Open Notebook

TC/TN#001 and IC/ON#001 mixed standards

Thu, 23 Aug 2012 10:47:51 GMT

C 22.6,56.4,112.3,167.6,330.2,641.8,936.2,1214 uM
N 6.8,17,34,50,99,193,281,364 uM

stacked and grouped bar plots in R

Tue, 10 Jul 2012 15:41:00 GMT

-> take the sum of each group and plot them with barplot, then overlay rectangles for subgroup stacking using symbol()

boxvector<-c(G_Wa,G_Wtot,A_Wa,A_Wtot,V_Wa, V_Wtot)	
boxes<-array(boxvector,dim=c(2,3),dimnames=list(c("W_a","W_tot"),c("Granite","Andesite","Vermiculite")))
	
# make the plot of Wa and Wtot beside each otehr
ylabel<-expression(paste("Mean weathered amount /  ",mu,"moles",sep=""))
barplot(	
	boxes,
        beside=TRUE,
	space=c(0,1),
	col=c("lightblue","darkblue"),
	cex.lab=1.2,
	cex.names=1.2,
	main=element,
	#ylab= ylabel,
	ylim=c(0,(max(boxes[2,])+max(Wtot_err)))
)
	
#arrange the data for Wmoss for an overlay rectangle
W_moss<-c(G_Wmoss,A_Wmoss,V_Wmoss)
rectmatrix<-matrix(nrow=3,ncol=2)
rectmatrix[,1]<-c(1,1,1)
rectmatrix[,2]<-W_moss
midpoints<-as.double(boxes[2,])-W_moss/2
symbols(x=c(2.5,5.5,8.5),y=midpoints,rectangles=rectmatrix,add=TRUE,inches=FALSE,bg="chartreuse3")

#draw error bars
abiox0<-c(1.5,4.5,7.5)
abioy0<-boxes[1,]-Wa_err
abioy1<-boxes[1,]+Wa_err
arrows(abiox0,abioy0,abiox0,abioy1,angle=90,code=3,length=0.125)
totx0<-c(2.5,5.5,8.5)
toty0<-boxes[2,]-Wtot_err	
toty1<-boxes[2,]+Wtot_err
arrows(totx0,toty0,totx0,toty1,angle=90,code=3,length=0.125)

which makes something that looks like this:

except groups are different - not element, as in image, but substrate (granite, vermiculite, andesite)

What's the deal with trace gas exchange between the ocean and atmosphere

Wed, 06 Jun 2012 13:39:00 GMT

Need to write 'trace gas flux 101' for a book chapter. I want to make it more useful to people new to the field than the usual regurgitation of the theory and presentation of the data that doesn't point out what it is you really need to know if you want to calculate a flux. What are the key points the chapter needs to get accross?

concentration uncertainty is greater than transfer velocity uncertainty for most or all of the (short-lived) trace gases
water side transfer velocity parameterisations that you might apply to your trace gas flux problem are empirically-derived (at some level) from observations of the flux of particular gases, or possibly heat and momentum
air-side transfer velocity parameterisations are based on micrometeorological models of water vapour transfer and poorly validated
It is difficult to reconcile the wind-speed relationships from different types of gas exchange experiments. Diffrerences, if real, tell us something important.
Wind is not a direct mechanistic predictor of transfer velocity, but is a useful and pragmatic choice of proxy!
Whilst there are significant uncertainties associated with the wind dependence of kw parameterisations, other effects are likely to introduce greater uncertainty into gas exchange estimates when applying the parameterisation to other gases, particularly more soluble ones for kw
bubble effects are parameterised into kw but is actually a separate flux term with k and delta C aspects.

and the 'classic' gas exchange text book sections are

gas exchange models
- thin / stagnant film model
  - k-delta-c
  - derive ka /kw etc
- other models
  - e.g. surface renewal, boundary layer etc.
bubbles
other processes
- e.g. chemical enhancement, rain, surfactants etc.
parameterisations
- wind tunnel studies
- global 14C constraint
- tracer
- direct measurement (e.g eddy covariance)
- noaa coare

so, how to organise the chapter?

