR CCh CCv CT r FR HB a
R 1.00
n 0.90* 1.00
B 0.93 0.92 1.00
H 0.49 0.63 0.48 1.00
LCR -0.17 -0.29 -0.21 -0.46 1.00
CCh 0.19 0.21 0.17 0.20 0.20 1.00
CCv 0.06 -0.04 -0.06 -0.03 0.36 0.48 1.00
CT 0.46 0.53 0.44 0.66 -0.30 0.25 -0.06 1.00
r 0.43 0.36 0.39 0.41 -0.22 -0.10 -0.24 0.19 1.00
FR 0.35 0.22 0.20 0.05 0.18 0.56 0.47 0.10 0.24 1.00
HB -0.05 -0.02 0.03 -0.10 -0.05 -0.38 -0.31 -0.24 0.17 -0.20 1.00
a 0.38 0.19 0.34 -0.23 -0.06 -0.09 0.06 -0.03 0.06 0.25 -0.11 1.00
ABS (Le(R)-Le(0)) 0.84 0.87 0.78 0.71 -0.32 0.26 0.03 0.54 0.44 0.26 -0.04 0.08
Note: only statistically significant correlation coefficients at p>0.95 (a>0.05) are highlighted in bold.
Considering that all the planting parameters depending on the distance to the noise source, as well as the density of plantings in horizontal and vertical projections, together with the "live crown" ratio, are included in one main component, which explains about 52% of the variation in the absolute decrease in the equivalent noise level, it seems useful to identify how plantings of various types in the large city affect such decline.
At this stage of the analysis, the entire array of data series characterizing the reduction of equivalent noise (both in absolute and relative values) on different profiles, perpendicular to the roads in different conditions of the residential area, was divided into 4 groups and analyzed separately in each group. These groups were distinguished according to the "type of plantings" classification criterion, namely: there are no plantings, a forest park, parks, and other plantings (street and inland blocks).
The series of data reflecting the reduction of the equivalent noise level on the profiles, grouped by the presence of plantings of different types or their absence, depending on the distance to the linear noise source are shown in Fig. 1 and Fig. 2. To establish how these data series, differ, we used the nonparametric Kolmogorov-Smirnov (KS) statistical criterion for pairwise comparison of data series [15]. We compared both the series of data reflecting changes in the equivalent noise level along the profiles as a whole, and the data sets characterizing the set of measurement points at the same distances on the profiles belonging to different groups according to the "type of plantings" classification criterion. If the empirically calculated criterion for the given pair of values is greater than the tabular critical value for these conditions, these series differ significantly from each other.
Fig. 1.Reduction of the equivalent noise level from vehicles depending on the distance to the noise source in the investigated sections in m. Kharkiv, grouped by the presence or absence ofplantings of different types (vegetation period 2017 and 2018)
Relative equivalent level of urban traffic noise in tree stands of different types at various distances (m) from the source of noise
X REL(UHO)/Le(R}) no tree stands
I REL(Le(0)/Le(R)) forest
A REL|Le(0)/Le(R)) parks
♦ REL(Le(0)/Le{R)) other
0.90
0,80
0,70
♦- ■ _____♦
I- t. ^^T^—- ♦ _^ * y « 0,971e"0 x ~ —----, R2= 0,5522 ---x
1
0.60
R1-0,6047
y ^ 03211e1""1 u = 0,7108"
y = 0,9001e ?-M* ■ R' = 0,7551
i
10
20
30
40
50
60
70
80
90
100
Fig. 2. Relative reduction of the equivalent noise level from vehicles depending on the distance to the noise source in the investigated sections in m. Kharkiv, grouped by the presence or absence ofplantings of different
types (vegetation period 2017 and 2018)
According to Kolmogorov-Smirnov criterion, all rows showing the absolute reduction of the equivalent noise level on profiles relating to different types of plantations, as well as on control profiles (without plantings) depending on the distance to the linear noise source, are significantly different from each other at the
Table 2
Results of comparison of sets of values reflecting the absolute reduction of equivalent noise level on profiles pertaining to different types of plantings as well as on control profiles
level significance p>0.95 (a>0.05). This means that such decrease can be described by a family of equations constructed by regression analysis methods that reflect the dependence of the form:
y = ax + b (1)
Type of planting Kolmogorov-Smirnov criterion
without plantings forest park parks
without plantings
forest park 0,88 (0,07)
parks 1,08 (0,07) 0,90 (0,05)
others 1,03 (0,08) 1,16 (0,06) 1,01 (0,06)
Note: in tabs there are tabular critical values of the criterion
MS Excel ® software builds linear regression noise depending on distance to noise sources by plant-models of absolute reduction of equivalent transport ings of different types: for the forest-park area:
ABS(L(R)-L(0)) = 0.2091R+7.3175 (R2=0.6923) ; (2)
for parks:
ABS(L(R)-L(0)) = 0.1839R+5.7251 (R2=0.6920) ; (3)
for street and interland plantings:
ABS(L(R)-L(0)) = 0.1647R+2.7492 (R2=0.6059) ; (4)
city building without plantings:
ABS(L(R)-L(0)) = 0.1168R+2.3131 (R2=0.5169) (5)
Instead, no statistically significant difference between the groups of profiles was found for the data showing the relative equivalent noise level at different distances from the highway (in all comparisons the empirical Kolmogorov-Smirnov criteria were calculated to be less than the critical tabular values). Only by comparing the regrouped data series on relative noise levels for points that are within the same distance range, but in different profile groups based on "planting type", is a significant difference between forestry and parkland plantings on the one hand, and non-plantation profiles or with street / intra-quarter plantings, on the other (moving average method, level 5) at the distance of 20 m from the carriageway (KS emp = 0.33 > K-S theor. = 0.30). In this case, the plantings of the first group reduce the noise load level by 16-25% of the equivalent noise level at the edge of the roadway, and in street plantings, or in the absence of plantings, this reduction reaches only 8-15%. At other intervals, the difference was not statistically significant.
