Microplastic pollution in urban streams across New Zealand: concentrations, composition and implications

Figures & data

Figure 1. Microplastic distributions across all 52 study sites, expressed as the number of microplastic items per m³ of water, by size categories (a) and shape categories (b). The violin plot outlines illustrate kernel probability density with the width of the shaded area representing the proportion of data found at this concentration (van der Bijl et al. 2015).

Figure 2. Microplastic composition in size classes (a) and identification classes (b).

Figure 3. (a) Mean total number (LN-transformed) of microplastic items caught during the 15-minute sampling period by region (DF = 4, p = 0.57, eta² = 0.06) and (b) mean fourth-root number of microplastics per m³ of water by region (DF = 4, p = 0.47, eta² = 0.07).

Table 1. The 16 most polluted stream sites based on microplastic concentrations per m³. Shown are total microplastic items per site caught in 15 minutes, total numbers extrapolated over 24 hours (based on the flow velocities measured during sampling), and microplastic concentrations/m³ for each size class and identification class.

Site	Stream name	Region	Total microplastics/15 min	Microplastics/m^3	Microplastics/24 hrs	Microplastic size class			Microplastic class
						250–500 µm	500–1000 µm	1000–5000 µm	Fibres	Fragments	Polystyrene
16	Tuhikaramea Stream	Waikato	4833	44.8	463,968	2613	1171	1049	901	514	3418
12	Rototuna Creek	Waikato	67	14.3	6432	6	58	3	57	10	0
36	Waiwhetu Stream	Wellington	264	8.3	25,344	2	240	22	72	6	186
37	Awamutu Stream	Wellington	53	7.7	5088	35	1	17	54	1	0
2	Papakura Stream	Auckland	572	7.3	54,912	21	547	4	543	29	0
18	Waitawhiriwhiri Stream	Waikato	254	5.1	24,384	66	179	9	164	85	5
29	Cannons Creek	Wellington	16	3.5	1536	1	1	14	14	2	0
48	Heathcote River	Canterbury	85	2.7	8160	33	18	34	41	44	0
7	Motions Creek	Auckland	208	2.3	19,968	16	77	115	137	68	3
1	Waimahia Stream	Auckland	47	2.1	4512	12	6	29	36	11	0
38	Kaiwharawhara Stream	Wellington	54	1.9	5184	1	49	4	5	36	13
27	Porirua Stream	Wellington	106	1.8	101,716	6	33	67	85	21	0

Figure 4. The four strongest (in relative terms) relationships from the multiple linear regression analysis. A. Total microplastic items (LN-transformed) versus channelisation (R² = 0.08, p = 0.05), B. 1000–5000 µm microplastics/m³ (LN) versus channelisation (R² = 0.07, p = 0.06), C. Microplastic fibres/m³ (LN) versus stream width (m) (R² = 0.10, p = 0.02), D. Polystyrene/m³ (LN) versus channelisation (R² = 0.06, p = 0.08). Note that 0.01 was added to the three microplastic concentration variables before LN-transformation because they contained zeros.

Table 2. Multiple linear regression outputs for microplastic (MP) response variables and the six continuous predictor variables (stream length, urban catchment length, proportion of urbanised catchment, water flow velocity, channelisation and stream width). Beads were too rare to be included in this analysis.

Multiple linear regression	r^2	p	Correlation/partial correlation
Total MP (LN-transformed)
Backward selection	0.14	0.02	–
Stream length	–	0.06	0.27
Channelisation	–	0.24	0.32
Forward selection	0.08	0.05	–
Channelisation	0.08	0.05	0.28
MP/m^3 (LN)
Backward selection	–	–	–
None	–	–	–
Forward selection	–	–	–
None	–	–	–
250–500 µm MP/m^3 (LN)
Backward selection	–	–	–
None	–	–	–
Forward selection	–	–	–
None	–	–	–
500–1000 µm MP/m^3 (LN)
Backward selection	–	–	–
None	–	–	–
Forward selection	–	–	–
None	–	–	–
1000–5000 µm MP/m^3 (LN)
Backward selection	0.07	0.06	–
Channelisation	–	0.06	0.27
Forward selection	–	–	–
None	–	–	–
Fibres MP/m^3 (LN)
Backward selection	0.10	0.02	–
Stream width	–	0.02	−0.32
Forward selection	0.10	0.02	–
Stream width	–	0.02	−0.32
Fragments MP/m^3 (LN)
Backward selection	–	–	–
None	–	–	–
Forward selection	–	–	–
None	–	–	–
Polystyrene MP/m^3 (LN)
Backward selection	0.06	0.08	–
Channelisation	–	0.08	0.24
Forward selection	–	–	–
None	–	–	–

Table 3. Comparisons of microplastic concentrations/m³ between our stream survey and Dikareva and Simon (2019) across the same five stream sites. All five sites were sampled at moderate flows in both studies.

Site	Mora-Teddy & Matthaei (sampled January 2019) items/m^3						Dikareva and Simon 2019 (sampled November 2016) items/m^3
	250–500 µm	500–1000 µm	1000–5000 µm	Fragments	Fibres	Total	63–500 µm	500–1000 µm	1000–5000 µm	Fragments	Fibres	Total
2/Pap	0.3	7.0	0.05	0.4	6.9	7.3	26	15	33	23	31	75
5/Ota	0.8	0.07	0.4	0.2	1.2	1.3	60	37	34	19	22	132
6/Pak	0.0	0.09	0.5	0.09	0.4	0.5	51	38	19	32	69	107
7/Mot	0.2	0.9	1.3	0.7	1.5	2.3	17	8	10	14	16	35
8/Meo	0.3	0.02	0.1	0.3	0.1	0.4	12	5	1	9	6	19

van der Bijl W, Thyselius M, Kotrschal A, Kolm N. 2015. Brain size affects the behavioural response to predators in female guppies (Poecilia reticulata). Proceedings of the Royal Society B: Biological Sciences. 282(1812):1132. doi: 10.1098/rspb.2015.1132

 Web of Science ®Google Scholar

Dikareva N, Simon K. 2019. Microplastic pollution in streams spanning an urbanisation gradient. Environmental Pollution. 250:292–299. doi: 10.1016/j.envpol.2019.03.105

 PubMed Web of Science ®Google Scholar

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.