meta data for this page
  •  

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
applications:applications [2022/03/02 22:41] – joetulenkoapplications:applications [2023/02/16 05:11] (current) – gregbalco
Line 2: Line 2:
  
 Here are some examples we have come up with so far. Please email any of us on the project [[balcs@bgc.org|Greg Balco]], [[benjamin.laabs@ndsu.edu|Ben Laabs]], and/or [[jtulenko@bgc.org|Joe Tulenko]] with your ideas so we can add them to the list! Here are some examples we have come up with so far. Please email any of us on the project [[balcs@bgc.org|Greg Balco]], [[benjamin.laabs@ndsu.edu|Ben Laabs]], and/or [[jtulenko@bgc.org|Joe Tulenko]] with your ideas so we can add them to the list!
- 
-**1) Analysis layer examples: the ICE-D X OCTOPUS web application** 
-The [[http://octopus.x.ice-d.org|ICE-D X OCTOPUS]] web application is not very much like the rest of the ICE-D focus area applications because it doesn't rely on an ICE-D-maintained back end database. Instead, its data layer is the [[https://earth.uow.edu.au|OCTOPUS database of cosmogenic-nuclide data used for erosion rate estimates]]. Also, the web application accesses those data by interacting with a Geoserver web feature server rather than a MySQL database. However, it does use ICE-D middle layer applications (the web service implementation of the online erosion rate calculator), so it's a good example of an analysis-layer application that uses both data-layer and middle-layer services. Also, it highlights the idea that you can mix and match data- and middle-layer servives from various places to do what you need to do. Finally, it's a pretty simple web application with a fairly minimal amount of code.  
- 
-The source code for the ICE-D X OCTOPUS web app can be viewed [[https://github.com/balcs/ice-d-x-octopus|here]]. It is written in Python 2.7, uses the [[https://webapp2.readthedocs.io/en/latest/|webapp2]] application framework, and runs on Google App Engine. Unfortuately this is kind of obsolete because GAE has been migrating to Python 3 and different web app frameworks (so I am not sure if you could install and run it on a newly created GAE project), but it is a nice simple example of how a transparent-middle-layer application can work.  
  
 ---- ----
  
-**2) Data-model comparison between LGM and penultimate moraine ages, and model output from simulations over multiple glaciations** (ie the ice sheet influence on regional climate example). **Hypothesis:** if the impact from ice sheets on re-arranging large-scale atmospheric circulation and thus modulating regional climate is actually significant, we should be able to use climate model output to predict the geospatial patterns of moraine preservation over multiple glacial cycles. This could be tested by comparing moraine ages and geospatial patterns from the database with model output to see where there is good fit between model output and data and where there isn't. +**1) Data-model comparison between LGM and penultimate moraine ages, and model output from simulations over multiple glaciations** (ie the ice sheet influence on regional climate example). **Hypothesis:** if the impact from ice sheets on re-arranging large-scale atmospheric circulation and thus modulating regional climate is actually significant, we should be able to use climate model output to predict the geospatial patterns of moraine preservation over multiple glacial cycles. This could be tested by comparing moraine ages and geospatial patterns from the database with model output to see where there is good fit between model output and data and where there isn't. 
  
 See the example output figure below and {{ :applications:ice_sheet_influence_exercise.pdf |find a link to the tutorial here.}} See the example output figure below and {{ :applications:ice_sheet_influence_exercise.pdf |find a link to the tutorial here.}}
Line 18: Line 13:
 ----  ---- 
  
-**3) Testing global expression of Younger Dryas**+**2) Testing global expression of Younger Dryas**
 Please find a tutorial and some matlab scripts used to generate some of the plots found in a recent paper from Greg Balco ([[https://www.annualreviews.org/doi/abs/10.1146/annurev-earth-081619-052609|Balco, 2021]]) in this example. Please find a tutorial and some matlab scripts used to generate some of the plots found in a recent paper from Greg Balco ([[https://www.annualreviews.org/doi/abs/10.1146/annurev-earth-081619-052609|Balco, 2021]]) in this example.
  
Line 28: Line 23:
 ----  ---- 
  
-**4) Post-Glacial Greenland ice-sheet retreat time-distance diagram** following up on a workshop at the University at Buffalo, we attempted to generate a time-distance diagram of SW Greenland Ice Sheet retreat and you can find the matlab script here.+**3) Post-Glacial Greenland ice-sheet retreat time-distance diagram** following up on a workshop at the University at Buffalo, we attempt to generate a time-distance diagram of SW Greenland Ice Sheet retreat and you can find the matlab script here to generate the following results:
  
-----+{{:applications:westgreenland_tdd_fig1.png?nolink&275|}}{{:applications:westgreenland_tdd_fig2.png?nolink&675|}}
  
-**5) Determining if measurement precision has gotten better through time** 
-This is a somewhat simple and fun exercise to investigate whether or not we as a community have been making progressively better measurements (ie improvements to field sample techniques, lab extraction procedures, AMS measurements, etc that should hopefully be leading to more precise cosmo measurements). 
  
-See the summary plot below that shows the story is a bit more complicated and perhaps we as a collective community have room for improvement.+The script combs through the new database and selects all ages from 'boulder' samples specifically from ICE-D Greenland that exist within the bounds box on the first figure and then plots the ages against longitude to convincingly show that as you move from the coast inland, ages generally get younger and track the recession of the Greenland Ice Sheet through the Holocene.
  