introduction
- take kDc out of thin film model and generalise in section on the air-water interface like in Bernt Jahne's book chapter.
- derive ka/kw and solubility dependence of partitioning
- turbulent forcing -> windspeed parameterisations-> form and schmidt number dependence
- approaches to calculating gas fluxes
Schmidt number dependence
- interfacial models
- micrometeorological models
processes driving gas exchange
- wind
- waves
- bubbles
- temperature, humidity and rain effects
- surfactants
- chemical and biological enhancement
- others
parameterising gas exchange with windspeed
- kw
  - Global constraint
  - Natural tracer
  - Deliberate tracer
  - Direct measurements
  - the role of bubbles
  - reconciling observations
- ka
non-windspeed based parameterisations
- the role of remote sensing
Gas-dependent and gas-specific effects
Other considerations

Elemental solubilities from XRF and aqueous phase analyses

Thu, 29 Mar 2012 15:24:00 GMT

Have XRF composition in umol/g and aqueous (equilibrium) concentrations in umol/l

elemental solubility is aqueous concentration / XRF

-> units are g/l

Aluminium has HIGH concentration in substrate but a relatively low concentration in the aqueous phase (elemental solubility 0.009 g/l)
Calcium has LOW concentration in substrate but high concentration in aqueous phase (elemental solubility ~10 g/l)

A good description of the meaning of the quantity is the mass of substrate which would be needed to dissolve stoichiometrically in 1L of water to give the observed aqueous concentration of the analyte. I.e. a disproportionately soluble element in a substrate requires a greater amount of stoichiometric dissolution than a disproportionately insoluble one.

Example: mineral has stoichiometry of elements X and Y: X6Y1 e.g. 6 umol/g X and 1 umol/g Y

In equilibrium with water the aqueous phase concentrations are observed to be 0.1 umol/L X and 10 umol/L Y

So elemental solubility of X is 0.1/6 = 0.017 and of Y is 10/1 = 10

This means that 10g of substrate would have to be stoichiometrically dissolved in 1L to give 10umol/L of Y, and 0.017g for X i.e. Y is disproportionately soluble relative to X.

ODV spreadsheet template

Mon, 10 Oct 2011 20:35:00 GMT

Using Ocean Data View at the moment:

http://odv.awi.de/

Useful to have a quick spreadsheet template with basic standard column headings:

https://docs.google.com/spreadsheet/pub?hl=en_GB&hl=en_GB&key=0AqRtHz1X9oQ6dElfU3RseTVmWkNyQ0VVWTZkVlFaeEE&output=html

I'm still here!

Tue, 23 Aug 2011 15:48:00 GMT

I've been busy with job interviews, paper writing and holidays, hence the lack of content for the last couple of months. Am just off to teach on the SOLAS summer school in Corsica for a couple of weeks, then I'll be beck generating research again. I haven't abandoned my notebook, honest.

Nitrogen cycling for archaen model

Tue, 14 Jun 2011 15:18:00 GMT

In preparation for my lectureship interview in a couple of weeks I've been thinking about the connections between the N and O cycles during the rise of oxygenic photosynthesis. This has been though about before, see the papers by Fennel et al (2005) and followers, and there are some reasonably obvious potential effects, not all of which appear to have been put together into a single model. This might be an interesting thing to have a bash at in advance of the interview and talk that I have to give at the end of the month.

Here are the processes which I think would be interesting to include:

Reduced iron supply, cycling by anoxygenic photosynthesis, sedimentation or oxidation
Oxygenic photosynthesis
Nitrogen fixation and the oxygen / reduced iron requirements thereof
Nitrification and denitrification and the oxygen dependencies of both
Dissimilatory nitrate reduction (denitrification) to ammonium
Annamox and N2O production?
NO3 / NH4 uptake.
remineralisation of organic N to NH4.
burial of organic material and associated net increase in O2
atmospheric O2

We'll use the model to asses which of the two nitrogen-oxygen feedbacks are likely to have kicked in first and therefore been dominant in the delaying of the rise of oxygen in the 'great oxidation' - i) nitrification in the presence of oxygen leading to subsequent denitrification and loss of N (which depends on the threshold values of oxygen for the availability of nitrification and the inhibition of denitrification) and ii) the inhibition of nitrogen fixation by either the lack of reduced iron or the presence of oxygen (are these two directly linked in terms of Nfixation inhibition?).

