Jason How

Jason How

Auto-generated profile, based on publicly available information

Jason How

Western Australian Fisheries and Research Laboratories, Department of Primary Industries and Regional Development, Government of Western Australia, P.O. Box 20, North Beach, WA 6920, Australia

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 May 2021 in Remote Sensing
Verified
Journal Article
Open access
PDF file
RRRyan R. Reisinger
Ryan R. Reisinger
AFAri S. Friedlaender
Ari S. Friedlaender
AZAlexandre N. Zerbini
Alexandre N. Zerbini
DPDaniel M. Palacios
Daniel M. Palacios
VAVirginia Andrews-Goff
Virginia Andrews-Goff
RCRochelle Constantine
Rochelle Constantine
Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar prediction of humpback whale habitat selection. We also fitted five regional models, the predictions of which we used as input features for four ensemble approaches: an unweighted ensemble, an ensemble weighted by environmental similarity in each cell, stacked generalization, and a hybrid approach wherein the environmental covariates and regional predictions were used as input features in a new model. We tested the predictive performance of these approaches on an independent validation dataset of humpback whale sightings and whaling catches. These multiregional ensemble approaches resulted in models with higher predictive performance than the circumpolar naive model. These approaches can be used to incorporate regional variation in animal habitat selection when fitting range-wide predictive models using machine learning algorithms. This can yield more accurate predictions across regions or populations of animals that may show variation in habitat selection.

7 January 2021 in Marine Mammal Science
Verified
Journal Article
Closed access
PDF file
JHJason R. How
Jason R. How
WMWilliam K. de la Mare
William K. de la Mare
DCDouglas K. Coughran
Douglas K. Coughran
MDMichael C. Double
Michael C. Double
SLSimon de Lestang
Simon de Lestang

1 August 2018 in Marine Policy
Verified
Journal Article
Closed access
NCNick Caputi
Nick Caputi
SLSimon de Lestang
Simon de Lestang
JHJason How
Jason How
FTFabian Trinnie
Fabian Trinnie
WFWarrick (Rick) Fletcher
Warrick (Rick) Fletcher

5 July 2022 in Ecology and Evolution
Verified
Journal Article
Open access
RHRichard A. How
Richard A. How
MCMark Cowan
Mark Cowan
JHJason R. How
Jason R. How
Determine seasonal, annual, and decadal patterns of abundance in reptile species and assemblages occupying central Bold Park (~338 ha), an isolated urban bushland remnant in Perth, Southwestern Australia. Fenced pitfall trapping in four sampling sites, representing different habitats and fire history, over the primary reptile activity period for 35 consecutive years with over 17,000 individuals captured during 3300 days of sampling; the trapping regime was modified for the last 28 years. Sampling occurred in one of 35 global biodiversity hotspots that has a Mediterranean climate experiencing a 15% decline from the century average rainfall over the last 50 years. Twenty‐nine species were recorded, with 16 captured in 32 or more years and accounting for nearly 97% of all captures; the six most common for 81%. Three taxa became locally extinct. Activity predominates in warmer and dryer months (October to April), peaking in November–December. Species richness remained relatively constant between years with around 73% of known taxa captured annually. Assemblages did not change when analyzing the presence/absence data but moved through five statistically significant assemblages analyzing relative abundance data. Over the last 28 years, relative abundance was significantly and positively correlated with annual rainfall residuals, uniquely for the 4 years preceding annual sampling, resulting in significant changes in total assemblages and significantly similar patterns in four sample sites; the presence/absence data indicated only minor assemblage changes across sites. The number of species recorded annually remained relatively constant, but relative abundance illustrated significant temporal changes in assemblages over decades. The modeled relationship between relative abundance and annual rainfall residuals for 4 years preceding annual sampling is supported by known ecological responses and reptile demographics within this Mediterranean climate. Maintenance of urban biodiversity should consider impacts of a significantly drying climate exacerbating the extinction debt already inherent in isolated bushland populations experiencing limited immigration.

20 October 2022 in Biology
Verified
Journal Article
Open access
PDF file
STStephen Mark Taylor
Stephen Mark Taylor
JHJason How
Jason How
MTMichael J. Travers
Michael J. Travers
SNStephen J Newman
Stephen J Newman
SMSilas Mountford
Silas Mountford
DGDaniel J. Gaughan
Daniel J. Gaughan
The management of human-shark interactions can benefit from the implementation of effective shark hazard mitigation measures. A Shark-Management-Alert-in-Real-Time (SMART) drumline trial in the Capes region of Western Australia was instigated after several serious incidents involving surfers and white sharks (Carcharodon carcharias). The project aimed to determine whether white sharks (target species), which were relocated after capture, remained offshore using satellite and acoustic tagging. Over a 27-month period, 352 fish were caught, 55% of which comprised tiger sharks (Galeocerdo cuvier). Ninety-one percent of animals were released alive in good condition. Only two white sharks were caught; both were relocated ≥ 1 km offshore before release and moved immediately further offshore after capture, remaining predominately in offshore waters for the duration of their 54-day and 186-day tag deployments. Our results confirm that desirable animal welfare outcomes can be achieved using SMART drumlines when response times are minimised. The low target catches and the detection of 24 other tagged white sharks within the study area supported the decision to cease the trial. Our results reiterate there is no simple remedy for dealing with the complexities of shark hazards and reinforce the importance of trialing mitigation measures under local conditions.

1 July 2018 in Bulletin of Marine Science
Verified
Journal Article
Closed access
NCNick Caputi
Nick Caputi
MFMing Feng
Ming Feng
ADAinslie Denham
Ainslie Denham
SLSimon de Lestang
Simon de Lestang
JPJames Penn
James Penn
JHJason How
Jason How
Western rock lobster larvae have a 9–11 mo pelagic larval phase that makes them susceptible to the oceanographic environment, such as the poleward-flowing Leeuwin Current, surface winds, and wave-induced Stokes drift. An innovative oceanographic-larval modelling approach, which incorporated these physical drivers and complex larval behavior, was used to investigate the cause of the major reduction in post-larval (puerulus) settlement in the late 2000s. Individual-based biophysical larval models, taking into account the sensitivity to different wind effects on the surface water movement and different levels of puerulus swimming, were developed utilizing 19 yrs of meteorological and oceanographic data (1994–2012). Model optimization was achieved by assessing modelled and observed puerulus settlements along the Western Australian coast for 24 model simulations with different combinations of model parameters. A postmodel analysis then assessed the effect of the timing of larval release, abundance and spatial distribution of larval release, and duration of larval development. Model results revealed that a mismatch between the seasonal timing of spawning, larval development, and onshore current systems needed to return the larvae to the coast appears to be the primary cause of the low settlement in the late 2000s. Since the mid-2000s, there has been an earlier start of the spawning season due to warmer winter temperatures, which may have caused the ocean currents and eddies to diffuse the larvae in the open ocean before they can be transported back to the coast by onshore currents and weakened winter storms.