Skip to main content
Dryad logo

Rubble Biodiversity Samplers (RUBS): 3D-printed coral models to standardise biodiversity censuses


Wolfe, Kennedy; Mumby, Peter (2020), Rubble Biodiversity Samplers (RUBS): 3D-printed coral models to standardise biodiversity censuses, Dryad, Dataset,


1. To ensure standardised, quantitative and repeatable methodologies, marine ecologists have engineered a range of artificial units to survey benthic communities with varying designs depending on target taxa, life history stage and habitat. In tropical ecosystems, autonomous units have typically lacked microhabitat complexity (e.g. planar tiles), short-term efficacy (> 1 y deployment) and/or a truly standardised design to sample cryptobenthic diversity.
2. Coral rubble is characterised by high microhabitat complexity, which is unresolved in sampling efforts. Yet, rubble can support the greatest density and diversity of metazoan taxa.
3. We engineered Rubble Biodiversity Samplers (RUBS) as a dismantlable lightweight 3D-printed model to standardise cryptobenthic biodiversity sampling.
4. We demonstrate the effectiveness of RUBS with preliminary results from short-term (1–14 days) deployments in Palau, Western Micronesia.
5. RUBS uphold the microhabitat complexity inherent in coral rubble (~50% interstitial space) and provide an effective and efficient methodology to sample cryptobenthic fauna, which could be applied in comparative studies.
6. The RUBS concept provides a platform for the broader implementation of 3D-printed models in marine ecology to overcome issues of standardisation regardless of habitat type.


PRINTING: The 3D-printed RUBS model consisted of four independent layers of artificial coral rubble designed to be stacked and joined using bolts and wingnuts through intentionally engineered holes. Each layer (ca. 140 x 90 x 20 mm) was created in Meshmixer (Version 3.5.474) by dissecting an existing branching coral (Acropora) design ( Replicate RUBS models were printed using grey polylactic acid (PLA) filament in Fused Deposition Modelling printers (Flashforge Guider IIs) at UQ Innovate (University of Queensland, Australia), which was relatively cost effective (~AU$15 RUBS-1).

FIELD COLLECTION: RUBS were deployed in shallow (1–2 m depth) coral rubble habitat in Palau, Western Micronesia (7˚16’4.3” N, 134˚27’7.6” E), November­–December 2019. The site was accessed by boat and fieldwork conducted on snorkel. Sixteen RUBS units were used on rotation, deployed with a small weight attached to the base in novel depressions (~8 cm depth) made carefully in the rubble by hand. For retrieval, RUBS were lifted from their depressions directly into individual plastic collection bags underwater and transferred in buckets to the Palau International Coral Reef Center (PICRC) where they were processed immediately. All fauna >250 µm were measured and identified (namely to the family level) using a dissecting microscope.

NATURAL RUBBLE COMPARISON: Small plastic mesh (5 x 5 mm) baskets were engineered to be the same dimensions as the RUBS (14 x 9 x 8 cm). Baskets were carefully filled with rubble in situ and embedded into novel depressions for 1–3 days before retrieval. Retrieval and sampling procedures were as for RUBS, above.

Usage Notes

The data are in two parts: motile cryptofauna collected from (1) RUBS and (2) rubble-filled baskets. The columns 'wdv' indicates water displacement volume of the RUBS and rubble baskets for density calculations. 

The RUBS models can be found at:


Centre of Excellence for Coral Reef Studies, Australian Research Council