ESTIMATING SITE OCCUPANCY AND SPECIES DETECTION PROBABILITY PARAMETERS FOR TERRESTRIAL SALAMANDERS

AbstractUnderstanding species trend, decline or growth, is vital to further conservation efforts. Species-habitat relationship studies are equally important for conservation as it helps in understanding the habitat a particular species depends upon, i.e. habitat conservation. However, rare and endemic species are inherently difficult to study and occupancy models are especially useful in such cases. We conducted the first detection, non-detection survey for the white winged duck in Dehing Patkai Wildlife Sanctuary, India to assess site occupancy and test habitat factors that explain its occupancy. We found that white winged duck occupancy was low (0.27 ± 0.21 SE) and detection probability was 0.44 ± 0.30 SE. We found that increasing tree richness and decreasing elevation increased species occupancy. Detection probability was influenced by our effort in that detection increased with increasing number of survey hours. Using two standard approaches, we estimated the optimal number of sites and replicate surveys for future occupancy studies. We further present considerations for future surveys. Considering the sporadic and fragmented information available, we recommend long-term ecological research to better understand the present and future population trends of the species.

Download Full-text

Estimating detection probability and site occupancy of Leiolepis guentherpetersi in the coastal sandy areas of Phu Loc district, Thua Thien Hue province

ACADEMIA JOURNAL OF BIOLOGY ◽

10.15625/2615-9023/v40n4.12839 ◽

2018 ◽

Vol 40 (4) ◽

Author(s):

Cao Thi Thanh Nguyen ◽

Ngo Van Binh ◽

Ngo Dac Chung

Keyword(s):

Environmental Factors ◽

Detection Probability ◽

Site Occupancy ◽

Weather Conditions ◽

Coastal Areas ◽

Available Information

The Peter’s Butterfly Lizard (Leiolepis guentherpetersi) is endemic to Vietnam. However, there is no available information related to detection probability and site occupancy of this species so far. Nine surveys were conducted at 50 plots in the coastal areas of Phu Loc district from September to December 2017 in order to detect the presence of Leiolepis guentherpetersi. Our results showed that the detection probability of L. guentherpetersi, when combined with environmental factors, was 0.383, which was higher than the naive detection probability of 0.34. The total AIC weight of the near sea ecosystem was 85.9% while the total AIC weight of the ecosystem far from the sea was only 13.5%. The AIC weight of weather conditions was 99.4% while the total AIC weight of temperature was 62.3% and the total AIC weight of humidity was 27.2%. These results indicated that the probability of detecting L. guentherpetersi influenced by both site covariates (near the sea or far from the sea) and sample covariates (temperature, humidity, and rainfall). In there, the near sea ecosystem is the best habitat for L. guentherpetersi and rainfall is sample covariates that had the greatest influence on detection probability and site occupancy of this species.

Download Full-text

Estimating Site Occupancy and Detection Probability Parameters for Meso- And Large Mammals in a Coastal Ecosystem

Journal of Wildlife Management ◽

10.2193/0022-541x(2006)70[1625:esoadp]2.0.co;2 ◽

2006 ◽

Vol 70 (6) ◽

pp. 1625-1633 ◽

Cited By ~ 123

Author(s):

ALLAN F. O'CONNELL ◽

NEIL W. TALANCY ◽

LARISSA L. BAILEY ◽

JOHN R. SAUER ◽

ROBERT COOK ◽

...

Keyword(s):

Detection Probability ◽

Site Occupancy ◽

Coastal Ecosystem ◽

Large Mammals

Download Full-text

The impacts of camera placement on species detectability in an Indonesian terrestrial community

10.26686/wgtn.15086754 ◽

2021 ◽

Author(s):

Eric Van Dam

Keyword(s):

