Improving data quality to build a robust distribution model for Architeuthis dux
Introduction
In recent years, niche models that estimate species distribution have become widely used in conservation biology (Guisan and Zimmermann, 2000). Rare species are examples where the prediction of suitable habitats is paramount to support fisheries management policies and conservation strategies (Pearce and Boyce, 2006, Márcia Barbosa et al., 2003). Defined by Cao et al. (1998) as species that occur at lower frequency or in low number in a sample of certain size, rare species have a key role in affecting biodiversity richness and by consequence they are indicators of degradation for aquatic ecosystems (Lyons et al., 1995, Cao et al., 1998). In this context, predictive models can considerably support the qualitative and quantitative criteria used to assign a “status” to a species (IUCN Species, 2001), by providing accurate, applicable and reliable spatial predictions to species population monitoring and sampling (Guisan et al., 2006). As discussed in many studies, the methodological progresses of Species Distribution Models (SDMs) allow nowadays to apply robust techniques to rare and endangered species (Guisan and Thuiller, 2005, Ferrier, 2002, Gibson et al., 2007, Razgour et al., 2011, Ovaskainen and Soininen, 2011, Rebelo and Jones, 2010, Wisz et al., 2008, Lomba et al., 2010).
Here, we propose a procedure to generate a niche model for a species of the giant squid family (Architeuthis dux), based on both presence and estimated absence locations. Our aim is to produce a map that is more accurate with respect to the ones that can be produced by commonly used models. Although giant squids have recently received special attention, little has been published regarding the population demographics and the ecology of these rare species. Most of the records refer to dead stranded animals, individuals captured alive by nets or from the remains found in the stomach of marine mammals (Clarke, 2006). When modelling the distribution of these species, high quality data are crucial but very scarce. This problem is especially important for rare species prediction, where models training is highly dependent on data quality.
Given this context, our study investigates a combination of presence only and presence/absences techniques to identify potentially suitable areas for A. dux subsistence. We also expect the results to help defining guidelines for use of SDMs for rare species.
We illustrate our approach using data from authoritative sources of observation records. Furthermore, we use an expert system to produce absence locations. In order to ensure high quality for the environmental variables associated to presence information, we use the maximum entropy (MaxEnt) model (Phillips et al., 2006, Berger, 1996) as a filter to select the variables that are important to define the potential habitat of the species. These are the variables that are mostly correlated to the species observations, among those we selected from reference studies. When possible, we make environmental variables values range from 450 to 1000 m, encompassing the deep ocean waters usually inhabited by A. dux (Guerra et al., 2010). Finally, we train an Artificial Neural Network on these datasets and compare the results with (i) a presence-only method, (ii) an expert system and (iii) an expert drawn map.
The paper is organized as follows: Section 2 reports the effort made to model or understand the potential habitat of rare species, and in particular of A. dux. Section 3 reports the details of our method and its expandability as a general approach to rare species modelling. Section 4 reports the results of both a qualitative and a quantitative comparison with other distribution maps for A. dux. Section 5 discusses the results and Section 6 draws the conclusions.
Section snippets
Overview
This section is divided into two subsections. The first reports the current understanding of the distribution of A. dux. The second describes the niche modelling approaches that have been applied or that can be applied to rare species.
Method
In this section we describe the technology which supported the experiments, and we also report our procedures for data preparation and environmental features selection. Furthermore, we explain our presence/absence approach to model the distribution of A. dux and its relevance for other rare species.
Results
In this section we describe the qualitative and quantitative approaches we used to compare the trained models with existing literature data. First, we report a “qualitative” comparison on coarse presence locations reported in literature for A. dux and Architeuthis spp. In order to investigate the differences between the models in detail, we also report the results of a quantitative comparison, with respect to a map drawn by an expert (Nesis, 2003).
Discussion
The results demonstrate that, according to a qualitative analysis, the simple topology FFNN gives the most promising results. In this scenario, the AquaMaps Suitable model is indeed the most stable. On the other hand, if we move to a quantitative evaluation with respect to an expert-drawn map, we better understand the differences between AquaMaps and the FFNN. AquaMaps presents few points in open ocean, because the model assigns more weight to the proximity of land, while the expert's map
Conclusions
In this paper, we have described a method to predict the distribution of A. dux at global scale. We have used a presence-only model to identify important environmental features possibly extracted at Architeuthis depth ranges indicated by other studies, we have generated absence locations using an expert system and we have retrieved presence records from two authoritative data sources. By means of a presence/absence model based on an Artificial Neural Network, we have produced a potential
Acknowledgments
The reported work has been partially supported by the i-Marine project (FP7 of the European Commission, INFRASTRUCTURES-2011-2, Contract No. 283644) and by the Giovanisì project of the Presidency of the Tuscan Regional Government.