Conclusions. According to mean values, a decrease in the equivalent noise level from vehicles, both in absolute and in relative values, is most significant on profiles located in the stands of the forest park zone (from 10.5 dBA to 25.3 dBA, or from 15% to 34 % at distances from 10 to 100 m). Park-type plantings provide equivalent noise reduction of 7.8 dBA to 21.0 dBA (10% to 28%) at distances of 10 to 100 m, and plantations of other types - from 5.0 dBA to 18.3 dBA (6% to 25%) at distances of 10 to 100 m deep from the edge of the roadway. At the same time, on profiles where there are no plantings, the absolute reduction of equivalent noise levels ranged from 5-6 dBA to 10.8 dBA (from 8% to 14%) at the same distances.
Spearman's rank correlation has revealed that the most significant factor in reducing noise from linear sources in space is, in fact, the distance to these sources and the planting parameters that are closely related to this factor: the width of the plantation stand and the number of rows of trees in the planting.
Using Kolmogorov-Smirnov criterion, linear regression dependences of the absolute noise reduction by plantings caused by the operation of vehicles were constructed at different distances from the roadway, which can be used separately for different types of urban plantings, such as: forest parks, parks, street and interland, urban development without plants.
Thus, it is shown that the presence of plantings in urban areas in areas with heavy traffic has a significant effect on the magnitude of the noise load reduction as it moves away from the linear noise source. Researches and subsequent analysis have made it possible to conclude that, first, the relative reduction of noise load correlates with a distance of 20 meters from the roadway and, secondly, the closer the structure of the planting to the park or forest, the difference the noise load level is higher. It is established that there is a need for further analysis of the possibility of additional environmentally sound solutions that will reduce the noise load in urban areas.
References
1. Matvieieva O.V., Myronrnko O.V. Modern problems of metropolitan areas and residential quarters. Renovation methods in the dense building environment. Scientific Bulletin of Construction. No. (4)/98 -2019
2. DSTU-N.B.V.1.1-33:2013. Guidance on the calculation and design of noise protection in rural areas. Kyiv. 2014. 46 p.
3. Didkovskyi V.S. Fundamentals of Acoustic Ecology / V.S. Didkovskyi, V.Ya. Akymenko, O.I. Za-porozhets and others. - Kirovograd: Impex LTD, 2001.
- 520 p.
4. Karagodina I.L., Osipov I.A., Shishkin I.A. Fighting noise in cities - M.: Medicine, 1972. - 159 p
5. Is there evidence that environmental noise is immuno- toxic? Prasher D. Noise Health. 2009, № 11 (44), P. 151-155
6. Cardiovascular effects of noise. Babisch W. Noise Health. 2011, №13, 201-204
7. The influence of subjective reactions to noise on health effects of the noise. Job R.F.S. Environ. Int. 1996, № 22 (1), P. 93-104
8. The effect of transportation noise on health and cognitive development: a review of recent evidence. Clark C., Stansfeld S.A. Int. J. Comp. Psychol. 2007, № 20 (2), P. 145-158
9. Petrenko O.K. Impact of vehicle noise on human body // Ukraine in the European space. Business, Politics, Law Issues: Mat. VI International Research Practice Conference, Lviv University of Business and Law. - Lviv, 2010. - P. 46-48
10. ISO 1996-1:2016. Acoustics - Description, measurement and assessment of environmental noise
— Part 1: Basic quantities and assessment procedures. URL: https://www.iso.org/standard/59765.html
11. GOST 23337-78. Noise. Noise measurement techniques in residential areas and in residential and public buildings. [The term of revocation has been postponed in accordance with the order dated December 14, 2017 No. 418. The document has been renewed from April 26, 2019 to January 01, 2022 according to the order dated April 24, 2019 No. 111]: Official publishing house Moscow: USSR State Committee on Construction, 1978. 22 p.
12. Monitoring and increasing the sustainability of anthropogenically disturbed forests. Collection of Recommendations of H.M. Vysotskyi Ukrainian Research Institute of Forestry and Agroforestry / Reference: V.P. Voron and others. - Kharkiv: The New Word, 2011. 304 p.
13. Lakin H.F. Biometrics. Issue 4, reworked and ext. M.: Higher school 1990, 352 p.
14. Computer Biometrics / Ed. V.N. Nosov. M.: Moscow State University, 1990. 232 p.
15. Ermatraut E.R., Prisyajnyuk O.I., Shevchenko I.L. Statistical analysis of agronomic research data in STATISTICA 6.0 package. Methodical instructions. Kyiv: UAAS Sugar Beet Institute, 2007. 55 p.
