Published on 24 September 2021

Analysing iNaturalist data from the no-take marine reserves of the Hauraki Gulf

Using the iNaturalist-Mātaki Taiao API was able to obtain a dataset of verified and geo-referenced observations from the different no-take marine reserves of the Hauraki Gulf Marine Park.

Mediterranean Fanworm

Mediterranean Fanworm, Sabella spallanzanii. February 2021, Long Bay-Okura Marine Reserve


The Hauraki Gulf Marine Park/Ko te Pataka kai o Tikapa Moana Te Moananui a Toi (HGMP) is a marine area of more than 1.2 million hectares. The park currently hosts six no-take marine reserves that represent about 0.3% of the spatial extent of the gulf. The HGMP also hosts two cable protection zones (CPZs) that prevent any fishing activity in approximately 5% of the waters of the gulf. Multiple benefits since the establishment of the Cape Rodney-Okakari Point in 1975 (the first in the HGMP) have been extensively documented. Biodiversity benefits include crayfish population recovery (Kelly et al., 2002), snapper density increase (Willis et al., 2003), restoration of trophic cascades (Shears & Babcock, 2003) and potential snapper larvae dispersal (Le Port et al., 2017; Le Port et al., 2014).

Auckland, the most populous city in the country, and its globally connected port, are in the Waitemata Harbor, which is part of the HGMP. With the increase in vessel traffic during approximately the last 100 years, the number of non-indigenous species being established in the HGMP has also increased. Such species pose a risk to native ecosystems and fauna by being either being a direct, e.g. depredation, or indirect (e.g. habitat modification) threat. Past relevant studies have shown how a foreign crab may cause cascading ecological effects by predating taxa that plays an important role in benthic ecosystems (Townsend et al., 2014) and how the Mediterranean fanworm is capable of altering communities of bacteria and macrofauna in soft-sediment habitats (Atalah et al., 2019), among others.

This study makes use of publicly available data from the New Zealand iNaturalist instance (iNaturalist, 2021) available at iNaturalist is a global citizen science platform that hosts open curated records of species observed by its registered users. iNaturalist data has been used in scientific studies as a way to record rare species (Wilson et al., 2020), potentially help detect disease in mangroves (Rossi, 2017) and help describe new species (Winterton, 2020) just to name a few examples.


Using the iNaturalist API, I was able to obtain a dataset of verified and geo-referenced observations from the different no-take marine reserves of the Hauraki Gulf. Verified observations (research grade) have been identified and validated by other users other than the owner. GPS data, often extracted from the observation's attached image files, allow to place an observation within a given area, in this case within the polygons representing the marine reserves. The list of iNaturalist places (and their respective polygons) for the reserves is as follows:

The total length of the dataset, as obtained on the 24th of September 2021 is of 1763 observations. The data was processed using the Python 3.8.10 language (Van Rossum & Drake, 1995). The code to pull, store, process and plot the data is publicly available in this Github repository.


The table below (Table 1) shows a breakdown of the dataset for each reserve and the species observed (native vs introduced).

Marine Reserve Native species obs Introduced species obs Total obs
Cape Rodney-Okakari Point 660 6 666
Long Bay-Okura 585 53 638
Tawharanui 203 2 205
Te Whanganui-A-Hei (Cathedral Cove) 124 10 134
Motu Manawa-Pollen Island 45 38 83
Te Matuku (Waiheke Island) 31 6 37
All 1648 115 1763

Long Bay and Goat Island are the reserves with most records, with 638 and 666 respectively. Despite having a total of just 83 records, Motu Manawa-Pollen Island boosts the second highest number of introduced species observations with a value of 38 (Table 1, Figure 1). The reserve with the highest number of introduced species observations is Long Bay-Okura with 53. The fact that they are in close proximity to the Waitemata Harbor (or within the harbor itself in the case of Motu Manawa) is likely related to the number of recorded observations of introduced species.

Number of observations by marine reserve

Figure 1: Number of observations of native (orange) and introduced (blue) species for each marine reserve

Cape Rodney and Tawharanui have near zero-ratios of native vs introduced species observations, with just 6 and 2 observations of introduced species respectively (Figure 1, Figure 2).

Percetage of introduced and native species observatios by location

Figure 2: Percentage of observations of native (orange) and introduced (blue) species for each marine reserve

Native taxa richness (number of native species) is higher at Long Bay-Okura (164 spp) and the highest number of introduced taxa is at Motu Manawa (30 spp), followed by Long Bay with 25 spp (Figure 3).

Taxa richness by location

Figure 3: Taxa richness for each marine reserve

Tawharanui and Cape Rodney yield 81 and 128 native spp respectively, and only 2 introduced spp each (Figure 3), which aligns with the low number of observations of introduced species on such reserves (Table 1, Figure 1).

The most frequent introduced species of the entire dataset is the Mediterranean fanworm, Sabella spallanzanii, while the most frequent native species is the Australasian snapper, Chrysophrys auratus (Figure 4, Figure 5).

Introduced species by count

Figure 4: Introduced species by number of observations

Native species by count

Figure 5: Native species by number of observations


For readability purposes, records previous to the year 2000 have been discarded in the data visualisations that follow. There are only 2 records previous to the year 200, one from 1981 and another from 1988 which were obivously uploaded to iNaturalist retrospectively.

Basic time-series visualization allows to detect an increasing trend in regards to the number of observations being recorded each year, as shown in the 95%-interval confidence of the linear regression (Figure 6). With approximately 3 months left to the end of the year, 2021 is the year with the highest number of observations (Figure 7).

