spatial pattern

Recognition of spatial pattern of plant species is essential to check the canopy position, regeneration, forest dynamics and identify biological relationships in an ecosystem. One of the main goals in analyzing the spatial pattern of trees in forest ecosystems is to discover more meaningful relationships between trees and their environment. For this purpose, special analytical methods are used to quantify the spatial pattern of plant communities. The aim of this study was to determine the spatial pattern of Populus euphratica Oliv. using various K, L, G and J functions and the ability of these functions in determining details spatial pattern of the species in riparian forests of Maroon, Behbahan.

Two stands of Populus euphratica including Pure and mixed were studied and callipered. Position (distance and azimuth) of all individuals with DBH larger than 5 cm was set. Spatial pattern of Populus euphratica in both pure and mixed stands was determined using Ripley's K function, L, G and J functions. For explanatory analysis in analysis of point statistics, the assumption test of complete randomness of points was considered. The usual hypothesis test in point analysis statistics was applied based on simulation tests such as the Monte Carlo significance test. This test was repeated 999 times so that the probability of the first type of error is at least close to 5%. Also, in order to check the homogeneity and heterogeneity of the studied areas, the Kolmogorov-Smirnov test was used, and R software (Var3.2.2) was used to analyze the distribution pattern.

Findings: 

The results related to the analysis of the spatial distribution pattern of the Populus using Ripley's K function in the pure stand indicate that the distribution pattern of the Populus up to a distance of approximately 150m is clumped and from this distance onwards, the pattern has changed to randomness and then to uniformity. A significant clumped pattern can be seen in the mixed stand up to a distance of about 125m. The results of the analysis of the heterogeneous L function in the pure stand show the spatial pattern of the field up to a distance of about 150m, a significant clump. In the mixed stand up to a distance of 125m, the pattern is aggregated and from this distance onwards, the pattern changes towards randomness and then significantly towards uniformity. The inhomogeneous function G in the pure stand of Populus up to a distance of about 100m shows a clumped pattern with a low slope. In the mixed stand of Populus from the initial intervals onwards, the curve with a low slope indicates a medium aggregated pattern. The value of the J index in the pure stand is equal to 1 up to a distance of about 30m, and the uniform state curve shows the existence of a random pattern up to this distance. The value of the index is less than 1 from the distance of 30 to about 75m with an almost steep slope, which indicates a relatively strong clumped pattern. In the mixed stand, the value of the index is less than 1 up to a distance of about 60m, with a steeper slope, and shows a strong clumped pattern.

All functions for two stands have confirmed similar results which are clumped up to a distance of about 100 meters. The J function provided more detail of the spatial pattern. Generally, the functions endorsed a type of ecological dependency between the Populus euphratica individuals.

Urbanization especially in big cities of developing and developed countries has major impacts on climate change by producing greenhouse gas and increasing average temperature, and thus creating urban heat islands(UHI). The unplanned urban development is one of the main factor, which is responsible for making such circumstances. Lack of enough attention to preserving natural and green infrastructure is one of the factors causes city warming and urban heat islands challenges that are important issues in urban environmental planning nowadays. Urban heat island consists of air temperature and surface temperature. Studies show that land cover planning and management can control surface temperature. The relationship between increasing the green spaces as an important element of the green infrastructure and decreasing surface temperature is already has been studied. Regarding the literature has been reviewed in this paper, the purpose of this study is to investigate and clarify the detailed relationship between the characteristics of spatial patterns of urban green spaces and their influences on surface temperature. Spatial composition and spatial configuration are two main elements of spatial patterns of urban green areas. Classification of green land cover based on Local Climate Zone (LCZ) help to discover the detailed relationship between each patterns’ components and the classified green spaces. The case under study is the city of Tehran, which has witnessed certain changes in relation to the development of built-up areas (both in form of planned and unplanned developments), reduction of green spaces and their spatial patterns, as well as rising average temperature.