Atmospheric lifetime of volatile selenium compounds

Wed, 01 Jun 2011 09:44:00 GMT

Atkinson, Roger et al. (1990) present experimentally-derived second order rate constants for dimethylselenide in the atmosphere for reactions with OH radical, O3 and NO3 radical:

reaction with	second order rate constant / cm3 molecules−1 s−1
OH	6.8 x10−11
NO3	1.4 x10−11
O3	6.8 x10−17

(1 OOM more rapid than the same reactions with DMS!)

Typical MBL concentrations of the reactants are as follows:

reactant	typical MBL concentration / molecules cm−3	ref/link
OH	1.4 x106	http://www.atmos-chem-phys.org/9/9225/2009/acp-9-9225-2009.html
NO3	1 x107	http://www.agu.org/pubs/crossref/2000/2000JD900314.shtml
O3	5 x1011	http://www.atmos-chem-phys.net/10/4611/2010/acp-10-4611-2010.html

We can calculate an e-folding lifetime for a second order bimolecular reaction:

Ax = 1k[B]

where A is in this case DMSe and [B] is the concentration of the reactant (OH, O3 or NO3)

This gives atmospheric lifetimes wrt each of the reactants as follows: OH = 2.9 hr; NO3 = 2.0 hr; O3 = 8.16 hr.

An overall lifetime can then be calculated as:

A = 11A1+1A2+1A3

which gives an overall night-time lifetime of 1.6 hours and daytime of 2.14 hours. Within the large uncertainties, an overall estimate of ~2 hours seems reasonable.

Atkinson, Roger, Aschmann, Sara M and Hasegawa, David (1990). Kinetics of the atmospherically important reactions of dimethyl selenide. Environ. Sci. Technol. 24, 1326–1332.

Latitudinally-resolved global ocean surface area, with sea-ice

Tue, 24 May 2011 13:18:00 GMT

Based on the spreadsheet here:
http://climateprediction.net/board/viewtopic.php?f=18&t=9312#p88228

which gives the percentage ocean cover in each 30 arc-second latitdinal band, based on the Globe 1.0 database.

The below scripts will calculate the ocean area in latitdunal bands of size x (in degrees). Sea-ice coverage from http://daac.ornl.gov/ISLSCP_II/guides/sea_ice_extent_xdeg.html for 1986-1995 is used to correct for minimum sea-ice coverage (for the purposes we need it for, this is the important term, seasonal sea-ice being considered effectively part of the open-water fraction of the ocean; in the future I'll modify this script to work with min, max or average sea-ice extent).

#R script to calculate the ocean area in latitude bands of a given extent

rad<-6371007 #Earth average radius in m (such that sphere area is equivalent to oblate spheroid area of actual planet

area<-read.csv("Ocean_area.data.csv",row.names=NULL)


#convert Lats to radians:
Lat_start<-area$Latitude.Start*pi/180
Lat_end<-area$Latitude.End*pi/180

lat_band_area<-function(Lat_start,Lat_end){
	#calculate total area of latitudinal band on planet
	2*pi*(rad^2)*(sin(Lat_start)-sin(Lat_end))
}

#calculate area of each tiny band
area$band_area<-lat_band_area(Lat_start,Lat_end)

#and ocean area
area$ocean_area<-area$band_area*area$Ocean.Fraction


#for bands of latitudinal extent in degrees
band_ocean_area<-function(start_lat, end_lat){
	area_subset<-subset(area,Latitude.Start<=max(start_lat, end_lat) & Latitude.End>=min(start_lat,end_lat))
	sum(area_subset$ocean_area)
}

#get areas at latitudinal spacing of x
get_lat_band_ocean_areas<-function(x){
	lat_s<-seq(from=-90,to=90,by=x)
	#strip the 90 off the end if it's there
	if(lat_s[length(lat_s)]==90){lat_s<-lat_s[1:(length(lat_s)-1)]}
	#now make the end lat array - starting at the 2nd member of lat_s	
	lat_e<-c(lat_s[2:length(lat_s)],90)
	x<-data.frame(lat_s,lat_e)
	x[,3]<-apply(x,1,function(row){band_ocean_area(row[1],row[2])})	
	colnames(x)<-c("Latitude.Start","Latitude.End","Ocean.Area")
	x		
}