-{{ :applications:precision_over_time_output.png?nolink |}} 
- 
-please find the script that you can copy into a new script editor written to produce this plot here. 
 <code> <code>
  
-%%%% A matlab script for extracting all samples from the database that have sample collection date info +% Does deglaciation TDD for west Greenland....
-%%%% and plotting their % measurement uncertainty against collection date +
-%%%% to see if we collectively as a community have gotten any better over the years  +
-%%%% at making these measurements. Largest improvements I would expect +
-%%%% should come from laboratory procedures (ie lower level blanks, better +
-%%%% 10Be isolation techniques, AMS precision improvement (maybe), etc. +
- +
-%clear out all pre-existing variables and windows to start fresh+
 clear all; close all; clear all; close all;
  
-% as always, one must first connect to ICE-D.  +%% For MAC users use this set of code to connect to the database 
-% Before running the script, be sure that you are connected to ICE-D in +% First set up SSH tunnel on port 12345. 
-% Heidi SQL! +%dbc = database('iced','reader','beryllium-10','Vendor','MySQL','server','localhost','port',12345);
-% The script for connecting from a windows OS is as follows, +
-% assuming you set up an ODBC connection:+
  
-dbc = database('ICED-ALPINE','reader','beryllium-10');+%% For WINDOWS users use this set of code to connect to the database 
 +dbc = database('ICED2', 'reader', 'beryllium-10');
  
-% here is some code from Greg Balco for connecting to the database from a +%% get all samples
-% Mac computer. I have no idea if it actually works but here it is for +
-% y'all.+
  
-% Get Sequel Pro running and use +q1 = ['select iced.base_sample.lat_DD, iced.base_sample.lon_DD ' ... 
-% the below to find the SSH tunnel port. +    'from iced.base_sample,iced.base_application_sites,iced.base_application ' ... 
 +    'where iced.base_sample.site_id = iced.base_application_sites.site_id ' ... 
 +    'and iced.base_application_sites.application_id = iced.base_application.id ' ... 
 +    'and iced.base_application.Name = "Greenland"'];
  
-%[sys1,sys2] = system('ps aux | grep 3306'); +result1 = fetch(dbc,q1);
-%portindex = strfind(sys2,':173.194.241.211:'); +
-%portstr = sys2((portindex(1)-5):(portindex(1)-1));+
  
 +%% get West Greenland samples
  
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%+blats = [64.8 71]; 
 +%blats = [61 71]; 
 +blons = [-60 -48];
  
-% Once connected to ICE-D, we want to extract ALL samples from the database +q2 = ['select iced.base_sample.lat_DD, iced.base_sample.lon_DD ' ... 
-% that have sample collection date information found in the 'samples' tab. +    ' from iced.base_sample,iced.base_application_sites,iced.base_application ' ... 
-% here is an SQL query formatted in matlab to extract all samples and +    ' where iced.base_sample.lat_DD < ' sprintf('%0.1f',blats(2)) ... 
-% format the sample names, date collected, concentration of 10Be atoms and +    ' and iced.base_sample.lon_DD < ' sprintf('%0.1f',blons(2)) ... 
-% AMS measurement error into one table:+    ' and iced.base_sample.lat_DD > ' sprintf('%0.1f',blats(1)) ... 
 +    ' and iced.base_sample.lon_DD > ' sprintf('%0.1f',blons(1)) ... 
 +    ' and iced.base_sample.site_id = iced.base_application_sites.site_id ' ... 
 +    ' and iced.base_application_sites.application_id = iced.base_application.id ' ... 
 +    ' and iced.base_application.Name = "Greenland"'];
  
-q1 = ['select samples.sample_name, samples.date_collected, Be10_Al26_quartz.N10_atoms_g, Be10_Al26_quartz.delN10_atoms_g, calculated_ages.t_LSDn ' ... +result2 = fetch(dbc,q2);
-    ' from samples, Be10_Al26_quartz, calculated_ages ' ... +
-    ' where samples.sample_name = Be10_Al26_quartz.sample_name ' ... +
-    ' and samples.sample_name = calculated_ages.sample_name '... +
-    ' and samples.date_collected is not null '... +
-    ' and Be10_Al26_quartz.N10_atoms_g is not null '... +
-    ' and Be10_Al26_quartz.delN10_atoms_g is not null '... +
-    ' and samples.date_collected not like "0000-00-00" '... +
-    ' and samples.date_collected not like "0001-01-12" '];+
  
-% now, we want to store these selections as a table in matlab +locs1 = cell2mat(table2cell(result1)); 
-% and convert the table into column vectors for plotting. +locs2 = cell2mat(table2cell(result2));
-% this is made somewhat complicated by the fact that the dates extracted +
-% from ICE-D are in string format so you have to break them up.+
  
-samples.table = fetch(dbc,q1); +%% get age data from everything
-samples.date_cell = table2array(samples.table(:,2)); +
-samples.date_cell_split = split(samples.date_cell,["-"]); +
-samples.date_raw = str2double(samples.date_cell_split); +
-samples.date_final = double(samples.date_raw(:,1) + (samples.date_raw(:, 2)/12) + (samples.date_raw(:,3)/30)); +
-samples.conc = table2array(samples.table(:,3)); +
-samples.err = table2array(samples.table(:,4)); +
-samples.errpercent = double(((samples.err(:,1))./(samples.conc(:,1)))*100); +
-samples.age = table2array(samples.table(:,5));+
  