Body Size ◽

Data Collection ◽

Study Design ◽

Detection Probability ◽

Camera Trap ◽

Camera Traps ◽

Camera Placement ◽

Species Detection ◽

Animal Populations ◽

Selective Placement

<p>Ecologists have increasingly favoured the use of camera traps in studies of animal populations and their behaviour. Because camera trap study design commonly implements non-random selective placement, we must consider how this placement strategy affects the integrity of our data collection. Selective placement of camera traps have the benefits of 1) maximizing the probability of encounter events by sampling habitats or microhabitats of known significance to a focus or closely-related species and 2) reducing data collection and maintenance effort in the field by situating cameras along more easily-accessible landscape features. Introducing a non-random survey method, such as selective placement, into a project studying a species or community that also expresses non-random habitat use may lead to unintentionally biased data and inaccurate results. By using a paired on-trail/off-trail camera-trap study design, my aim is to investigate potential differences in popular ecological indices, species detection probability (p) using multi-method occupancy models, and intraspecific temporal activity for a terrestrial community in Gunung Palung National Park in Indonesian Borneo. Differences in detection probability between on and off-trail cameras were compared against species characteristics (including body size, diet, and taxonomic group) to find potential correlations. While several species exhibited a significant difference in detection probability between cameras placed on foot trails and those placed randomly off-trail, there was no measured community trend. This stresses my conclusion further that a non-random study design leaves results open to bias from unknown patterns in detection due to underlying variation in behaviour and microhabitat use. Selective placement may be effective for increasing detection probability for some species but can also lead to substantial bias if the features selected for are not explicitly taken into account within the analysis or balanced with a control in the study design. In addition, a positive interactive effect was found between on trail species detection and body size for the terrestrial omnivore guild, and three species presented significant variation in temporal activity between camera placement types. This provides evidence that camera placement not only affects species state parameters and indices but has a noticeable impact on behavioural observations that require accountability as well.</p>

Download Full-text

Okupansi Kukang Jawa (Nycticebus javanicus E. Geoffroy 1812) di Hutan Tropis Dataran Rendah di Kemuning, Bejen, Temanggung, Jawa Tengah

Jurnal Ilmu Kehutanan ◽

10.22146/jik.46141 ◽

2019 ◽

Vol 13 (1) ◽

pp. 15 ◽

Cited By ~ 2

Author(s):

Mahfut Sodik ◽

Satyawan Pudyatmoko ◽

Pujo Semedi Hargo Yuwono

Keyword(s):

Detection Probability ◽

Negative Impact ◽

Site Occupancy ◽

Water Source ◽

Critically Endangered ◽

Primate Species ◽

Slow Loris ◽

Occupancy Model ◽

Central Java ◽

Site Use

Faktor kehilangan/berkurangnya habitat, dan fragmentasi habitat dapat memberikan dampak buruk terhadap kukang Jawa (Nycticebus javanicus), satwa primata nokturnal yang tergolong dalam kategori Critically Endangered. Kukang Jawa yang hidup di hutan yang terfragmentasi merasakan dampak negatif dari faktor- faktor tersebut dan hal tersebut juga dapat memengaruhi okupansi dalam sebuah kawasan. Penelitian ini bertujuan untuk mengetahui faktor-faktor yang memengaruhi okupansi habitat oleh kukang Jawa di hutan dataran rendah yang terfragmentasi di Kemuning, Temanggung, Jawa Tengah, Indonesia. Untuk memperkirakan proporsi penggunaan wilayah, probabilitas deteksi (detection probability) dan faktor – faktor yang berpengaruh terhadap okupansi habitat oleh kukang Jawa, kami menggunakan occupancy model of a single-season. Sebanyak 5 kali ulangan survei malam pada tahun 2017 digunakan sebagai data pokok di dalam model okupansi. Metode pengambilan data lingkungan dan data anthropogenic menggunakan observasi lapangan dan interview dengan masyarakat lokal. Kami membagi lokasi penelitian menjadi 141 grid dengan ukuran 200 m x 200 m (4 ha) sebagai acuan dalam survei malam dengan jalur. Data kovariat lingkungan yang diukur adalah jarak dari jalan, jarak dari tepi hutan, jarak dari pemukiman, jarak dari sumber air, ketinggian tempat, dan kemiringan lahan. Hasil penelitian menunjukan bahwa kukang Jawa menghuni habitat sekitar 23,2% dari keseluruhan areal di hutan Kemuning. Jarak dari jalan dan jarak dari sumber air (sungai) berkorelasi positif terhadap tingkat hunian, sedang jarak dari pemukiman berkorelasi negatif terhadap tingkat hunian dari kukang Jawa. Data dan informasi kuantitatif yang dihasilkan dari penelitian ini penting untuk mengetahui kebutuhan sumber daya jangka panjang populasi kukang Jawa khususnya di hutan Kemuning. Selanjutnya diharapkan pemerintah Indonesia atau stakeholder terkait dapat melakukan upaya konservasi dan rencana strategi pengelolaan spesies kukang Jawa dengan baik khususnya di hutan dataran rendah yang terfragmentasi.Occupancy of Javan Slow Loris (Nyticebus javanicus E. Geoffroy 1812) in Kemuning Tropical Low Land Forest, Bejen, Temanggung, Central Java Abstract Habitat loss and landscape fragmentation have a negative impact on the Javan slow loris (Nycticebus javanicus), a Critically Endangered nocturnal primate species. Slow lorises in remaining forest fragments might be suffered and affect their occupancy behavior. We aim to investigate the determinant factors for the probability of habitat occupancy by the javan slow loris in Kemuning forest fragment of Temanggung District, Central Java. To estimate the site occupancy rate, detection probability, and the determinant factor of site use by Nycticebus javanicus, we employed the occupancy model of a single-season using night surveys. Five repeated night surveys in 2017 were used as the main basis data for the occupancy model. We used direct observation and interview with locals to collect data on environmental and anthropogenic features. We divided the study area into 141 grids with 200 m x 200 m (4 ha) each which were the basis for the night survey following existing walking paths. The influence of six covariates was assessed to determine of site use by Nycticebus javanicus: distance to road, distance to forest edge, distance to the settlement, distance to water source, altitude, and elevation. The result shows that the probability of site use occupied by Nycticebus javanicus was 23.2% of the total area. Distance to roads and distance to water source have a positive correlation with the probability of site use, whereas the influence of distance to settlements has a negative correlation with the site use of the species. Such quantitative data and information gained in this research are important to know for the long term resource needs of the Nycticebus javanicus, especially in the Kemuning forest. Therefore, the Indonesian Government or related stakeholders can formulate the detail conservation plans of the species, especially in the lowland fragmented tropical forest.