#read and process sea-ice file to get minimum annual ice coverage per bands of latitudinal range, x
get_lat_band_min_ice_cover<-function(x,ice_file="sea_ice_data_1deg/sea_ice_data_1d_1995.csv"){
	data<-read.table(ice_file,sep=",",header=TRUE)
	data2<-subset(data, LAND_MASK==0)
	data2$minice<-apply(data2[,3:14],1,min)
	data2$ice_area<-apply(data2,1,function(row){lat_band_area((row[2]+0.5)*pi/180,(row[2]-0.5)*pi/180)*0.01*row[16]/360})
	lat_s<-seq(from=-90,to=90,by=x)
	#strip the 90 off the end if it's there
	if(lat_s[length(lat_s)]==90){lat_s<-lat_s[1:(length(lat_s)-1)]}
	#now make the end lat array - starting at the 2nd member of lat_s	
	lat_e<-c(lat_s[2:length(lat_s)],90)
	x<-data.frame(lat_s,lat_e)
	x[,3]<-apply(x,1,function(row){sum(subset(data2,LAT>row[1] & LAT<row[2])[,"ice_area"])})
	colnames(x)<-c("Latitude.Start","Latitude.End","Seaice.Area")
	x
}

yearlist=seq(from=1986,to=1995,by=1)
get_average_sea_ice_cover<-function(x,years=yearlist){
	lat_s<-seq(from=-90,to=90,by=x)
	#strip the 90 off the end if it's there
	if(lat_s[length(lat_s)]==90){lat_s<-lat_s[1:(length(lat_s)-1)]}
	#now make the end lat array - starting at the 2nd member of lat_s	
	lat_e<-c(lat_s[2:length(lat_s)],90)
	y<-data.frame(lat_s,lat_e)
	for(year in years){
		filename=paste("sea_ice_data_1deg/sea_ice_data_1d_",year,".csv",sep="")
		y<-cbind(y,get_lat_band_ice_cover(x,filename)[,"Seaice.Area"])	
	}
	y$mean_sea_ice_area<-apply(y[,3:length(yearlist+2)],1,mean)
	colnames(y)<-c("long","lat",yearlist,"mean_seaice_area")
	y
}

get_ocean_area_minus_seaice<-function(x){
	lat_s<-seq(from=-90,to=90,by=x)
	#strip the 90 off the end if it's there
	if(lat_s[length(lat_s)]==90){lat_s<-lat_s[1:(length(lat_s)-1)]}
	#now make the end lat array - starting at the 2nd member of lat_s	
	lat_e<-c(lat_s[2:length(lat_s)],90)
	y<-data.frame(lat_s,lat_e)
	ocean_area<-get_lat_band_ocean_areas(x)[,"Ocean.Area"]
	Ice_area<-get_average_sea_ice_cover(x)[,"mean_seaice_area"]
	y$Water.Area<-ocean_area-Ice_area
	y
}

And Ocean_area.data.csv: https://spreadsheets.google.com/spreadsheet/ccc?key=0AqRtHz1X9oQ6dEJhU2R2akdxME9vSGtEdUJ1UmRjRXc&hl=en_GB#gid=0

}}}

SmartBuoy sample log for incoming samples

Fri, 20 May 2011 14:29:00 GMT

https://spreadsheets.google.com/spreadsheet/ccc?key=0AqRtHz1X9oQ6dHhUMW1jUXF1N2VLX2ZVcVNPS3ZnY0E&hl=en_GB

What average concentrations... results

Mon, 16 May 2011 12:28:00 GMT

Let's assume we need to get between 1 and 10 Tg-C out of the ocean for each of the compounds. The figure below shows the results of back-calculation of the required seawater concentration for a range of compounds, required source strengths (s, in g-C per year) and gas phase concentrations.

click on image to view full size

Because of the low solubility of isoprene and the monoterpenes, the gas phase concentration barely affects the result - i.e. gas phase concentration is close to zero at 1000ppt relative to sw concentration (except for the much more soluble a-terpineol, which will not be discussed further). Sensitivity analysis (not shown) also demonstrates that the results are rather robust to values of global mean T, S and winspdeed (only varying by about +/- 5% over realistic ranges of mean values).