-% most important variables we created here are the samples.errpercent +q3 = ['select iced.base_sample.lat_DD, iced.base_sample.lon_DD, ' ... 
-% (sample error divided by sample concentration * 100), the final date +    'iced.base_calculatedages.t_St,iced.base_calculatedages.t_LSDn,iced.base_calculatedages.dtint_St, ' ... 
-% collected (year + month/12 + day/30) and the age of each sample to make +    'iced.base_calculatedages.dtext_St ' ... 
-% a scatterplot of percent error versus time (with marker size dictated by age)+    'from iced.base_sample, iced.base_calculatedages, iced.base_application_sites,iced.base_application' ... 
 +    ' where iced.base_sample.id = iced.base_calculatedages.sample_unique_id ' ... 
 +    ' and iced.base_sample.site_id = iced.base_application_sites.site_id ' ... 
 +    ' and iced.base_application_sites.application_id = iced.base_application.id ' ... 
 +    ' and iced.base_application.Name = "Greenland"' ... 
 +    ' and lat_DD < ' sprintf('%0.1f',blats(2)) ' and lon_DD < ' sprintf('%0.1f',blons(2)) ' and lat_DD > ' sprintf('%0.1f',blats(1)) ' and lon_DD > ' sprintf('%0.1f',blons(1))];
  
-scatter(samples.date_final, samples.errpercent, 'filled',... +result3 = fetch(dbc,q3);
-     'SizeData',(samples.age ./ 1000), 'MarkerEdgeColor',[0 0 0],... +
-     'MarkerFaceColor',[0 .7 .7],'LineWidth', 0.75) +
-ylim([0 100]) +
-set(gca, 'YScale', 'log') +
-ylabel('percent error','FontSize',16) +
-xlabel('sample measurement date','FontSize',16)+
  
-lsline+%% 
  
-% and that is it! Does this plot actually show that we are improving as a +ages1 = cell2mat(table2cell(result3));
-% community?+
  