Download Full-text

Multispecies site occupancy modeling and study design for spatially replicated environmental DNA metabarcoding

10.1101/2021.02.14.431192 ◽

2021 ◽

Author(s):

Keiichi Fukaya ◽

Natsuko Ito Kondo ◽

Shin-Ichiro S. Matsuzaki ◽

Taku Kadoya

Keyword(s):

Study Design ◽

Detection Efficiency ◽

Site Occupancy ◽

Environmental Dna ◽

List Type ◽

Bayesian Decision ◽

Imperfect Detection ◽

Efficient Manner ◽

Species Detection ◽

Multi Stage

AbstractEnvironmental DNA (eDNA) metabarcoding has become widely applied to gauge biodiversity in a noninvasive and cost-efficient manner. The detection of species using eDNA metabarcoding is, however, imperfect owing to various factors that can cause false negatives in the inherent multi-stage workflow.Imperfect detection in the multi-stage workflow of eDNA metabarcoding also raises an issue of study design: namely, how available resources should be allocated among the different stages to optimize survey efficiency.Here, we propose a variant of the multispecies site occupancy model for eDNA metabar-coding studies where samples are collected at multiple sites within a region of interest. This model describes the variation in sequence reads, the unique output of the high-throughput sequencers, in terms of the hierarchical workflow of eDNA metabarcoding and interspecific heterogeneity, allowing the decomposition of the sources of variation in the detectability of species throughout the different stages of the workflow. We also introduced a Bayesian decision analysis framework to identify the study design that optimizes the efficiency of species detection with a limited budget.An application of the model to freshwater fish communities in the Lake Kasumigaura watershed, in Japan, highlighted a remarkable inhomogeneity in the detectability of species, indicating a potential risk of the biased detection of specific species. Species with lower site occupancy probabilities tended to be difficult to detect as they had lower capture probabilities and lower dominance of the sequences. The expected abundance of sequence reads was predicted to vary by up to 23.5 times between species.An analysis of the study design suggested that ensuring multiple within-site replications of the environmental samples is preferred in order to achieve better species detection efficiency, provided that a throughput of tens of thousands of sequence reads was secured.The proposed framework makes the application of eDNA metabarcoding more error-tolerant, allowing ecologists to monitor ecological communities more efficiently.

Download Full-text

Variable detectability and El-Niño associations with riparian snakes in Sabah, Malaysian Borneo

Journal of Tropical Ecology ◽

10.1017/s0266467421000468 ◽

2021 ◽

pp. 1-6

Author(s):

Sami Asad ◽

Shi Teng Ng ◽

Julsun Sikui ◽

Mark-Oliver Rödel

Keyword(s):

