Spatial and temporal variations of PM_2.5 concentrations in Chinese cities during 2015-2019

Figures & data

Figure 1. Spatial distribution maps of PM_2.5 concentrations in China during 2015–2019.

Table 1. Global spatial autocorrelation of PM_2.5 in China, 2015–2019.

Year	Moran’s I	Z-Score	p-Value
2015	0.1871	14.8168	<0.0001
2016	0.1977	15.5776	<0.0001
2017	0.1726	13.7575	<0.0001
2018	0.2164	17.1120	<0.0001
2019	0.1976	15.7266	<0.0001

Figure 2. Local spatial autocorrelation cluster maps of PM_2.5 in China, 2015 to 2019.

Table 2. Results for yearly spatial clusters of PM_2.5 in China from 2015 to 2019.

Cluster type	Cities (n)	Radius (km)	Number of cases	Mean inside	Mean outside	Standard deviation	Log likelihood ratio(LLR)	p-Value
Most likely	25	272.48	101,058	85.85	54.82	31.75	36,278.02	<0.0001
Secondary	1	0	3796	110.55	60.47	33.77	4111.39	<0.0001
2nd Secondary	11	521.75	33,541	67.65	60.37	33.99	718.97	<0.0001
3rd Secondary	1	0	2968	75.23	60.75	34.02	266.87	<0.0001
4th Secondary	28	362.33	86,878	63.05	60.39	34.02	219.98	<0.0001
5th Secondary	5	182.57	14,777	62.17	60.8	34.04	11.75	0.0013

Figure 3. The spatial and spatiotemporal clusters of PM_2.5 in China, 2015–2019. (a) The spatial clusters of PM_2.5. (b) The spatiotemporal clusters of PM_2.5.

Table 3. Results for spatiotemporal clusters of PM_2.5 in China from 2015 to 2019.

Cluster type	Cluster time	Cities (n)	Radius (km)	Number of cases	Mean inside	Mean outside	Standard deviation	Log likelihood ratio (LLR)	p-Value
Most likely	2015/1-2017/2	25	272.48	53,017	100.76	56.31	31.27	44,131.64	<0.0001
Secondary	2016/12-2017/1	36	938.64	6683	108.33	60.22	33.6	6673.72	<0.0001
2nd Secondary	2015/11-2015/12	37	708.92	7113	105.78	60.21	33.62	6362.47	<0.0001
3rd Secondary	2015/12-2016/1	35	366.74	6284	93.96	60.43	33.84	3029.51	<0.0001