 +% And just from boulders
 +q4 = [q3 ' and iced.base_sample.what like "%boulder"'];
 +
 +result4 = fetch(dbc,q4);
 +
 +ages2 = cell2mat(table2cell(result4));
 +
 +close(dbc);
 +
 +%% 
 +
 +figure(1); clf; 
 +% Center on Greenland Summit
 +gisplat = 72 + 36/60;
 +gisplon = -38.5;
 +
 +xl = [-0.18 0.15];
 +yl = [-0.25 0.25];
 +
 +tx = xl(1) + 0.95*diff(xl);
 +ty = yl(1) + 0.06*diff(yl);
 +ts = 14;
 +
 +mw = 0.005;
 +awx = (1 - 5*mw)/4;
 +awy = (1-2*mw);
 +
 +axesm('mapprojection','eqdazim','origin',[gisplat gisplon 0]);
 +
 +%axesm('globe','grid','on')
 +
 +
 +g1 = geoshow('landareas.shp');
 +set(get(g1,'children'),'facecolor',[0.9 0.9 0.9]);
 +set(gca,'xlim',xl); 
 +set(gca,'ylim',yl);
 +hold on;
 +
 +
 +% Plot all samples
 +plotm(locs1(:,1),locs1(:,2),'ko','markerfacecolor',[0.7 0.7 1]);
 +% Plot only samples in west greenland selection box
 +plotm(locs2(:,1),locs2(:,2),'ko','markerfacecolor',[0.2 0.2 1]);
 +
 +
 +plotm(blats([1 1 2 2 1]),blons([1 2 2 1 1]),'b','linewidth',1);
 +
 +
 +
 +%
 +
 +set(gca,'box','off','xcolor',[1 1 1],'ycolor',[1 1 1]);
 +%plotm(gisplat,gisplon,'ko','markerfacecolor','w');
 +%text(tx,ty,[sprintf('%0.0f',100*f2) '%'],'fontsize',ts,'horizontalalignment','right');
 +
 +hg = gridm('on');
 +set(hg,'clipping','on');
 +set(hg,'linestyle','-','color',[0.75 0.75 0.75]);
 +
 +temp = jet(12);
 +for a = 1:12
 +    maxage = 11000 - (a-1).*500;
 +    minage = maxage - 500;
 +    these = find(ages2(:,3) < maxage & ages2(:,3) > minage);
 +    if length(these) > 0
 +        plotm(ages2(these,1),ages2(these,2),'o','color',[0.5 0.5 0.5],'markerfacecolor',temp(a,:));drawnow;
 +    end
 +end
 +
 +
 +
 +
 +
 +%% Plot
 +
 +figure(2); clf; 
 +%plot(ages1(:,2),ages1(:,3),'bo','markerfacecolor',[0.5 0.5 1]);
 +hold on;
 +
 +for a = 1:size(ages2,1)
 +    xx = [1 1].*ages2(a,2);
 +    yy = ages2(a,3) + [-1 1].*ages2(a,5);
 +    plot(xx,yy,'r'); hold on;
 +end
 +
 +plot(ages2(:,2),ages2(:,3),'ko','markerfacecolor','r');
 +set(gca,'ylim',[5000 15000]); grid on;
 +
 +%% Plot proposed cold periods from Young and others, etc. 
 +
 +hold on;
 +
 +events = [11620 10410 9090 8050 7300];
 +devents = [430 350 260 220 310];
 +
 +for a = 1:length(events)
 +    xx = [-54 -49]; yy = [events(a) events(a)];
 +    plot(xx,yy,'b','linewidth',1);
 +    xx = [-54 -49 -49 -54 -54];
 +    yy = [events(a)-devents(a) events(a)-devents(a) events(a)+devents(a) events(a)+devents(a) events(a)-devents(a)];
 +    patch(xx,yy,[0.8 0.8 1],'edgecolor','none','facealpha',0.1);
 +end
 + 
 +%% filter for better AEP performance
 +p1 = polyfit([-53.5 -49.88],[11500 6426],1);
 +px = [-54 -49]; py = polyval(p1,px); %plot(px,py,'g');
 +
 +predt = polyval(p1,ages2(:,2));
 +okclip = find(abs(predt - ages2(:,3)) < 1000);
 +plot(ages2(okclip,2), ages2(okclip,3),'ko','markersize',8);
 +
 +aept = ages2(okclip,3);
 +aepdt = ages2(okclip,5);
 +aepz = -ages2(okclip,2).*100 - 4900;
 +
 +% Get bounding curve
 +
 +out = aep_bound(aept,aepz,0);
 +figure(2);
 +aept_out = out.t; aepx_out = -(out.z + 4900)./100;
 +plot(aepx_out,aept_out,'k'); 
 +
 +%% MCS bounding curve
 +
 +% Interestingly, this is weirdly pathological because at the westings where
 +% there are a lot of data, almost all the random iterations have one
 +% outlier that pulls it out. So it actually doesn't really work. 
 +
 +if 1
 +
 +figure; 
 +plot(aept,aepz,'ro'); hold on; drawnow;
 +p1 = plot(1000,1000,'ro');
 +ni = 100;
 +intt = 6500:10:12000;
 +for a = 1:ni
 +    thist = randn(size(aept)).*aepdt + aept;
 +    thisz = aepz;
 +    out = aep_bound(thist,thisz,0);
 +    delete(p1);
 +    p1 = plot(thist,thisz,'bo');
 +    plot(out.t,out.z,'k'); drawnow; pause;
 +    %inty(a,:) = interp1(out.t,out.z,intt);
 +    %plot(intt,inty(a,:),'k');
 +    disp(a);
 +end
 +
 +end
 +    
 +%% 
 +
 +if 1
 +    
 +    % Make figure comparing this to Holocene "events" 
 +
 +    figure;
 +    diffx = diff(aepx_out)./diff(aept_out);
 +    for a = 1:length(diffx)
 +        xx = [aept_out(a) aept_out(a+1) aept_out(a+1)];
 +        if a == length(diffx)
 +            yy = -[diffx(a) diffx(a) diffx(a)];
 +        else
 +            yy = -[diffx(a) diffx(a) diffx(a+1)];
 +        end
 +        plot(xx,yy,'r'); hold on;
 +    end
 +
 +
 +    set(gca,'xlim',[6000 12000]);
 +    grid on;
 +    for a = 1:length(events)
 +        yy = [0 5e-3]; xx = [events(a) events(a)];
 +        plot(xx,yy,'b','linewidth',1);
 +        yy = [0 5e-3 5e-3 0 0];
 +        xx = [events(a)-devents(a) events(a)-devents(a) events(a)+devents(a) events(a)+devents(a) events(a)-devents(a)];
 +        patch(xx,yy,[0.8 0.8 1],'edgecolor','none','facealpha',0.1);
 +    end
 +    
 +end
  
 </code> </code>
  
 +----
  
-**Follow up** as it should be obvious from this follow up plot, there is a strong relationship between sample concentration and percent error. This should make sense because the higher the concentration of Be-10 in the sample, the smaller % of that is from background Be-10 (ie Be-10 that was not originally trapped within sample quartz grains but was rather either from meteoric Be-10 that contaminated samples, Boron contamination that has the same atomic mass as Be-10, and/or Be-10 that came with the Be-9 carrier). See the plot below and the matlab script used to make the plot.+**4) Determining if measurement precision has gotten better through time** 
 +This is a somewhat simple and fun exercise to investigate whether or not we as a community have been making progressively better measurements (ie improvements to field sample techniques, lab extraction procedures, AMS measurements, etc that should hopefully be leading to more precise cosmo measurements).
  
-{{ :applications:precision_v_conc.png?nolink |}}+See the summary plots below that show the story is a bit more complicated; essentially one must remove the effect of sample concentration on sample percent error.
  
