Spatial approaches to ecological population monitoring and management

Tuesday, June 15 at 05:45pm (PDT)
Wednesday, June 16 at 01:45am (BST)
Wednesday, June 16 09:45am (KST)

SMB2021 SMB2021 Follow Tuesday (Wednesday) during the "MS09" time block.
Note: this minisymposia has multiple sessions. The second session is MS10-ECOP (click here).

Share this


Tae-Soo Chon (Pusan National University/Ecology and Future Research Association, Republic of Korea), Fugo Takasu (Nara Women’s University, Japan)


Population stability is at stake in ecosystems due to abrupt establishment or unexpected extinction of species, resulting from anthropogenic disturbances (e.g., industrial development, climate change) directly and indirectly. In ecology population dispersal phases of entering, establishment and proliferation/extinction are presented on spatial domain through individual life processes including reproduction, death, movement and environmental adaptation. Consequently, the degree of internal and external constraints would be enormous during the dispersal processes: objective methods are required for expressing population dispersal effectively. In the symposium we focus on how methodologies could be applied to addressing complex spatial phases for monitoring and management either locally or globally. In small scale stochastic processes on spatial habitat change will be dealt with, utilizing models on behavior state determination (e.g., Markov chain, Hidden Markov model, machine learning). In large scale the causality of population increase/decrease will be addressed in spatio-temporal domain using spatially explicit models, individual based models, network models and other data-driven models to obtain information on monitoring, risk factor analysis, prognosis and management in population dispersal. The target populations for model applications include terrestrial/aquatic species and disease. The participants could join the session and share information on model developments and practical applications.

Tae-Soo Chon

(Ecology and Future Research Institute/Pusan National Univ., Republic of Korea)
"Patterning local cooccurrence patterns of Nutria individuals using Geo-self-organizing map applied to telemetry data"
Nutria populations expanded rapidly in Korea since 1990s, causing biodiversity loss, local habitat disturbances and agricultural damages in ecosystems. The geo-self-organizing map (Geo-SOM) was applied to radio-tracked individual data to cluster geographical areas in association with plant types, land cover states and biological parameters. The minimum nearest neighbor distances for the different sexes were overall in accord with the minimum distances for the same sex. Local cooccurrences of female and male individuals were negatively associated with male-male cooccurrences compared with female-female cooccurrences, suggesting male dominance in group formations. Movement and cooccurrence information extracted by Geo-SOM aids understanding of population dispersal to help formulating management strategies for nutria populations.

Thakur Dhakal

(Kangwon National Univ., Republic of Korea)
"Unraveling Behavior States of Wild Boar Movements in Habitat Transitions Using Hidden Markov Model"
Wild boar (Sus scrofa) population dispersal is a critical issue in Korea nowdays, being closely linked with epizootics of African swine fever. Understanding movement of wild boar is a key issue in predicting spatial advancement patterns of the population. Movements of animals, however, are highly complex and difficult to analyze. We addressed behavior states of wild boar individuals by applying the hidden Markov model (HMM) to field data. Movements of wild boar individuals were continuously tracked at the Bukhan Mountain, Seoul, Korea, with the interval of approximately 2 hours up to 313 days from June, 2018 to May, 2019. Observable events were expressed as visiting by wild boar individuals to habitats with different resources (forest, leaf types and water). Transition probability matrices (TPMs) and emission probability matrices (EPMs) were estimated according to different initial conditions. Self-organizing map (SOM) was utilized to cluster output parameters produced from initial conditions to find the global optimum of parameters. Characteristic TPMs were observed according to different number of states. The event with most favorable habitat with “broad-leaf and water” shows the maximum probability of visit in EPM, followed by the habitats with “coniferous-leaf and water”. As the number of states increased, other habitats including “coniferous-leaf without water” and “no-forest without water” had higher probabilities of visit in EPMs. HMM in linking with SOM is useful for addressing behavior states of movements of wild boar individuals and would provide basic information on monitoring wild boar population dispersal.

Sung-Won Hong

(Kyungpook National Univ., Republic of Korea)
"Ensemble species distribution models proved habitat characteristics coincidence of dead and living long-tailed gorals (Naemorhedus caudatus) according to extreme snowfall"
Ensemble species distribution models (SDMs) have been used to define the vulnerable areas for critically endangered species and establish the conservation planning. The long-tailed goral (Naemorhedus caudatus) is a critically endangered herbivore in South Korea. Despite government efforts to recover the population through reintroduction programs, the animal remains vulnerable to heavy snowfall. From March to June 2010, 24 animals were found dead due to heavy snowfall in the Wangpi Stream basin. In this study, we hypothesized that gorals that died due to snowfall are low-status individuals that lived in the sub-optimal or non-suitable areas. Using the occurrence data from extensive field surveys from 2008 to 2010 in the Wangpi Stream and the carcass location data as well, we (1) defined the goral habitat characteristics and (2) compared the habitat characteristics between dead and living gorals using ensemble species distribution modeling (BIOCLIM, Domain, generalized linear models, generalized additive models, random forests, boosted regression trees, classification and regression trees and Maxent). The ensemble models had high levels of goodness-of-fit and suggested that the sites where dead gorals were found were closely related to typical goral habitats. These results implied that the optimal goral habitats could become uninhabitable following heavy snowfall. Most of the dead animals were pregnant females or were young, implying that they could not escape their primary habitats due to lower mobility. Thus, when there is a climate catastrophe, the optimal goral habitats should be considered for rescue and artificial feeding.

Taeyong Shim

(Korea University, Republic of Korea)
"Evaluating Distribution Shifts of Invasive Largemouth Bass under Climate Change"
The spread of largemouth bass (Micropterus salmoides) is a rising concern in South Korea. This study aims to evaluate the distribution shifts of largemouth bass in South Korea using classification algorithms. The candidate classification algorithms include RF (Random Forest), C5.0 and cforest (Conditional Inference Random Forest) which are built in the caret package in R. Largemouth bass occurrence records and environmental variables (temperature, precipitation, flow, water quality, and topography) from 2011 to 2015 were used in model training. In training, grid and random searching methods were compared for identifying the hyperparameters within an algorithm (RF, C5.0, and cforest). As a result, grid searching applied RF showed the highest accuracy. RF showed that largemouth bass will shift to the upstream regions in the Han river. This study is expected to be helpful for predicting distribution shifts and establishing management policy of largemouth bass.

Hosted by SMB2021 Follow
Virtual conference of the Society for Mathematical Biology, 2021.