Ignoring gas phase concentration (fixing it at 100ppt), the table below shows concentration required for source strengths of 1, 5, and 10 Tg-C from each compound, and also observed marine concentrations, where available. All water phase concentrations in pM.

compound	s=1 Tg-C	s=5 Tg-C	s=10 Tg-C	Observed (REF)
Isoprene	60	290	580	12-94 (Matsunaga et al, 2002); 10-50 (Milne et al, 1995); 4-70 (Kurihara et al, 2010) 20 probably representative (Shaw, Gantt and Meskhidze, 2010)
alpha-Pinene	36	170	350	5-65000!! (Button and Juttner, 1989)

So, it looks like Isoprene might account for a flux of up to 1 Tg-C out of the ocean, and the jury's out on the monoterpenes as there is just the one study on seawater concentrations, which found average a-Pinene concentration of 65000pM in June of 1985 and 5 in June 1986. Seems a little peculiar!!

However if the monoterpenes are the source of the SOA, then a number of them (10ish) need to be at or above the concentration of isoprene to constitute a total organic C flux of roughly 10 Tg or more to the atmosphere. Seawater measurements needed! If alpha pinene is at anything like the average of the 1985 and 1986 coastal concentrations seen by Button et al, then 500Tg-C could be coming from the oceans in the form of a-Pinene alone.

plotting code for figure:

plot_it<-function(compound,source_strengths=c(1e12,5e12,10e12),gas_phase_concentration_range=c(1,10,100,1000)){

 x<-data.frame(gasconc<-gas_phase_concentration_range)
 leglist<-NULL
 for(s in source_strengths){
 	x[,paste("s=",as.character(s),sep="")]<-1e6*Cxsw(compound,x[,1],s,1)
	leglist<-c(leglist,paste("s=",as.character(s),sep=""))
 }
 ymax=100*(ceiling(max(x[-1])/100))
 y=seq(from=0,to=ymax,length.out=nrow(x))
 matplot(x[,1],y,type="n",xlab="gas phase concentration / ppt",ylab="water phase concentration / pM",main=compound)
 matlines(x[,1],x[2:length(x)],lty=1:(length(x)-1),col="black")
 legend("topleft",leglist,lty=seq(from=1,to=length(x)-1, by=1))
}

What average global surface seawater isoprene and monoterpene concentrations are required for a given flux out of the ocean?

Mon, 16 May 2011 10:05:00 GMT

'Top down' (from global transport and atmospheric chemistry models based on amount of secondary organic aerosol) and 'bottom up' (estimating marine biogenic production) approaches to closing the budget for the contribution of isoprene and the monoterpenes' contribution to SOA are miles apart. Taking modelled and measured atmospheric concentrations and a required flux per unit are from the top down approach, we will be able to estimate the seawater concentration required to drive the necessary flux. If this comes out at e.g. >uM levels we can be pretty sure that the budget won't close and we need to look to other compounds for SOA. If it's nM or less then the jury's still out and we need more investigation. A really simple bit of work but hopefully useful and informative.

So, given that we need to account for a given mass of organic compounds getting into the aerosol per year (i.e. a flux), F_x (where x is the compound of interest), and there are a reasonable number of observations of gas phase concentrations of monoterpenes and isoprene, C_{x_{(g)}}, we can extrapolate the required average seawater concentration from the above terms, the average temperature and salinity, solubility and transfer velocity like so:

C_{x_{(sw)}} = \frac{C_{x_{(g)}}}{K_H} - \frac{F}{K_w}

Where K_H and K_w are the dimensionless gas/liquid transfer velocity and total transfer velocity with respect to the water side of the interface, both of which are T and S dependent (and the latter wind dependent) and can be calculated from my numerical scheme. F needs to be negative for a flux out of the ocean