-and the code to make this plot below:+**Plot One** 
 +This is what you get when you plot measurement errors (%) for all samples in the database (with sample collection dates) against their sample collection dates. There is a linear trend fitted to the data with a very poor R^2 value. 
 + 
 + 
 +{{ :applications:precision_vs_time.png?nolink |}} 
 + 
 + 
 +**Plot 2**  
 +As it should be obvious from the plot, there is a notable power relationship between sample concentration and percent error. This should make sense because the higher the concentration of Be-10 in the sample, the smaller % of that is from background Be-10 (ie Be-10 that was not originally trapped within sample quartz grains but was rather either from meteoric Be-10 that contaminated samples, Boron contamination that has the same atomic mass as Be-10, and/or Be-10 that came with the Be-9 carrier). 
 + 
 +{{ :applications:precision_vs_concentration.png?nolink |}} 
 + 
 +**Plot 3** 
 +So what we need to do is detrend the data based on this power relationship. An easy way to do this is to simply calculate an expected value (based on the power relationship) and subtract that expected value from the observed measurement error. I could have made another plot to demonstrate this, but essentially if you plot detrended measurement errors against sample concentrations, the trendline you could fit to the data will be horizontal and centered at zero. 
 + 
 +The final plot below shows what happens when you plot detrended measurement errors against sample collection date. Notice that the R^2 value is still relatively low but did increase by a factor of 4 and the slope became slightly more negative. Maybe this very loose correlation does say something about community improvement... What do you think? 
 + 
 +{{ :applications:detrended_precision_vs_time.png?nolink |}} 
 + 
 + 
 +and the code to make these plots is below:
  
 <code> <code>
 +
 +%%%% A matlab script for extracting all 10Be measurements from the alpine/entire database that have sample collection date info
 +%%%% and plotting their % measurement uncertainty against collection date
 +%%%% to see if we collectively as a community have gotten any better over the years 
 +%%%% at making these measurements. Largest improvements I would expect
 +%%%% should come from laboratory procedures (ie lower level blanks, better
 +%%%% 10Be isolation techniques, AMS precision improvement (maybe), etc.
  
 %clear out all pre-existing variables and windows to start fresh %clear out all pre-existing variables and windows to start fresh
Line 144: Line 320:
 % assuming you set up an ODBC connection: % assuming you set up an ODBC connection:
  
-dbc = database('ICED-ALPINE','reader','beryllium-10');+dbc = database('ICED2','reader','beryllium-10');
  
 % here is some code from Greg Balco for connecting to the database from a % here is some code from Greg Balco for connecting to the database from a
Line 166: Line 342:
 % AMS measurement error into one table: % AMS measurement error into one table:
  
-q1 = ['select samples.sample_name, samples.date_collected, Be10_Al26_quartz.N10_atoms_g, Be10_Al26_quartz.delN10_atoms_g, calculated_ages.t_LSDn ' ... +%% Query to extract all 10Be data from the alpine database 
-    ' from samples, Be10_Al26_quartz, calculated_ages ' ... + 
-    ' where samples.sample_name = Be10_Al26_quartz.sample_name ' ... +q1 = ['select iced.base_sample.name, iced.base_sample.date_collected, iced._be10_al26_quartz.N10_atoms_g, iced._be10_al26_quartz.delN10_atoms_g, iced.base_calculatedages.t_LSDn ' ... 
-    ' and samples.sample_name = calculated_ages.sample_name '... +    ' from iced.base_sample, iced._be10_al26_quartz, iced.base_calculatedages, iced.base_site, iced.base_application_sites ' ... 
-    ' and samples.date_collected is not null '... +    ' where iced.base_sample.id = iced._be10_al26_quartz.sample_id ' ... 
-    ' and Be10_Al26_quartz.N10_atoms_g is not null '... +    ' and iced.base_sample.id = iced.base_calculatedages.sample_unique_id ' ... 
-    ' and Be10_Al26_quartz.delN10_atoms_g is not null '... +    ' and iced.base_sample.site_id = iced.base_site.id ' ... 
-    ' and samples.date_collected not like "0000-00-00" '... +    ' and iced.base_site.id = iced.base_application_sites.site_id ' ... 
-    ' and samples.date_collected not like "0001-01-12" '];+    ' and iced.base_sample.date_collected is not null ' ... 
 +    ' and iced._be10_al26_quartz.N10_atoms_g is not null ' ... 
 +    ' and iced._be10_al26_quartz.delN10_atoms_g is not null ' ... 
 +    ' and iced.base_sample.date_collected not like "0000-00-00" ' ... 
 +    ' and iced.base_sample.date_collected not like "0001-01-12" ' ... 
 +    ' and iced.base_application_sites.application_id = 2 ']; 
 + 
 +%% If you would like to try the entire database (not just the alpine database) use this query instead of the first one 
 + 
 +%q1 = ['select iced.base_sample.name, iced.base_sample.date_collected, iced._be10_al26_quartz.N10_atoms_g, iced._be10_al26_quartz.delN10_atoms_g, iced.base_calculatedages.t_LSDn ' ... 
 +%    ' from iced.base_sample, iced._be10_al26_quartz, iced.base_calculatedages ' ... 
 +%    ' where iced.base_sample.id = iced._be10_al26_quartz.sample_id ' ... 
 +%    ' and iced.base_sample.id = iced.base_calculatedages.sample_unique_id ' ... 
 +%    ' and iced.base_sample.date_collected is not null ' ... 
 +%    ' and iced._be10_al26_quartz.N10_atoms_g is not null ' ... 
 +%    ' and iced._be10_al26_quartz.delN10_atoms_g is not null ' ... 
 +%    ' and iced.base_calculatedages.t_LSDn not like "0" ' ... 
 +%    ' and iced.base_sample.date_collected not like "0000-00-00" ' ... 
 +%    ' and iced.base_sample.date_collected not like "0001-01-12" ']; 
 + 
 +%% gather the data and organize it into cell arrays that can be used for plotting 
 % now, we want to store these selections as a table in matlab % now, we want to store these selections as a table in matlab
 % and convert the table into column vectors for plotting. % and convert the table into column vectors for plotting.
Line 189: Line 386:
 samples.errpercent = double(((samples.err(:,1))./(samples.conc(:,1)))*100); samples.errpercent = double(((samples.err(:,1))./(samples.conc(:,1)))*100);
 samples.age = table2array(samples.table(:,5)); samples.age = table2array(samples.table(:,5));
 +
  
