Table 2. Diversity estimates of the microbial communities at different locations and units at investigated waterworks.

Water works	Unit	Zone	Number of sequencesa	Observed OTUsb	Richness estimates			Evenness	Diversity
					Singleton OTUs0.03s	Chao1	ACE	GiniCoff	H
DWTP1	GW		6758	372	192	1005	890	0.841	8.5±0.02
	PF	Top	12 050±2396	322±4	175±3	546±81	641±99	0.890±0.01	4.1±0.18
		Depth	12 414±3826	361±10	193±34	604±130	742±159	0.881±0.01	4.2±0.24
	AF	Top	13 908±787	439±66	257±56	779±130	909±161	0.862±0.01	4.8±0.17
		Depth	14 361±1988	497±69	296±47	959±155	1100±178	0.855±0.01	4.8±0.40
DWTP2	GW		6392	1105	772	3046	3352	0.683	6.1±0.03
	PF	Top	7738±553	393±47	284±31	1142±182	1337±182	0.895±0.01	4.3±0.20
		Depth	7113±622	406±30	286±26	1406±59	1410±97	0.888±0.01	4.5±0.05
	AF	Top	9869±3891	512±45	340±49	1361±362	1525±330	0.862±0.01	4.6±0.09
		Depth	11 757±4330	480±62	301±71	1275±473	1371±505	0.868±0.01	4.5±0.02
DWTP3	GW		18746	863	527	1695	1895	0.775	6.8±0.04
	PF	Top	6803±1332	470±69	343±55	1639±302	1719±336	0.929±0.01	3.1±0.60
		Depth	6988±751	564±50	427±58	1893±233	2241±401	0.879±0.04	3.3±0.43
	AF	Top	6406±1228	802±81	589±79	2694±494	3008±520	0.721±0.03	7.6±0.01
		Depth	6803±1332	784±105	590±114	2580±591	2973±708	0.716±0.03	7.5±0.41
DWTP4	GW		16 480	1169	676	2170	2373	0.667	8.6±0.03
	PF	Top	6397±930	239±45	161±39	750±202	770±296	0.879±0.01	4.0±0.37
		Depth	7498±888	357±42	258±34	1222±165	1311±236	0.852±0.01	4.6±0.26
	AF	Top	6880±952	1106±192	850±184	4368±1194	4932±1526	0.784±0.02	6.4±0.08
		Depth	6390±782	1104±176	844±164	4599±1715	5013±1689	0.792±0.01	6.3±0.47
DWTP5	GW		16 648	962	593	1808	2183	0.735	7.5±0.03
	AF	Top	8084±740	343±68	241±45	1131±317	1233±354	0.909±0.01	3.4±0.36
		Depth	7650±1026	562±181	422±142	1913±742	2136±853	0.859±0.04	4.6±1.02

Abbreviations: ACE, abundance-based coverage estimator; AF, after-filter; DWTP, drinking water treatment plant; G_coff, Gini coefficient; GW, groundwater; H, shannon index; OTU, operational taxonomic unit; PF, prefilter.

All sample sizes were equalized at 4000 sequences. Standard deviations (±) were obtained from biological replicates; standard deviations in the top and bottom layers were assessed from triplicate samples from the same filter. GW samples were taken as individual samples.

^aNumbers of sequences in PF and AF rows show the total amount of high-quality sequences that were obtained from 10 ng of DNA extracted from 0.5 g of drained wet sand grains. Mean values were calculated by grouping the samples according to zones, units and waterworks.

^bOTU calculations were carried out at 97% phylogenetic similarity.