Page 437 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

2. STHCSA

STHCSA is an important component of Spatio-temporal analysis since location and time are two critical
aspects of important events such as the Coronavirus epidemic. The outputs of such analyses can provide useful
information to guide the activities aimed at preventing, detecting, and responding to pandemic problems [12]. There
are various methods for Spatio-temporal analysis. In this study, Mann-Kendall statistics were used to detect trend
data and Getis-Ord Gi* statistic was used to identify hot/cold spots.
Network Common Data Form (NetCDF) is a file format to store multi-dimensional scientific data such as
temperature, humidity, disease, and crime. The netCDF cube is generated using the COVID-19 x, y, and time data as

x, y, and z-axis. Summarizes a collection of points into a netCDF by aggregating them into space-time bins. The
Mann-Kendall p-value and z-scores measuring the statistical significance of the trend in a hot spot (spatial clusters
of high values) or cold spot (spatial clusters of low values) at a location. A positive and negative z-score indicates an
upward and downward trend respectively [9].
Then the pattern in the spatial-temporal data was identified with Getis-Ord Gi* statistic based on
neighborhood distance and neighborhood time step parameter. The Getis-Ord Gi* statistic was calculated for each

bin as follows [13, 14]:
∑
= =1 (1)
̅

2
∑
= √ =1 − (2)
2
̅

∑ − ̅ ∑
=1
∗
=1
= (3)

2
∑ −(∑ ) 2
√ =1 =1

−1
Where xj is the value of feature x at location j, n is the number of data and wij are the elements of the weight
matrix.
∗
The is recorded as a z-score for each variable in the dataset. The more intense the clustering of high

values or hot spot is statistically for the larger positive z-scores. The more intense the clustering of low values or
cold spot is statistically for smaller negative z-scores. Based on p-value and z-scores of this statistic, 17 pattern types
include No Pattern Detected, New Hot Spot, Consecutive Hot Spot, Intensifying Hot Spot, Persistent Hot Spot,
Diminishing Hot Spot, Sporadic Hot Spot, Oscillating Hot Spot, Historical Hot Spot, New Cold Spot, Consecutive
Cold Spot, Intensifying Cold Spot, Persistent Cold Spot, Diminishing Cold Spot, Sporadic Cold Spot, Oscillating
Cold Spot and Historical Cold Spot were extracted [9].
3. Results
According to the WHO reports, data on COVID-19 confirmed cases and deaths were collected until March
21. The spatiotemporal cube was created for the global map of the world with the deaths and cases reported every

day. The pixel size was 200,000 meters, and the Behrmann was selected for the projected coordinate system. first, a
hot/clod spot map based on the Getis-Ord Gi * statistic was prepared for COVID-19 cases and deaths. Based on the