References (94)
Species distribution models and ecological theory: a critical assessment and some possible new approaches
Ecol. Model.
(2007)- et al.
An infrastructure-oriented approach for supporting biodiversity research
Ecol. Inform.
(2015) - et al.
Combining simulated expert knowledge with neural networks to produce ecological niche models for Latimeria chalumnae
Ecol. Model.
(2013) - et al.
The giant squid Architeuthis: an emblematic invertebrate that can represent concern for the conservation of marine biodiversity
Biol. Conserv.
(2011) - et al.
Predictive habitat distribution models in ecology
Ecol. Model.
(2000) - et al.
Overcoming the rare species modelling paradox: a novel hierarchical framework applied to an Iberian endemic plant
Biol. Conserv.
(2010) - et al.
Otter (Lutra lutra) distribution modeling at two resolution scales suited to conservation planning in the Iberian Peninsula
Biol. Conserv.
(2003) - et al.
SPECIES: a spatial evaluation of climate impact on the envelope of species
Ecol. Model.
(2002) - et al.
Maximum entropy modeling of species geographic distributions
Ecol. Model.
(2006) - et al.
Using multi-scale modelling to predict habitat suitability for species of conservation concern: the grey long-eared bat as a case study
Biol. Conserv.
(2011)
Predicting the distributions of marine organisms at the global scale
Ecol. Model.
The encyclopedia of life
Trends Ecol. Evol.
Some aspects of the systematics and biology of squid of the genus architeuthis based on a study of specimens from newfoundland waters
Bull. Mar. Sci.
First record of a giant squid Architeuthis sp. Steenstrup, 1857 (Cephalopoda, Architeuthidae) in Brazilian waters
Bol. Inst. Pesca Sao Paulo
Definition and detection of vulnerable marine ecosystems on the high seas: problems with the “move-on” rule
ICES J. Mar. Sci. [Journal du Conseil]
Ocean monitoring and forecasting core services: the European Myocean example
Proc. OceanObis
Feed-forward neural networks
IEEE Potentials
A Brief Maxent Tutorial
Data integration: the ocean biogeographic information system
Life World's Oceans: Divers. Distrib. Abund.
Neural Networks for Pattern Recognition
Gut contents of a giant squid Architeuthis dux (Cephalopoda: Oegopsida) from New Zealand waters
N. Z. J. Zool.
Giant squids may die when exposed to warm water currents
Nature
Presence-absence versus presence-only modelling methods for predicting bird habitat suitability
Ecography
Species distribution modeling in the cloud
Concurr. Comput.: Pract. Exp.
D4science: an e-infrastructure for supporting virtual research environments
IRCDL
How important are rare species in aquatic community ecology and bioassessment?
Limnol. Oceanogr.
The iMarine Data Bonanza: improving data discovery and management through an hybrid data infrastructure
Boll. Geofis. Teor. Appl.
New records of the giant squid Architeuthis dux in the southern Indian Ocean
J. Mar. Biol. Assoc. UK
Do habitat suitability models reliably predict the recovery areas of threatened species?
J. Appl. Ecol.
Oceanic cephalopod distribution and species diversity in the Eastern North Atlantic Malcolm R. Clarke
Life Mar. Sci.
Cephalopod species in the diet of a sperm whale (Physeter Catodon) stranded at Penzance, Cornwall
J. Mar. Biol. Assoc. UK
Inferring the past to predict the future: climate modelling predictions and phylogeography for the freshwater gastropod Radix balthica (Pulmonata, Basommatophora)
Mol. Ecol.
The D4Science MaxEnt Implementation
Automatic procedures to assist in manual review of marine species distribution maps
Adaptive and Natural Computing Algorithms
Comparing heterogeneous distribution maps for marine species
GISci. Remote Sens.
Modeling species distribution with GIS
Research Techniques in Animal Ecology
Predicting the distribution of the crested Tinamous, Eudromia spp. (Aves, Tinamiformes)
J. Ornithol.
Interoperability of biodiversity databases: biodiversity information on every desktop
Science
Species distribution models: ecological explanation and prediction across space and time
Ann. Rev. Ecol. Evol. Syst.