 % most important variables we created here are the samples.errpercent % most important variables we created here are the samples.errpercent
Line 195: Line 393:
 % a scatterplot of percent error versus time (with marker size dictated by age) % a scatterplot of percent error versus time (with marker size dictated by age)
  
 +%% Figure 1 - percent error by date measured without detrending the influence of sample concentration
 +
 +fig1 = figure(1)
 +scatter(samples.date_final, samples.errpercent, 'filled',...
 +     'SizeData',(samples.age ./ 1000), 'MarkerEdgeColor',[0 0 0],...
 +     'MarkerFaceColor',[0 .7 .7],'LineWidth', 0.75)
 +ylim([0 100])
 +set(gca, 'YScale', 'log')
 +ylabel('measurement error [%]','FontSize',16)
 +xlabel('sample measurement date [CE]','FontSize',16)
 +dim1 = [.15 .9 0 0];
 +str1 = {['Linear best fit:'], ['m = -0.0669'], ['b = 140.15'], ...
 +    ['R^2 = 0.0003']};
 +annotation('textbox',dim1,'String',str1,'FitBoxToText','on', 'FontSize', 16);
 +hold on
 +x1 = samples.date_final;
 +y1 = -0.0669 * x1 + 140.15;
 +plot(x1,y1, "Color", [0 0 0],"LineStyle","-","LineWidth", 2)
 +
 +% and that is the simple approach but as we know, there is a dependence 
 +% of precision on sample concentration (ie higher concentration = generally
 +% lower error since the background Be is less impactful for hot samples.
 +% to get around this, let's weigh each sample based on this relationship
 +% (see the script below and the resulting figure that demonstrates this 
 +% what-appears-to-be a power relationship.
 +
 +%% Figure 2 establishing the relationship between sample concentration and percent error
 +
 +fig2 = figure(2)
 scatter(samples.conc, samples.errpercent, 'filled',... scatter(samples.conc, samples.errpercent, 'filled',...
      'MarkerEdgeColor',[0 0 0],...      'MarkerEdgeColor',[0 0 0],...
Line 200: Line 427:
 ylim([0 100]) ylim([0 100])
 set(gca, 'Yscale', 'log', 'Xscale', 'log') set(gca, 'Yscale', 'log', 'Xscale', 'log')
-ylabel('percent error [%]','FontSize',16)+ylabel('measurement error [%]','FontSize',16)
 xlabel('sample concentration [atom/g]','FontSize',16) xlabel('sample concentration [atom/g]','FontSize',16)
 +dim2 = [.15 .9 0 0];
 +str2 = {['Power best fit:'], ['a = 65.186'], ['b = -0.216'], ...
 +    ['R^2 = 0.1005']};
 +annotation('textbox',dim2,'String',str2,'FitBoxToText','on', 'FontSize', 16);
 +hold on
 +x2= samples.conc;
 +y2=65.185 * x2.^-0.216;
 +plot(x2,y2, "Color", [0 0 0],"LineStyle","-","LineWidth", 2)
 +
 +
 +% so now, let's plot percent errors weighted by the equation fit to 
 +% the precision vs concentration plot above. Simplest way to do this is to 
 +%detrend the data (ie generate a curve of expected values given the
 +%distribution - the fitted power curve - and subtract the expected values
 +%from the actual measurements. Then, plot the difference against the
 +%measurement date to remove the influence of sample concentration on error
 +%percent since we only want to know if error percent is decreasing over
 +%time.
 +
 +%the way I have been thinking of it is like this: if we made the same
 +%measurement on the same exact sample every time (ie the expected
 +%concentration should be the same every time) for the past 30 years, and if
 +%we actually have been making better measurements, we should still see a
 +%decreasing trend in percent error over time (even though it is the exact
 +%same sample that we are measuring every time). Obviously this is not true,
 +%we've been making all sorts of measurements over the past 30 years and so
 +%this is an attempt to detrend the data in concentration vs percent error
 +%space and plot the detrended data against date of measurement. Anyway...
 +
 +%% Figure 3 plotting percent error by date measured after accounting for influence of concentration on percent error
 +
 +fig3 = figure(3)
 +y3 = samples.errpercent - y2;
 +scatter(samples.date_final, y3, 'filled',...
 +     'SizeData',(samples.age ./ 1000), 'MarkerEdgeColor',[0 0 0],...
 +     'MarkerFaceColor',[0 .7 .7],'LineWidth', 0.75)
 +ylim([-10 10])
 +ylabel('detrended error [%]','FontSize',16)
 +xlabel('sample measurement date [CE]','FontSize',16)
 +dim3 = [0.15 .9 0 0];
 +str3 = {['Linear best fit:'], ['m = -0.1289'], ['b = 260.3'], ...
 +    ['R^2 = 0.0012']};
 +annotation('textbox',dim3,'String',str3,'FitBoxToText','on', 'FontSize', 16);
 +hold on
 +x3= samples.date_final;
 +y3=-0.1289 * x3 + 260.3;
 +plot(x3,y3, "Color", [0 0 0],"LineStyle","-","LineWidth", 2)
  
