Methods for broad‐scale plant phenology assessments using citizen scientists’ photographs

Abstract

Plant phenology-the timing of plant life cycle stages such as flowering or leaf senescence-plays a critical role in terrestrial ecosystems and is known to be responsive to environmental changes (Rathcke and Lacey, 1985; Ollerton and Lack, 1992; Cleland et al., 2007; Chuine, 2010). The fingerprint of accelerating global change, including both global-scale climatic changes and their local-scale outcomes, along with human disturbance, may show its first biotic signs in disrupted phenologies. These disruptions can have significant consequences if they lead to phenological mismatches between plants and the animals that depend on them (Kudo and Ida, 2013; Mayor et al., 2017). Plant phenology data that cover broad scales have until recently only been available through monitoring networks, such as the National Phenology Network (NPN) (Schwartz et al., 2012; Rosemartin et al., 2014) in the United States, which coordinates amateur and professional scientists to make phenological observations. Although such networks provide critical data, reporting remains sparse because such networks often focus on key taxa or repeat sampling at a relatively small number of locations. Promising new resources, as well as historical resources (e.g., imaged herbarium sheets), that provide wider taxonomic and spatial coverage are becoming available digitally (Davis et al., 2015; Willis et al., 2017; Silva et al., 2018); however, these have attendant issues with sampling protocols and with proper annotation of phenological traits and species identification. An alternative set of resources that has yet to be broadly tapped for phenology studies comprises repositories of naturalist citizen science images. Here we focus in particular on iNaturalist (http://www.inatu ralist.org/) as a source of phenology data because it: (1) enforces the provision of species occurrence metadata required for scientific use; (2) manages taxonomic resources, putting a premium on quality identification, and sets objective requirements for records to be considered "research grade"; (3) allows reporting of cultivation status along with annotation of traits including phenology in metadata fields, although trait annotation is still not often used; (4) is growing at a rapid and

The accuracy of phenology estimators for use with sparsely sampled presence‐only observations

“…Researchers are increasingly using incidental data to estimate phenological metrics, in part due to exponential growth of digitized museum specimens and observers uploading their photographs on platforms like iNaturalist. These platforms, now containing tens of millions of imaged and identified organisms, can be leveraged to document phenological signals (Taylor & Guralnick, 2019), and provide a unique opportunity to answer questions across greater spatial and taxonomic scales (Barve et al., 2020). Additionally, more than 390 million specimens of preserved plants reside in herbaria across the world (Thiers, 2019) and more than 300 million arthropod specimens are housed in North American natural history collections (Cobb et al., 2019).…”

“…However, such standardized data are limited in space and time. Instead, the fastest growing data resources usable for documenting phenology come from incidental or opportunistic community science platforms, for example, iNaturalist, where users share and identify photographs of biodiversity across the globe (Barve et al., 2020). In 2019, 538,906 digital vouchers of butterflies were added to iNaturalist, a 1.8‐fold increase from 2018 (as of February 2020).…”

Using iNaturalist to monitor adherence to best practises in bat handling

“…Bats are implicated as reservoir hosts for a wide variety of potentially zoonotic viruses, including filoviruses, henipaviruses, lyssaviruses, and coronaviruses (Anderson et al 2019, Drexler et al 2014, Olival and Hayman 2014. It is rare that contact with bats is fatal to humans; it is probably more often fatal to the bat due to injury or stress from being handled.…”

Potential for online crowdsourced biological recording data to complement surveillance for arthropod vectors

“…Data from the iNaturalist platform are starting to be used by researchers for a variety of applications, and have been utilized to study urban biodiversity [ 17 ], confirm new country reports of the invasive ladybird Harmonia axyridis in Latin America [ 18 ], monitor urban red fox and coyote populations [ 19 ], aid mapping of global termite distribution and diversity [ 20 ], record the phenology of flowering in plants [ 21 ], monitor fish for signs of black spot disease [ 22 ], and study the distribution, phenology, and host plant associations of the queen butterfly Danaus gilippus thersippus [ 23 ]. Observations from iNaturalist could also be useful for the study of vector species of importance to public and veterinary health, but this potential has yet to be explored.…”