Daily rainfall nearest neighbor pattern using point data series in Iran

In this data study, assessment of daily rainfall nearest neighbor׳s patterns (DRBBP) was described in Iran. This article presents some spatial patterns of daily rainfall nearest neighbor׳s patterns for Iran from 170 stations and 31195 rainfall points by comparing ordinary kriging techniques based on the forecast models. For the nearest neighbor׳s patterns of the daily rainfall, rainfall data series of 1975–2014 was employed to estimate the point data of daily rainfall. The statistical properties were analyzed to indicate an increase in dispersed variability patterns of daily rainfall in Iran. Dispersed patterns were selected as the best nearest neighbor׳s models to model daily rainfall variability. The data results will help climatologists and hydrologists in model assessment and planning of natural environment in Iran.


a b s t r a c t
In this data study, assessment of daily rainfall nearest neighbor's patterns (DRBBP) was described in Iran. This article presents some spatial patterns of daily rainfall nearest neighbor's patterns for Iran from 170 stations and 31195 rainfall points by comparing ordinary kriging techniques based on the forecast models. For the nearest neighbor's patterns of the daily rainfall, rainfall data series of 1975-2014 was employed to estimate the point data of daily rainfall. The statistical properties were analyzed to indicate an increase in dispersed variability patterns of daily rainfall in Iran. Dispersed patterns were selected as the best nearest neighbor's models to model daily rainfall variability. The data results will help climatologists and hydrologists in model assessment and planning of natural environment in Iran.
& 2018 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Subject area
Climatology and Atmospheric environment More specific subject area Spatial statistics, GIS applications, Environmental science Type of data

Value of the data
The data could be suitable for climatic analysis in predicting of rainfall variability for climate classification in Iran.
The data base could provide viewpoints of variability patterns of the daily rainfall using nearest neighbor's techniques in Iran.
The dataset will help climatologists and hydrologists in model fitting and planning of natural environment in Iran.
The data will be helpful in predicting of the variability patterns of the daily rainfall through which the environmental planning.

Data
The rainfall data series for this study was extracted from the Meteorological Organization of Iran (IRIMO) from 1975 through 2014 from 170 synoptic and climatology stations in Iran. The point dataset to estimate the daily rainfall nearest neighbor's patterns (DRBBP) for all stations is extracted from a rainfall dataset of the stations. Rainfall data series obtained from the IRIMO differs from rainfall datasets predicted by the ordinary kriging in the data distributions are estimated on models of the ordinary kriging. In addition, rainfall validation datasets (RVD) were extracted from ordinary kriging models based on its spatial coordinates calculated by [1,2]: ð Þ this is a second-order polynomial trend surface on the spatial x-and y-coordinates. The point dataset collected are used based on observed mean distance, expected mean distance, nearest neighbor index, z-score, and p-value to assess the nearest neighbor's patterns. However, the point daily rainfall dataset across the rainfall points 31195 is a suitable tool to analyze the nearest neighbor's patterns. DRBBP depends on dataset extracted methods such as the ordinary kriging, and the Table 1 Daily rainfall data mean from Ordinary kriging models.

Method Circular Spherical Tetraspherical Pentaspherical Exponential Gaussian Rational Quadratic
Hole Effect  Table 3 The statistical properties of the data extracted from ordinary kriging models.
OMD is the observed mean distance between each rainfall point and its nearest neighbor, EMD is the expected mean distance in a random distribution, d i is the distance between rainfall point and its nearest neighboring, n is the total number of rainfall points, and A is the area of a minimum enclosing rectangle around all rainfall points. For the average nearest neighbor ratio (Moran's I index), a Moran's I index value less than 1 shows a pattern clustering (less spatial variability), while a Moran's I index value greater than 1 indicates a dispersion pattern (more spatial variability). In this analysis, Moran's I index value equal to 1, shows a random pattern the statistically significant the random of daily rainfall series. The quality and characteristics of the point dataset can be studied using spatial statistical methods tools such as calculating areas(for each rainfall point in a polygon rainfall points dataset), estimating distance band from neighbor count (the minimum, maximum, and average distance to the specific Nth nearest neighbor), switching rainfall points data (transfers rainfall points to weighted rainfall point data), and converting spatial weights matrix to table (transforms a binary spatial weights matrix file to a table) for a dataset of rainfall points. The rainfall point dataset is a climatic time series affecting the climatic variability in the climate classification. Rainfall point data sets were extracted from the GIS-based layers using the ordinary kriging based on the Circular, Spherical, Tetraspherical, Pentaspherical, Exponential, Gaussian, Rational Quadratic, Hole Effect, K-Bessel, J-Bessel, and Stable models. Therefore, the estimated daily rainfall dataset over the forty years is a suitable typical data series of the spatial variability of RVD as revealed in Tables 1-5. The nature and characteristics of the daily rainfall dataset can be studied using spatial statistical methods tools such as daily rainfall nearest neighbor's patterns (DRBBP).   Table 1 presents the mean estimation of the data extracted from ordinary kriging. Fig. 1 also presents the data. Fig. 1 describes the statistical variability properties revealing the spatial variability of the daily rainfall in Iran.
1.2. The mean estimation of the point data from ordinary kriging Table 2 presents the mean estimation of the data extracted from ordinary kriging. Fig. 2 also presents the data. Fig. 2 depicts the statistical variability properties revealing the point daily rainfall in Iran.
1.3. The statistical properties of the data from ordinary kriging Table 3 presents the statistical properties of the data extracted from ordinary. Fig. 3 also presents the data. Fig. 3 describes the statistical variability properties revealing the spatial variability of the daily rainfall in Iran. Table 4 presents the statistical properties of the data extracted from point ordinary kriging. Fig. 4 also presents the data. Fig. 4 depicts the statistical distribution variability properties revealing the spatial variability of the abnormal distribution of the daily rainfall in Iran. Table 5 presents the statistical estimation of the data extracted from DRBBP in Iran. Fig. 5 also presents the data. Fig. 5 depicts the nearest neighbor's patterns properties revealing the dispersed variability pattern of the daily rainfall in Iran.

Materials and methods
Several studies have been presented on nearest neighbor's patterns properties of rainfall [4][5][6][7][8][9][10]. The rainfall data for this analysis were collected from the IRIMO from 1975 through 2014 from 170 synoptic and climatology stations in Iran. The rainfall dataset extracted and compared in the climatology primary data such as daily rainfall point series was employed as recorded variables for the present article result.