 </code> </code>
  
-And so, the plot made comparing percent error with date of measurement could potentially be improved by attempting to control for sample concentration based on the relationship between sample concentration and percent error plotted in this follow up (eg weigh each sample by its concentration based on the relationship between concentration and percent error). Nice!+
  
 ---- ----
  
-**6) Is there a correlation between Al/Be ratios and sample elevation?**+**5) Is there a correlation between Al/Be ratios and sample elevation?**
 This example is based off a recent publication ([[https://www.mdpi.com/2076-3263/11/10/402|Halsted et al., 2021]]) that found there is a correlation between Al / Be ratios and elevation that likely needs to be taken into account. This example is based off a recent publication ([[https://www.mdpi.com/2076-3263/11/10/402|Halsted et al., 2021]]) that found there is a correlation between Al / Be ratios and elevation that likely needs to be taken into account.
  
Line 234: Line 508:
 % assuming you set up an ODBC connection: % assuming you set up an ODBC connection:
  
-dbc = database('ICED-ALPINE','reader','beryllium-10');+dbc = database('ICED2','reader','beryllium-10');
  
 % here is some code from Greg Balco for connecting to the database from a % here is some code from Greg Balco for connecting to the database from a
Line 258: Line 532:
 % dictating marker size). % dictating marker size).
  
-q1 = ['select samples.sample_name, samples.elv_m, Be10_Al26_quartz.N10_atoms_g, Be10_Al26_quartz.delN10_atoms_g, Be10_Al26_quartz.N26_atoms_g, '... +%% Query to extract all 10Be Al26 pairs from the alpine database
-' Be10_Al26_quartz.delN26_atoms_g, calculated_ages.t_LSDn '... +
-' from samples, Be10_Al26_quartz, calculated_ages '... +
-' where samples.sample_name = Be10_Al26_quartz.sample_name '... +
-' and samples.sample_name = calculated_ages.sample_name '... +
-' and samples.elv_m is not null and Be10_Al26_quartz.N10_atoms_g is not null and Be10_Al26_quartz.N26_atoms_g is not null '... +
-' and Be10_Al26_quartz.N26_atoms_g > 1000 and Be10_Al26_quartz.N26_atoms_g not like 0 '];+
  
-% this will store all data in a table that we can format into cell arrays +q1 = ['select iced.base_sample.name, ' ... 
-% for plotting with the following code:+' iced.base_sample.elv_m, ' ... 
 +' iced._be10_al26_quartz.N10_atoms_g, ' ... 
 +' iced._be10_al26_quartz.delN10_atoms_g, ' ... 
 +' iced._be10_al26_quartz.N26_atoms_g, ' ... 
 +' iced._be10_al26_quartz.delN26_atoms_g, ' ... 
 +' iced.base_calculatedages.t_LSDn ' ... 
 +' from iced.base_sample, iced._be10_al26_quartz, iced.base_calculatedages, iced.base_site, iced.base_application_sites ' ... 
 +' where iced.base_sample.id = iced._be10_al26_quartz.sample_id ' ... 
 +' and iced.base_sample.id = iced.base_calculatedages.sample_unique_id ' ... 
 +' AND iced.base_sample.site_id = iced.base_site.id ' ... 
 +' and iced.base_site.id = iced.base_application_sites.site_id ' ... 
 +' and iced.base_sample.elv_m is not NULL ' ... 
 +' and iced._be10_al26_quartz.N10_atoms_g is not NULL ' ... 
 +' and iced._be10_al26_quartz.N26_atoms_g is not null ' ... 
 +' and iced._be10_al26_quartz.N26_atoms_g > 1000 ' ... 
 +' and iced._be10_al26_quartz.N26_atoms_g not like 0 ' ... 
 +' and iced.base_application_sites.application_id = 2 ']; 
 + 
 +%% If you would like to try the entire database (not just the alpine database) use this query instead of the first one 
 +% Worth noting that in this query, we have not isolated the samples that 
 +% have strictly simple exposure histories so there are probably a lot of 
 +% samples in the dataset that are below the simple exposure ratio line 
 + 
 +%q1 = ['select iced.base_sample.name, ' ... 
 +%' iced.base_sample.elv_m, ' ... 
 +%' iced._be10_al26_quartz.N10_atoms_g, ' ... 
 +%' iced._be10_al26_quartz.delN10_atoms_g, ' ... 
 +%' iced._be10_al26_quartz.N26_atoms_g, ' ... 
 +%' iced._be10_al26_quartz.delN26_atoms_g, ' ... 
 +%' iced.base_calculatedages.t_LSDn ' ... 
 +%' from iced.base_sample, iced._be10_al26_quartz, iced.base_calculatedages ' ... 
 +%' where iced.base_sample.id = iced._be10_al26_quartz.sample_id ' ... 
 +%' and iced.base_sample.id = iced.base_calculatedages.sample_unique_id ' ... 
 +%' and iced.base_sample.elv_m is not NULL ' ... 
 +%' and iced._be10_al26_quartz.N10_atoms_g is not NULL ' ... 
 +%' and iced._be10_al26_quartz.N26_atoms_g is not null ' ... 
 +%' and iced._be10_al26_quartz.N26_atoms_g > 1000 ' ... 
 +%' and iced._be10_al26_quartz.N26_atoms_g not like 0 ' ... 
 +% and iced.base_calculatedages.t_LSDn not like 0 ' ... 
 +%' AND iced.base_sample.name NOT LIKE "MH12-47"']; 
 + 
 +%% gather the data and organize it into cell arrays that can be used for plotting
  