A statistical explanation of MaxEnt for ecologists
Divers. Distrib.
An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo-absence data
J. Appl. Ecol.
Mapping spatial pattern in biodiversity for regional conservation planning: where to from here?
Syst. Biol.
Prediction and validation of the potential global distribution of a problematic alien invasive species-the American bullfrog
Divers. Distrib.
Climate change, plant migration, and range collapse in a global biodiversity hotspot: the Banksia (Proteaceae) of Western Australia
Global Change Biol.
The marine fauna of New Zealand: Cephalopoda:Oegopsida:Architeuthidae (Giant Squid)
NIWA Biodivers. Mem.
Mapping Species Distributions: Spatial Inference and Prediction
Dealing with uncertain absences in habitat modelling: a case study of a rare ground-dwelling parrot
Divers. Distrib.
Cited by (25)
Habitat distribution change of commercial species in the Adriatic Sea during the COVID-19 pandemic
2022, Ecological InformaticsAll the spermatangia on a female were implanted by single-pair copulation in giant squid Architeuthis dux
2021, Deep-Sea Research Part I: Oceanographic Research PapersCitation Excerpt :Current knowledge of these species has been gained primarily from anatomical (Norman and Lu, 1997; Bolstad and O'Shea, 2004; Hoving et al., 2004), chemical (Landman et al., 2004; Rosa et al., 2005) and genetic (Deagle et al., 2005; Winkelmann et al., 2013; Wada et al., 2020) investigations with a limited number of specimens, and from a very few expeditions by remotely operated and manned vehicles (Kubodera and Mori, 2005; Hoving et al., 2016; Shea et al., 2018; Burford and Robison, 2020). The giant squid Architeuthis dux is the largest invertebrate presumed to inhabit the world's mesopelagic (or even deeper) zone except in polar regions (Coro et al., 2015; Xavier et al., 2018). Curiously, along the Japanese coastline of the Sea of Japan, a markedly large number of well-grown individuals has been stranded on the shores or trapped in fishery set-nets in winter.
Studying the swift, smart, and shy: Unobtrusive camera-platforms for observing large deep-sea squid
2021, Deep-Sea Research Part I: Oceanographic Research PapersCitation Excerpt :Thus, there is a need to develop reliable methods for recording these elusive species in their natural habitats (Roper and Shea 2013; Guerra et al., 2018). Moreover, many large deep-sea cephalopods, such as the giant squid Architeuthis dux and the colossal squid Mesonychoteuthis hamiltoni, are ecologically important deep-sea predators (Cherel et al., 2009), serve as prey for deep-diving cetaceans (Judkins et al., 2015), and have extensive geographic ranges (Coro et al., 2015; Remeslo et al., 2019). Knowledge of the behaviour, distribution, and abundance of these species is therefore a key component to understanding deep-sea ecosystems (Hoving et al., 2014).
Predicting geographical suitability of geothermal power plants
2020, Journal of Cleaner ProductionA global-scale ecological niche model to predict SARS-CoV-2 coronavirus infection rate
2020, Ecological ModellingCitation Excerpt :ENMs have been extensively used to discover these characteristics directly, or indirectly by tracing viruses’ principal vectors (Fuller et al., 2013; Linden, 2006; Medley, 2010; Peterson et al., 2006; Samy et al., 2016; Signorini et al., 2014; Tachiiri et al., 2006; Valiakos et al., 2014; Walton et al., 2010; Zhu and Peterson, 2014). In particular, the Maximum Entropy model (MaxEnt) has been often used as an ENM due to its flexibility to work with both presence and presence/absence data scenarios (Coro et al., 2015b; 2013; Elith et al., 2011; Phillips et al., 2004). Also, MaxEnt can estimate the influence of each parameter on the identification of the niche, i.e. the most important parameters to understand a virus’ preferred conditions.
Enacting open science by D4Science
2019, Future Generation Computer SystemsCitation Excerpt :In fact, they have been designed and developed in the context of several projects and initiatives supporting real communities, often multidisciplinary ones, involved in research investigations in fields including biological sciences, earth and environmental sciences, agricultural sciences, social sciences and humanities. Concrete use cases have been discussed in previous works, e.g. [27–31]. For the sake of this paper we briefly report one recent experience in exploiting the just described offering for a computational biology case.