El Niño ◽

Detection Probability ◽

El Nino ◽

Occupancy Modelling ◽

Species Detection ◽

Snake Species ◽

Factors Associated ◽

Modelling Framework ◽

Malaysian Borneo

Abstract Although snake populations are suffering numerous local declines, determining the scale of these declines is problematic due to the elusive nature of snakes. Determining the factors associated with species detection is therefore essential for quantifying disturbance effects on populations. From 2017 to 2019, we assessed the detectability associations of five river-associated snake species and all snake detections in general within two logging concessions in Sabah, Malaysian Borneo. Data collected from both stream transects and visual encounter surveys at 47 stream sites were incorporated into an occupancy-modelling framework to determine the climatological, temporal and survey distance associations with species detection probability. Detection probability of riparian snake species was significantly associated with humidity, month (2 spp. each), survey distance and total rainfall over 60 days (1 spp. each). Pooled snake species detectability was significantly positively associated with transect distance and the 2019 El-Niño year, whilst yearly pooled snake species detections in stream transects spiked during El-Niño (2017 = 2.05, 2018 = 2.47, 2019 = 4.5 snakes per km). This study provides new insights into the detectability of riparian rainforest snakes and suggests that future studies should account for short-term (climatological and temporal) and long-term (El-Niño) factors associated with detection probability when surveying and assessing snake populations.

Download Full-text

Estimating site occupancy and detectability of an endangered New Zealand lizard, the Otago skink (Oligosoma otagense)

Wildlife Research ◽

10.1071/wr05045 ◽

2006 ◽

Vol 33 (3) ◽

pp. 193 ◽

Cited By ~ 12

Author(s):

Catherine M. Roughton ◽

Philip J. Seddon

Keyword(s):

New Zealand ◽

Detection Probability ◽

Site Occupancy ◽

Implicit Assumption ◽

Monitoring Methods ◽

Analytical Technique ◽

Robust Estimates ◽

Simple Index ◽

Management Interventions ◽

Reptile Species

There is a need to develop appropriate monitoring methods for cryptic reptile species, such as the endangered New Zealand endemic Otago skink (Oligosoma otagense), that take into account incomplete detectability. A recently developed analytical technique was applied to presence–absence data for Otago skinks to derive robust estimates of both detection probability and site occupancy. The estimated detection probability for Otago skinks was 0.4 (s.e. 0.052), and the estimate of the proportion of the study area occupied by Otago skinks was 0.71 (s.e. 0.11). The data derived from presence–absence surveys in consistently sunny weather, with a standardised scanning and searching technique, provided a good basis for robust estimates of detection probability and of the percentage of area occupied by skinks. This survey technique could be applied to other similarly cryptic reptile species to derive rigorous estimates of site occupancy in order to track changes over time or in response to management interventions. It will be particularly appropriate where precise estimation of absolute abundance is not warranted, or where a simple index of relative abundance may be invalid due to failure of the implicit assumption of constant detectability.

Download Full-text

Understanding distribution and occupancy of Himalayan Monal in Uttarkashi district, Uttarakhand

10.1101/2021.02.16.431367 ◽

2021 ◽

Author(s):

Amira Sharief ◽

Hemant Singh ◽

Bheem Dutt Joshi ◽

Tanoy Mukherjee ◽

Kailash Chandra ◽

...

Keyword(s):

Detection Probability ◽

Site Occupancy ◽

Null Model ◽

Average Density ◽

Camera Traps ◽

Occupancy Probability ◽

Sign Surveys ◽

Reserve Forest ◽

Anthropogenic Threats ◽

Conservation And Management

AbstractThe Himalayan Monal is a conservation priority species in its entire distribution range. Its population is declining in many areas due to various anthropogenic threats. The information on species distribution and its abundance is lacking in many areas which are vital for conservation and management planning. Hence, through the present study, we have assessed the abundance and occupancy of Himalayan monal in Uttarkashi district (Uttarakhand). We used camera traps and conventional sign surveys for documenting the species during 2018-2019. We installed a total of 69 camera traps (2819 trap nights) and surveyed 54 trails (650 km) which represents entire habitat and topographic variability of the landscape. The occupancy and detection probability was modelled using the habitat variables. The top model showed that occupancy probability of Himalayan monal was positively influenced by the slope (β =27.52 ±16.25) and negatively influenced by Reserve Forest (RF) (β= −8.14 SE ± 4.99). The observed naïve occupancy of Himalayan Monal was 0.69 in the study area, which was slightly lower than the estimated occupancy. However, in the null model, the site occupancy estimated was found to be 0.82±0.08 and with detection probability 0.23±0.03. The overall abundance of monal was estimated about 171.58 ±10.2 in the study area with an average density of 0.62/ km2. The activity pattern analysis indicates that monal remains very active between 6.00 hrs −12.00 hrs and relatively less active during mid-day when humans are most active 11.30 hrs-16.30 hrs. The present study is a first attempt to estimate occupancy and abundance using camera traps as well as sign survey for the species primarily from non-Protected Area (PA). We found that Himalayan monal is abundant outside the PAs, which is a good indication for its long-term viability and also identified areas for conservation and management prioritization in Uttarkashi.

Download Full-text