A hierarchical approach to defining marine heatwaves

In the paper

“A variety of impacts 4 have been associated with these anomalous events, including shifts in species ranges, local 5 extinctions and economic impacts on seafood industries through declines in important fishery 6 species and impacts on aquaculture.”

What the sources say

• ?Not checked — no full text of the sourcecites Garrabou et al. 2009 source
• ?Not checked — no full text of the sourcecites Wernberg et al. 2013 source
• Partly supported

  The “Ecological and Economic Impacts” section describes species range shifts, lobster mortality and substantial economic effects on fisheries. It does not document local extinctions or aquaculture impacts, and the economic harm it describes is not specifically attributed to declines in fishery species; lobster prices fell amid an influx of catch. No page number is available in our copy. (LLM, conf=0.86)

  cites Mills et al. 2013 sourceExperimental — accuracy not yet measured

In the paper

“The high temporal resolution provided by the in situ logger data reveals higher temporal SST variability, but a lower peak intensity than the daily satellite dataset, consistent with previous analysis of sub-surface in situ and daily satellite data in this region (Smale and Wernberg 2009).”

What the sources say

• Partly supported

  The ‘Satellite-derived vs. logger-derived data’ section (page not available) says in situ recorders capture short-term temperature fluctuations that satellite data cannot resolve. It also reports that daily AVHRR temperatures were generally higher than logger temperatures, but gives this comparison for annual and seasonal means, not peak intensity. The claimed lower logger peak intensity is therefore not established by the cited analysis. (LLM, conf=0.82)

  cites Smale and Wernberg 2009 sourceExperimental — accuracy not yet measured

In the paper

“The general definition of atmospheric heatwaves is a prolonged period where temperatures are substantially hotter than normal (Perkins and Alexander 2013).”

What the sources say

• Supported

  The Introduction defines heat waves broadly as consecutive days when conditions are excessively hotter than normal. This supports the citing sentence’s paraphrase of a prolonged period of unusually high temperatures; no page number is supplied in the passage. (LLM, conf=0.94)

  cites Perkins and Alexander 2013 source

In the paper

“From examples such as the 2003 MHW in the northern Mediterranean Sea (Garrabou et al. 2009), the 2011 Ningaloo Niño in Western Australia (Feng et al. 2013) and the 2012 MHW in the northwest Atlantic (Mills et al. 2013; Chen et al. 2014), it is clear that the atmosphere, land surface and ocean all may have important driving roles in how and where these prolonged heat events play out.”

What the sources say

• ?Not checked — no full text of the sourcecites Garrabou et al. 2009 source
• Supported

  The Discussion (page unavailable) identifies the 2011 marine heat wave off Western Australia as a Ningaloo Niño and describes oceanic and atmospheric processes driving it. Feng et al. is cited as an example in a sentence drawing a broader conclusion from several events; that citation does not specifically attribute a land-surface driver to the Australian event. (LLM, conf=0.91)

  cites Feng et al. 2013 source
• Partly supported

  The Northwest Atlantic temperature discussion says atmospheric blocking brought warm air into the region and contributed to the 2012 ocean heat wave; it also refers to atmosphere–ocean interactions. It does not identify the land surface as a driver of that event: Greenland melting is described as another likely consequence of the atmospheric conditions. No page number is available in our copy. (LLM, conf=0.84)

  cites Mills et al. 2013 sourceExperimental — accuracy not yet measured

In the paper

“Based on these criteria, Perkins and Alexander (2013) define an atmospheric 138 heatwave when at least three consecutive days exceed a calendar day threshold defined as the 139 90 th percentile value for temperature.”

What the sources say

• Supported

  The Conclusions define CTX90pct as three or more consecutive days exceeding the calendar-day 90th percentile, and CTN90pct applies the corresponding threshold to nighttime temperatures. The heat-wave indices section specifies that these thresholds are calculated from maximum and minimum temperature, respectively. (LLM, conf=0.93)

  cites Perkins and Alexander (2013) source

In the paper

“protecting high diversity coral reef areas with a high probability of catastrophic disturbances in the near future, including MHWs, may be less favourable in comparison with protecting an area with less biodiversity but a low probability for disturbance (Game et al. 2008).”

What the sources say

• Partly supported

  The Discussion (page not available) says that, all else equal, protecting reefs at lower risk of catastrophic bleaching is preferable, and that improving persistence can require trade-offs with biodiversity coverage. It does not establish the specific comparison between a high-diversity, high-risk reef and a less-diverse, low-risk reef. The study models bleaching risk through 2100 but does not explicitly assess marine heatwaves as a broader category of disturbance. (LLM, conf=0.84)

  cites Game et al. 2008 sourceExperimental — accuracy not yet measured

In the paper

“A recent atmospherically driven marine heatwave in the northeast Pacific during the boreal winters of 2013-2015 had significant downstream effects on North American weather, and also disrupted northeast Pacific fisheries and coastal ecosystems (Bond et al. 2015; Hartman, 2015; Whitney 2015).”

