Classification of unlabeled observations in Species Distribution Modelling using Point Process Models

Abstract

71 Abstract 8 1. Species distribution modelling, which allows users to predict the spatial distribution of species with the 9 use of environmental covariates, has become increasingly popular, with many software platforms providing 10 tools to fit species distribution models. However, the species observations used in species distribution 11 models can have varying levels of quality and can have incomplete information, such as uncertain species 12 identity. 13 2. In this paper, we develop two algorithms to reclassify observations with unknown species identities 14 which simultaneously predict different species distributions using spatial point processes. We compare the 15 performance of the different algorithms using different initializations and parameters with models fitted 16 using only the observations with known species identity through simulations. 17 3. We show that performance varies with differences in correlation among species distributions, species 18 abundance, and the proportion of observations with unknown species identities. Additionally, some of the 19 methods developed here outperformed the models that didn't use the misspecified data. 20 4. These models represent an helpful and promising tool for opportunistic surveys where misidentification 21 happens or for the distribution of species newly separated in their taxonomy. 22 23

Goitered Gazelle’s (Gazella Subgutturosa) Habitat Desirability Modeling by Using Maximum Entropy (Maxent) Method

“…One of the main shortcomings of distribution model predictions is a lack of reliable information on species absence (Jiménez-Valverde et al ., 2007). Species distribution models (SDMs) based on presence-absence or presence-only data (Guilbault et al ., 2019) are used widely in biogeography to characterize the ecological niche of species and to predict the geographical distribution of their habitat (Naimi et al ., 2014). The presence-only data only contains information about species presence, in contrast to presence-absence data which records both where species have been found present and where they have not been found (Warton & Shepherd, 2010; Renner et al ., 2015).…”

“…The presence-only data only contains information about species presence, in contrast to presence-absence data which records both where species have been found present and where they have not been found (Warton & Shepherd, 2010; Renner et al ., 2015). Although presence-absence data is generally of higher quality, it is also less common than presence-only data because it requires more rigorous planning to visit a set of predetermined sites (Van Strien et al ., 2013; Ruete & Leynaud, 2015; Guilbault et al ., 2019). The presence-only data allow for easy public and private involvement in biological monitoring and are the dominant source of species occurrence data (Elith et al ., 2011; Pédarros et al ., 2020).…”

“…In recent years, many techniques for modeling species’ niches and distributions have been developed (Guisan & Zimmermann, 2000; Guisan & Thuiller, 2005; Kozak et al ., 2008; Peterson et al ., 2011; Radosavljevic & Anderson, 2014). Computer tools such as a large number of algorithms and software platforms (Guilbault et al ., 2019) with Geographic Information System (GIS) (Traill & Bigalke, 2007), and statistical modeling techniques to draw up predictive maps (Jiménez-Valverde et al ., 2007; Naqibzadeh et al,. 2021) have been become more widely used in ecology and conservation biology (Guisan & Zimmermann 2000; Warren et al ., 2008; Kamyo & Asanok, 2020).…”