In the process of investigating the relationship between urban spatial patterns of greenspaces in city of Tehran and land surface temperature, different methods and techniques are applied. The greenspace classification map of the city of Tehran was produced with the help of Landsat 8 satellite (2019) and LCZ method of land use classification, which divides green areas into 4 classes as follows:A; heavily wooded landscape of deciduous and/or evergreen trees. Land cover mostly pervious(low plants). Zone function is natural forest, tree cultivation, or urban park.
B; Lightly wooded landscape of deciduous and/or evergreen trees. Land cover mostly pervious(low plants). Zone function is natural forest, tree cultivation, or urban park.
C; Open arrangement of bushes, shrubs, and short, woody trees. Land cover mostly pervious (bare soil or sand). Zone function is natural scrubland or agriculture.
D; Featureless landscape of grass or herbaceous plants/crops. Few or no trees. Zone function is natural grassland, agriculture, or urban park.
Kappa coefficient and overall accuracy of this map was 0.8706, 88.172 percent, which confirms its accuracy. The next step was selecting landscape metrics. Based on the aim of the study and the reviewed literature spatial composition and spatial configuration are selected as two main elements of spatial patterns of urban green areas. The relationship between land cover patterns and surface temperature is analyzed and discussed by using Pearson and Pearson Partial correlation method.

The result of Pearson correlation analysis showed that there is a significant and negative correlation between spatial composition of A, B and D land cover classes with surface temperature. The highest negative correlation belongs to class B (scattered trees) and the lowest belongs to class A (dense trees). In contrast to these negative correlations, the correlation coefficient of class C with surface temperature is positive and significant.
The result of Pearson correlation analysis regarding spatial configuration showed that the average size of each green space class has a continuous and significant negative relationship with the surface temperature, though, the size of these correlations varies in different classes. The correlation also showed that besides size and significance, the direction of green marginal density of each class also differs. It should be noted that the surface area of green space classes (as a composition metric) has a great impact on the results so that the correct and clear correlation of configuration metrics with temperature could not be distinguished. This issue was resolved by using Partial Pearson correlation coefficient and controlling the effect of Class Area metric. As a result, the relationship between configuration metrics and LST changed significantly. Before controlling the Class Area metric, almost all metrics were correlated with LST, however, the new detailed findings showed that only the Mean Size of Patches in A and D classes and Edge Density in B and C classes had a significant relationship with surface temperature.
The study shows that spatial composition of green spaces in Tehran in relation to the Class area of classes A, B and D had a negative and inverse relationship with surface temperature. Class B, located in the east and west of Tehran, has the highest negative correlation. Class A in the east and center of the city with the lowest surface area and its scattered distribution pattern in comparison to other classes has the least negative correlation with surface temperature (95% confidence level). Class D, located mostly in the south and west of the city, has a negative relationship between Class area and temperature at the 99% confidence level. The correlation of spatial composition of class C in the northern half of the city is not like the other three classes and indicates a positive and significant relationship with surface temperature due to the presence of shrubs and grasslands with low density, scattered shrubs, and soil.
Regarding the partial Pearson correlation of spatial configuration metrics, the Mean Patch size of Class A at 99% confidence level shows a negative and significant relationship with temperature, but due to its subdivision and uneven distribution of green space in this class, the effect of this class in the reduction of temperature is not significant. The Mean Patch size of Class D has a significant negative relationship with surface temperature at 95% confidence level, although its cooling effect is not considerable.
Both Edge Densities of classes B and C at 95% confidence level had a significant positive correlation with surface temperature, but as trees did not exist in a dominant and dense manner to cause shading and temperature adjustment in these type of greenspace classes, a positive correlation between the Edge Density of them and surface temperature is occurred.

This paper has demonstrated the relationship between urban heat islands and spatial patterns of green spaces in Tehran city. The literature based study showed the scope of the problem explaining that urban green spaces contribute to mitigate climate change impacts through decreasing the surface temperature. The spatial form and pattern of urban green spaces have different effect on surface temperature as indicated in several studies. Importantly, planners and designers, need more detailed studies to take into account the relation between effects of spatial composition and configuration of different classification of plants and their effects on urban surface temperature. In this research, greenspaces patterns was studied using Local Climate Zone(LCZ) method and correlation of spatial pattern (composition and configuration) of each of LCZ green classes with the surface temperature were provided. The results of the analysis of the spatial composition of these classes showed that tree canopy greenspaces in both cases of high / low density and low plants has a reducing effect on temperature, but low-density grasslands and scattered shrubs with soil cover, has a positive relationship with temperature. More detailed results on the spatial configuration shows that only the mean patch size in dense tree areas and low plants has a significant negative correlation with temperature. But Edge density of scattered trees and open arrangement of bushes had a significant positive relationship with temperature. Thus urban green space planning and management, through determining the type, composition, and configuration of existing patterns and their improvements based on their effect on the reduction of surface temperature will help to decrease urban heat island impacts.