What the sources say

• Partly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites Bond et al. 2015 sourceExperimental — accuracy not yet measured
• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Hartman, 2015
• Partly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites Whitney 2015 sourceExperimental — accuracy not yet measured

In the paper

“but a lower maximum intensity (imax=4.24°C) than a 56 day event in 2012 (10 April - 4 June) where the imax was 4.89°C”

In the paper

“The most intense event was in 2008 (imax = 5.05) lasting only 9 days (26 June to 4 July) with an imean of 3.87°C.”

In the paper

“The longest event was still ongoing at the time of analysis, with 110 days (13 September - 31 December 2014 - the end of the dataset), with a lower imax (3.31°C) and imean (2.51°C), but the highest icum (276°C days).”

In the paper

“Note that the 56 and 187 day events in 2012 are considered distinct, as the temperature dropped below the threshold for at least 3 days (5 days) between the two events.”

In the paper

“However, the variability in SST magnitude clearly differs between the four datasets, and would result in different metrics of heatwave intensity.”

In the paper

“with the longest MHW lasting for 95 days (13 May 1999 - 15 August 1999) with a maximum (imax), mean (imean) and cumulative intensity (icum) of 3.60°C, 2.50 °C and 237°C days above the climatological mean, respectively.”

In the paper

“The corresponding imean and icum for this 56 day event was 2.59°C and 145 °C days, respectively.”

In the paper

“The 2003 MHW was the largest event based on the imean (4.06°C; not shown in Figure 3b) and lasted 30 days by our definition (2 June - 1 July) with an imax of 5.02°C.”

In the paper

“This is followed by the daily satellite data, which are finer in spatial (0.25° degree grid) and temporal (daily) resolution.”

In the paper

“Sensitivity testing on daily data suggests that a 30-day 'moving window' is appropriate for smoothing climatology from daily data.”

In the paper

“The WA and northwest Atlantic MHWs are the largest by maximum intensity, such that the red and yellow bars are the same.”

In the paper

“In situ temperatures (e.g. loggers) | 4 | 4 | 4 | 4 | 1 (if multiple loggers), else N/A | 2 | High | High | Low | Low/ Med”

In the paper

“The northwest Atlantic event is not the largest according to duration or cumulative intensity, but the red bar obscures the yellow bar since they are so close in time.”

In the paper

“A clear 8 definition will facilitate retrospective comparisons between MHWs, enabling the synthesis 9 and a mechanistic understanding of the role of MHWs in marine ecosystems.”

In the paper

“Marine heatwaves (MHWs) have been observed around the world and are expected to 3 increase in intensity and frequency under anthropogenic climate change.”

• Also stated in: § 1. Introduction - Marine heatwaves and their ecological impact · p. 4 · hedged

  “which may increase in frequency and magnitude as a result of anthropogenic climate change (IPCC 2012)”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites IPCC 2012

In the paper

“extreme temperature events can affect species distributions and alter ecosystem structure”

In the paper

“Furthermore, monitoring heatwaves can lead to a better understanding of their development, characteristics and impacts.”

In the paper

“Under this definition, a MHW can be caused by a combination of atmospheric forcing (e.g. heating) and oceanic conditions (e.g. faster advection or advection of unusually warm water).”

In the paper

“Many of the MHW metrics can be calculated from gridded products, such as SST datasets, reconstructed observational data, and model/reanalysis data.”

In the paper

“The qualitative definition of a MHW applies to ocean regions (including subsurface waters, estuarine, or enclosed seas such as the Mediterranean Sea or Baltic Sea)”

In the paper

“These metrics can be useful at different time scales in forecasting for the following days to weeks and for long-term projections.”

In the paper

“Australia's Bureau of Meteorology OceanMAPS system predicts daily SSTs with a one-week lead time (www.csiro.au/bluelink/)”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites www.csiro.au/bluelink/

In the paper

“deployment of submerged data loggers close to the coastline and the use of oceanographic arrays for the open sea could provide information about heat penetration depths and durations”

In the paper

“This is generally due to the resolution of the data, but can also relate to other issues of quality, consistency and instrumentation.”

In the paper

“Lower spatial resolution data results in reduced intensity because neighbouring non-heatwave areas are included in mean values.”

In the paper

“For example, Coral Reef Watch is based on near real-time monitoring during the warmest months of the year and is used to identify areas where conditions may be approaching those conducive to coral inform management actions (to reduce additional stressors for example) which can be implemented quickly (e.g. Beeden et al. 2012).”

What the sources say

• ?Not checked — no full text of the sourcecites Beeden et al. 2012 source

In the paper

“This also means that a MHW is not just limited to the warmer months, since for some biological applications the consideration of heatwaves in colder months is essential.”

In the paper

“Besides near real-time monitoring, the metrics can be used to estimate the prevalence of future MHWs.”