Observations regression plot

Figure 6: Linear regression for the annual number of observations (native and introduced spp combined)

Number of observations by year and type (introduced vs native spp)

Figure 7: Annual number of introduced (blue) and native (orange) species observations

The southern hemisphere summer months, ranging from November to March, have the higher rates of recorded observations (Figure 8). However, observations of introduced species remain at similar levels throughout the year, except during the months of June and July, when they sustain a drop.

Observations by month and type (introduced vs native spp)

Figure 8: Monthly number of introduced (blue) and native (orange) species observations

Spatial trends

Heat maps of observations within the marine reserves show how density of observations is higher near the shore, with the exception of the Cape Rodney reserve, where high density can be also found offshore (between the research center and Goat Island).

Note: zoom in and out of the map below to explore heat maps of each of the marine reserves.


The number of observations within each of the marine reserves is likely to be related to the accessibility and popularity among New Zealanders, particularly Aucklanders. Goat Island is a popular recreational destination and is estimated to receive 337,000 visitors per year (Hunt, 2008), probably more nowadays. The Long Bay Regional Park, which is adjacent to the Long Bay-Okura marine reserve, is known to attract nearly one million visitors each year (Auckland Regional Council, 2010), making it one of the most popular parks in the Auckland region. The third reserve with most observations is Tawharanui, which is also adjacent to a regional park. Tawharanui receives an estimate of 160,000 visitors per year and is popular for campers, surfers and scuba divers alike (Auckland Regional Council, 2010). The popularity of these three reserves and their proximity to the most populous city in the country are probably the reasons of why they hold the highest numbers of observations from the whole of the dataset.

The number of introduced species at both Manawa-Pollen Island and Long Bay-Okura is likely be related to their physical location in regards to the Waitemata Harbor, which hosts the port of Auckland, known to be the main point of entry of foreign marine species. Popular reserves located further away from the Waitemata such as Tawharanui or Cape Rodney-Okakari Point present a lower number of introduced species observations and introduced taxa richness, suggesting a better ecological state than those closer to the Waitemata.

Mediterranean fanworm, Sabella spallanzani, is the introduced species with the highest number of observations which is hardly a surprise knowing how well-established this species is in the Waitemata, Waiheke Island and the inner Hauraki Gulf. Perhaps more surprising is the presence of two plant species in the top-5 of introduced species; the mignonette vine (or Madeira vine), Anredera cordifolia, and pampas grass, Cortaderia jubata. The results reveal how land-based species are also included in the dataset either because they are species found below the high-tide mark (preferred habitat or washed up) or due to inaccuracies of the GPS data. Applying a taxa filter to the dataset would see observations of such species removed, which could provide a clearer picture of invasive vs native marine species only.

The increasing number of annual observations is a trend that is likely follow in the upcoming years. A similar tendency has been registered when analysing data from the iNaturalist platform, where the number of observations have been growing almost by two-fold each year. Despite the known limitations of citizen-science data, more submitted observations are likely to help provide a better foundation for future analysis, as long as curators or artificial intelligence engines can keep up with the task of identifying and validating submissions.


Atalah, J., Floerl, O., Pochon, X., Townsend, M., Tait, L., & Lohrer, A. M. (2019). The introduced fanworm, Sabella spallanzanii, alters soft sediment macrofauna and bacterial communities. Frontiers in Ecology and Evolution, 7, 481.

Auckland Regional Council (2010). Regional Parks Management Plan.

Kelly, S., Scott, D., & MacDiarmid, A. B. (2002). The value of a spillover fishery for spiny lobsters around a marine reserve in northern New Zealand. Coastal Management, 30 (2), 153-166. doi:10.1080/089207502753504689

Hunt, L. (2008). Economic Impact Analysis of the Cape Rodney Okakari Point (Leigh) Marine Reserve on the.Rodney district. A report prepared for the Department of Conservation

iNaturalist. Available from Accessed September 2021

Le Port, A., Montgomery, J. C., Smith, A. N. H., Croucher, A. E., McLeod, I. M., & Lavery, S. D.(2017). Temperate marine protected area provides recruitment subsidies to local fisheries. Proceedings of the Royal Society B-Biological Sciences, 284 (1865), 20171300

Le Port, A., Montgomery, J., & Croucher, A. (2014). Biophysical modelling of snapper Pagrus Auratus larval dispersal from a temperate MPA. Marine Ecology Progress Series, 515 , 203-215.

Rossi, R. E. (2017). Using inaturalist observations to detect disease in red mangroves (Rhizophora mangle) (No. e3326v1). PeerJ Preprints.

Shears, N. T., & Babcock, R. C. (2003). Continuing trophic cascade effects after 25 years of no-take marine reserve protection. Marine Ecology Progress Series, 246 , 1-16.

Townsend, M., Lohrer, A. M., Rodil, I. F., & Chiaroni, L. D. (2015). The targeting of large-sized benthic macrofauna by an invasive portunid predator: evidence from a caging study. Biological invasions, 17(1), 231-244.

Van Rossum, G., & Drake Jr, F. L. (2009). Python 3 reference manual. Centrum voor Wiskunde en Informatica Amsterdam.

Willis, T. J., Millar, R. B., & Babcock, R. C. (2003). Protection of exploited fish in temperate regions: High density and biomass of snapper Pagrus Auratus (Sparidae) in northern New Zealand marine reserves. Journal of Applied Ecology , 40(2), 214-227.

Winterton, S. L. (2020). A new bee-mimicking stiletto fly (Therevidae) from China discovered on iNaturalist. Zootaxa, 4816(3), 361-369.

Wilson, J. S., Pan, A. D., General, D. E. M., & Koch, J. B. (2020). More eyes on the prize: an observation of a very rare, threatened species of Philippine Bumble bee, Bombus irisanensis, on iNaturalist and the importance of citizen science in conservation biology. Journal of Insect Conservation, 24(4), 727-729.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 New Zealand License.