background-image: url(libs/slide_templates/GMRI_template_slide1.png) background-size: cover class: center middle inverse # 2021 WARMEM Size Spectra Update #### Adam A. Kemberling ##### Gulf of Maine Research Institute updated: 2021-12-07 <style type="text/css"> .remark-slide-content { font-size: 24px; padding: 1em 4em 1em 4em; } </style> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover # WARMEM SizeSpectrum Analysis: 2021 ## **About:** This slide deck covers the major findings of the size-spectrum analyses performed on NMFS trawl data. ## **Goals:** Present relevant background and current progress for feedback. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## **Background:** Ecosystem Impacts of Warming **From the WARMEM Proposal:** > #### Warming is the most robust change projected by global climate models (Deser et al. 2010), and most marine ecosystems, including the northwest Atlantic Shelf are projected to experience temperatures that regularly exceed past variability by the middle of the century (Henson et al. 2017, Alexander et al. 2018). --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## **Background:** Warming in the NW Atlantic <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-1-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Physiological Impacts of Warming: > Temperature is the most important physical factor affecting processes at multiple scales of organization from cellular chemical reaction rates, organismal physiological processes, population productivity and distribution, and interactions among species in complex communities (Fry 1971, Magnuson et al. 1979, Claireaux and Lefrançois 2007). Temperature related effects on marine organism physiology impacts: 1. The acquisition of biomass through feeding 2. The loss of biomass through metabolism 3. The rate of development --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Physiological Patterns In most species differences in the temperature-dependence of these rates causes **<u>young animals to grow more quickly</u>** in warmer water. However, development rates also increases, **<u>causing individuals in warmer conditions to reach maturity earlier and at a smaller size.</u>** --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Impact of Warming on Size Spectrum: In the context of a strongly size-structured ecosystem, growth and maturity changes alter fitness and ultimately determine whether a species is successful in the given environment. **In the case of the NW Atlantic:** sustained increases in temperature should have a physiological impact on the community. This leads to our second hypothesis: > #### H2. Warming alters the community through the direct influence of temperature on metabolism, growth, and population productivity. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Research Objectives: For each region, we will develop several time series indicators: >a) Community composition metrics, >b) Mean size of the aggregate community and key functional groups, >c) **Slope and intercept of the size spectrum.** These indicators will allow us to explore the three hypotheses and will support the size-spectrum and dynamic food-web modeling. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## What is a Size Spectra A **"Size spectra"** describes the relationship between organism size (length or bodymass) and abundance. ![Petchey Size Spectrum Schematic](libs/spectra_theory_images/spectra_theory_petchey.png) --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Features of a Community Size Spectrum Size spectra are characterized with two parameters **Slope & Intercept** > <h4> **<u>Slope:</u>** </h4> Size spectrum slope capture the rate at which abundance declines with increasing body size. This relationship captures the efficiency of energy transfer from small to large individuals, and emerges from the scaling of individual-level processes (Guiet et al. 2016b). > Variation in the slope has been associated with **fishing pressure** (Bianchi et al. 2000, Shin et al. 2005) and **environmental conditions** (Guiet et al. 2016a). --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Size Spectrum as an Ecological Indicator **Jennings & Blanchard, 2004** > The Slopes of size spectra become increasingly negative following fishery exploitation (Duplisea & Kerr 1995; Rice & Gislason 1996) > This is due to (1) the differential vulnerability of larger species, (2) within-population changes in mean body size and life history, (3) genetic changes in life history, (4) predator–prey relationships or (5) changes in competitive interactions (Gislason & Rice 1998; Jennings, Greenstreet & Reynolds 1999; Bianchi et al. 2000; Law 2000). --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Research Expectations: We expect that changes in growth rates, especially in immature fish, will be the first response in the community to increases in temperature... > ### After several years of warm conditions, the size of adults will decline. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Constructing NMFS Trawl Size Spectrum The Steps to construct size spectra from the NMFS Trawl Data: 1. Perform area-stratification to estimate stratified abundances at length 2. Use published weight-at-length relationships to get individual weights for different lengths caught, stratified biomass at length 3. Using stratified abundances, estimate size spectra for the study regions (GoM, GB, SNE, MAB) * Assign abundances to bodymass bins on log10 scale * Total the stratified abundances for each bin * **Normalize** abundances by dividing by the bin-width * Fit size spectra --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Survey Data Prep Data on fish abundance, size, and age come from the bottom trawl surveys conducted by the Northeast Fisheries Science Center (NEFSC) in the autumn (since 1963) and spring (since 1968). There are two main datasets the "catch" data and the "biological" data:
--- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Step 1: Stratified Abundances Stratified abundances at length were calculated for each year, season, and survey strata sampled. These area-stratified abundances were used to account for differences in sampling effort and the relative size of each stratum. For details on the area-stratification and Length-Weight QA/QC please see [Area stratification validation markdown](https://adamkemberling.github.io/nefsc_trawl/R/qaqc_reports/stratification_validation.html) --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Step 2: Bodymasses from Length **After QA/QC 68 species were used:** *_acadian redfish_, _american plaice_, _american shad_, _atlantic angel shark_, _atlantic cod_, _atlantic croaker_, _atlantic halibut_, _atlantic herring_, _atlantic mackerel_, _atlantic sharpnose shark_, _atlantic spadefish_, _atlantic sturgeon_, _atlantic thread herring_, _atlantic wolffish_, _barndoor skate_, _black sea bass_, _blackbelly rosefish_, _blueback herring_, _bluefish_, _buckler dory_, _bullnose ray_, _butterfish_, _chain dogfish_, _clearnose skate_, _cownose ray_, _cunner_, _cusk_, _fawn cusk-eel_, _fourspot flounder_, _goosefish_, _greater amberjack_, _haddock_, _little skate_, _longhorn sculpin_, _northern kingfish_, _northern searobin_, _ocean pout_, _offshore hake_, _pollock_, _red hake_, _rosette skate_, _roughtail stingray_, _round herring_, _sand tiger_, _sandbar shark_, _scup_, _sea raven_, _silver hake_, _smooth butterfly ray_, _smooth dogfish_, _smooth skate_, _southern kingfish_, _spanish mackerel_, _spanish sardine_, _spiny butterfly ray_, _spiny dogfish_, _spot_, _spotted hake_, _striped bass_, _summer flounder_, _thorny skate_, _weakfish_, _white hake_, _windowpane flounder_, _winter flounder_, _winter skate_, _witch flounder_ and _yellowtail flounder_* --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Step 3: Build Size Spectra Using data from the Gulf of Maine as an example, here is the distribution of abundances before normalizing the size spectra: <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-4-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Normalize Size Spectra Before fitting the size spectrum slope the last step is to normalize each bin by the bin width to get the *"normalized biomass size spectra"* <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-5-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Fitting Size Spectra Once the data is normalized it is then fit with a linear model to get the size spectrum slope and intercept. <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-6-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Regional Progress Update: Size Spectrum Results for **Slope, intercept, and exponent of size spectra (alternative method)** have been estimated for <u>all regions and for each season</u>. These are stored in a results table and are staged for testing relationships to external drivers. Here is some background on the differences among the different regions. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Distribution of Biomass By Region and Species Type <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-7-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Gulf of Maine: Deep Dive <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-8-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Gulf of Maine Abundance Trends <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-10-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Gulf of Maine Biomass Trends <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-11-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Key Species Trends Some of the largest drivers within of Gulf of Maine come from a handful of species. For these it is helpful to look how their individual populations have changed. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Key Species (Atlantic Cod) <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-14-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Key Species (Haddock) <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-15-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Key Species (Spiny Dogfish) <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-16-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Key Species (Acadian Redfish) <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-17-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## GOM: Size Spectra Results <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-19-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## GOM: Size Spectra Intercepts <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-20-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## GOM: Exponent of Size Spectra (MLE Method) <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-21-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## How does that relate to patterns in biomass ex. area #### **Broad Gulf of Maine Patterns** > Increasing temperatures, particularly warm since ~2008 > Increase in **<u>biomass & abundance</u>** over time. Decreases in average size and weight of the individuals, across all groups of species. > **<u>Decreasing</u>** size spectrum slope (from both methods) --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Size Spectrum Slopes: All Areas <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-22-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Average Size Changes: All Areas The following two slides use the stratified catch data, prior to the removal of individuals smaller than 1g. The changes in size are across all areas in the survey and are done individually by survey season. --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Changes in Average Seasonal Weight |Change in Avg. Seasonal Weight | n| |:------------------------------|--:| |Decreasing | 52| |Increasing | 16| |No Change | 57| |NA | 2| <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-25-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ### Changes in Average Seasonal Length |Change in Avg. Seasonal Length | n| |:------------------------------|--:| |Decreasing | 44| |Increasing | 17| |No Change | 64| |NA | 2| <img src="Northeast_Trawl_Size_Spectrum_files/figure-html/unnamed-chunk-27-1.png" style="display: block; margin: auto;" /> --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Things we are still working on: 1. Size and growth characteristics of specific species 2. Relationships with external drivers --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Analyses to think of? > How much attention should we give to the internal structures of the size spectra? > How important is the community composition to the inference of size spectra as ecosystem indicators? > Is this consistent with the community indicators? > Is the mechanism just temperature as a departure from normal, or is it an accumulation or successive temperature impact? --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Potential Drivers? > Temperature Data: (surface temperature, anomalies, lagged, deltas) > Zooplankton Community (CPR) > Fishing effort? > Fishing landings > Community metrics --- background-image: url(libs/slide_templates/GMRI_template_slide_white.png) background-size: cover ## Bonus: Full Research Pipeline