In the paper

“metrics discussed here and applied to daily SST data would allow warnings to be issued when bleaching (http://coralreefwatch.noaa.gov/satellite/index.php).”

In the paper

“For other research applications, such as studies of large-scale climate variability, MHW metrics may require further modification based on the resolution of datasets being used.”

In the paper

“Not only can different datasets generate different values for the same metric, certain indices may simply not be appropriate or derivable from some data sources.”

In the paper

“However, resolution may be reduced due to the coarseness of spatial grids and, in the case of regional models, the domain they cover.”

In the paper

“Therefore, we recommend five days as a balance to achieve relatively uniform global MHW counts under current climatic conditions.”

In the paper

“their POAMA model predicts monthly SSTs for the upcoming nine months (www.bom.gov.au/climate/poama2.4/poama.shtml)”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites www.bom.gov.au/climate/poama2.4/poama.shtml

In the paper

“Primary metrics allow for the most general comparison between duration and magnitude (intensity).”

In the paper

“This set of metrics allows different MHW events to be uniquely described and compared ( Figure 2 ).”

In the paper

“A hierarchy is useful as different temperature datasets, based on their spatial and temporal resolution, have different abilities to provide different metrics.”

In the paper

“Moreover, by using a percentile rather than standard deviation definition no assumption is made regarding the underlying distribution of anomalies.”

In the paper

“Near real-time monitoring of ocean surface temperatures based on satellite data is possible”

In the paper

“these metrics will likely be specific to particular habitats, regions and species via potential local adaptation to extremes (Palumbi et al. 2014)”

What the sources say

• ?Not checked — no full text of the sourcecites Palumbi et al. 2014 source

In the paper

“A percentile threshold is recommended rather than an absolute value above the climatological value as the magnitude of variability across a range of timescales varies considerably by region.”

In the paper

“Using them within a forecasting framework would lead to near-term prediction of MHWs.”

In the paper

“While coarser resolution datasets may provide information about larger areas and/or longer time periods, this information may not be particularly relevant for marine managers or policy makers who require accurate local scale information, particularly on magnitude, to assess likely impacts.”

In the paper

“An absolute threshold (e.g. 2°C anomaly) would only be relevant in terms of impacts in some regions but not in others (e.g. due to species acclimation).”

In the paper

“A lower intensity MHW (panel d ) is distinguished by its maximum intensity (imax), while a short MHW (panel e ) is distinguished by its duration ( D ).”

In the paper

“The three-tier framework allows for an over-arching and consistent measurement of heatwaves, while also providing flexibility in specifying additional metrics, if necessary.”

In the paper

“Including MHW metrics in the forecasting based on daily predictions would help to identify areas where MHWs may occur and actions could be implemented weeks ahead of time”

In the paper

“It will also facilitate the comparison of observed events against those simulated in model projections, which will be very useful in understanding plausible future changes in MHWs.”

In the paper

“Projections beyond the near-term could identify future MHW risk areas.”

In the paper

“This will facilitate comparisons between different MHW events, across seasons and at regional scales.”

In the paper

“While 10% of days will be above this threshold”

In the paper

“The likelihood of an area experiencing extreme climatic events could thus be used to decide which areas should be protected and which are less resilient and prone to strong impacts with low expectations of recovery.”

In the paper

“Marine heatwaves are an emerging area of interdisciplinary research with potential for collaborative initiatives in understanding these phenomena.”

In the paper

“This event, along with the 2003 Mediterranean Sea, 2011 Western Australia and 2012 Northwest Atlantic MHW, provide an opportunity to investigate the drivers and anomalous properties of MHWs under a hierarchical framework.”

In the paper

“Such an approach can be used to identify regions that may be particularly susceptible to MHWs (i.e. areas subjected to intense human impacts) or regions where ecological impacts may be particularly severe (i.e. hotspots of biodiversity).”

In the paper

“The adverse impacts of these events span a vast range of marine ecosystems.”

In the paper

“Daily climatological threshold time series (e.g. 90 th percentile) may need to be smoothed in order to extract a useful climatology from inherently variable data.”

In the paper

“This asymmetric MHW (b) is distinguished from one with a slow onset and rapid decline (panel c ) by the rate of warming (ronset) metric.”

In the paper

“Assessments of spatial and temporal variability in the occurrence of MHWs can be combined with analyses of other important aspects of the marine environment”

In the paper

“The use of standardised software would ensure consistency in calculating metrics”

In the paper

“Factors such as time of year relative to the onset of the MHW, or periods of above mean temperature preceding the MHW may lead to greater impacts.”

In the paper

“A lower intensity MHW (panel d ) is distinguished by its maximum intensity (imax), while a short MHW (panel e ) is distinguished by its duration ( D ).”

In the paper

“A marine heat wave (MHW) with regular warming onset and decline (panel a ) can be distinguished from one with similar duration and maximum intensity but asymmetric warming (panel b ) by the cumulative intensity metric (icum).”