 samples.table = fetch(dbc,q1); samples.table = fetch(dbc,q1);
Line 278: Line 587:
 samples.error = double((samples.cell_array(:,4)./samples.cell_array(:,2)).*(((samples.cell_array(:,3)./samples.cell_array(:,2)).^2)+((samples.cell_array(:,5)./samples.cell_array(:,4)).^2)).^0.5); samples.error = double((samples.cell_array(:,4)./samples.cell_array(:,2)).*(((samples.cell_array(:,3)./samples.cell_array(:,2)).^2)+((samples.cell_array(:,5)./samples.cell_array(:,4)).^2)).^0.5);
  
-%now time to plot it all!+%% The code to maka a figure plotting up all of the samples by elevation vs Al/Be ratio
  
 +fig1 = figure(1)
 errorbar(samples.cell_array(:,1), (samples.cell_array(:,4)./samples.cell_array(:,2)), samples.error, 'vertical', 'LineStyle','none') errorbar(samples.cell_array(:,1), (samples.cell_array(:,4)./samples.cell_array(:,2)), samples.error, 'vertical', 'LineStyle','none')
 hold on hold on
Line 286: Line 596:
     'MarkerEdgeColor',[0 0 0], 'MarkerFaceColor', [0 .7 .7], 'SizeData', (samples.cell_array(:,6)./1000),...     'MarkerEdgeColor',[0 0 0], 'MarkerFaceColor', [0 .7 .7], 'SizeData', (samples.cell_array(:,6)./1000),...
     'LineWidth', 0.5)     'LineWidth', 0.5)
 +ylim([0 15])
 xlabel('Elevation (m)','FontSize',16) xlabel('Elevation (m)','FontSize',16)
 ylabel('[Al-26] / [Be-10]','FontSize',16) ylabel('[Al-26] / [Be-10]','FontSize',16)
 +dim1 = [.8 .9 0 0];
 +str1 = {['Linear best fit:'], ['m = -0.0002'], ['b = 6.4274'], ...
 +    ['R^2 = 0.0375']};
 +annotation('textbox',dim1,'String',str1,'FitBoxToText','on', 'FontSize', 16);
 hold on hold on
- +x1 = samples.cell_array(:,1); 
-% from here, one could do some regressions, fit a line to the data and one +y1=-.0002 * x1 + 6.4274; 
-% might find a negative correltaion between Al/Be ratios and elevation.  +plot(x1,y1, "Color", [0 0 0],"LineStyle","-","LineWidth", 2)
- +
-% At the moment I just added this lsline function in the Statistics and +
-% Machine Learning toolbox that fits a linear regression to the data. Still figuring out how to do more statistics on +
-% it (ie confidence interval, displaying regression equation and r2, etc.) +
- +
-lsline+
  
 %thanks for following along! %thanks for following along!
Line 305: Line 614:
 ---- ----
  
-**7) Heinrich Stadials aridity drives glacier retreat in the Mediterranean?**+**6) Heinrich Stadials aridity drives glacier retreat in the Mediterranean?**
 This example is a follow up to a paper recently published in Nature Geoscience ([[https://www.nature.com/articles/s41561-021-00703-6|Allard et al., 2021]]) that found that glaciers in the region may have been retreating during Heinrich Stadials due to more arid conditions.  This example is a follow up to a paper recently published in Nature Geoscience ([[https://www.nature.com/articles/s41561-021-00703-6|Allard et al., 2021]]) that found that glaciers in the region may have been retreating during Heinrich Stadials due to more arid conditions. 
  
Line 316: Line 625:
 ---- ----
  
-**8) Identifying regions of possible heavy moraine degradation** (using the moraine ages and land degradation models incorporated into the middle layer of calculations) and **comparing identified areas of high degradation to geohazards** (plate boundaries and areas of high seismic activity). +**7) Identifying regions of possible heavy moraine degradation** (using the moraine ages and land degradation models incorporated into the middle layer of calculations) and **comparing identified areas of high degradation to geohazards** (plate boundaries and areas of high seismic activity). 
 **Hypothesis:** if geohazards actually present a notable obstacle to moraine dating through moraine degradation, then we should be able to find instances of high moraine degradation coinciding with high seismic activity. **Hypothesis:** if geohazards actually present a notable obstacle to moraine dating through moraine degradation, then we should be able to find instances of high moraine degradation coinciding with high seismic activity.