Therefore we can write a simple script to calculate this for a given compound and values of F_x, C_{x_{(g)}} and global average values of temperature, salinity and windspeed.

source("K_calcs_Johnson_OS.R")

#SOAMass<- 1e12 #annual mass of carbon in SOA from emitted compound
#SOA_conversion_factor<- 0.1 #amount of emitted gas which ends up as SOA (0-1)


mgws<-7.6 #mean global wind speed in m/s (second product moment - assuming squared K-u relationship)
mgT<-15 #mean global ocean surface temp
mgS<-35 #mean global surface salinity		

Cxsw<-function(compound,Cxg,SOAmass=1e12,SOA_conversion_factor=0.1){
	#input gas phase concentration in ppt, get seawater concentration in uM
	
	#convert gas phase concentration to mol/m3
	Cg<-(Cxg*1e-12*101325)/(8.314472*(mgT+273.15))

	Fx<- -SOAmass/SOA_conversion_factor #required annual flux of gas out of ocean
	print(paste("For oceanic isoprene flux of Fx",Fx,"g-C per year"))
	#convert flux to moles C per year
	F<-Fx/12

	#and to moles of compound per year
	F<-F/compounds[compound,"C"]

	#and to moles per unit area per second (ocean s.a. 3.6e14m2)
	F<-F/(3.6e14*3600*24*365.25)

	Csw_moles_per_metrecubed<-(Cg/KH(compound,mgT,mgS))-(F/Kw(compound,mgT,mgws,mgS))
 
	pM<-Csw_moles_per_metrecubed*1e9
	print(paste(pM,"pM"))

	#return value in umol/l
	Csw_moles_per_metrecubed*1e3
}

Default plot for Henry's law with T and S

Wed, 11 May 2011 12:45:00 GMT

Create a method for a T and S dependent Henry's law solubility plot for the transfer velocity scheme.

source("K_calcs_Johnson_OS.R")

TSSolPlot<-function(compound){
S_list<-c(0,17.5,32,33,34,35,36)
plot(T_LOTS,KH(compound,T_LOTS,max(S_list)),type="n",xlab="Temperature / Celcius", ylab="KH / dimensionless (gas/liquid)")
linetype<-1
legendlist<-NULL
for(S in S_list){
 lines(T_LOTS,KH(compound,T_LOTS,S),lty=linetype)
 linetype<-linetype+1
 legendlist<-c(legendlist,paste("S=",S,sep=""))
}

legend("topleft",legendlist,lty=seq(from=1,to=length(S_list),by=1))
}

TSSolPlot<-function(compound){
S_list<-c(0,17.5,32,33,34,35,36)
plot(T_LOTS,KH_Molar_per_atmosphere(compound,T_LOTS,min(S_list)),type="n",xlab="Temperature / Celcius", ylab="KH / dimensionless (gas/liquid)")
linetype<-1
legendlist<-NULL
for(S in S_list){
 lines(T_LOTS,KH_Molar_per_atmosphere(compound,T_LOTS,S),lty=linetype)
 linetype<-linetype+1
 legendlist<-c(legendlist,paste("S=",S,sep=""))
}

legend("topright",legendlist,lty=seq(from=1,to=length(S_list),by=1))
}

TSSchmidtPlot<-function(compound){
S_list<-c(0,17.5,32,33,34,35,36)
plot(T_LOTS,schmidt(compound,T_LOTS,max(S_list)),type="n",xlab="Temperature / Celcius", ylab="Schmidt number (dimensionless)")
linetype<-1
legendlist<-NULL
for(S in S_list){
 lines(T_LOTS,schmidt(compound,T_LOTS,S),lty=linetype)
 linetype<-linetype+1
 legendlist<-c(legendlist,paste("S=",S,sep=""))
}

legend("topleft",legendlist,lty=seq(from=1,to=length(S_list),by=1))
}

This code can also be found in the git repository

Henry's law solubilities and molar volume data for isoprene and monoterpenes

Sun, 24 Apr 2011 11:59:00 GMT

Following the SOFI marine trace gas meeting in Plymouth last week, it seems particularly important that isoprene and the monoterpenes should be included in the compounds.dat input data to the transfer velocity scheme. These weren't included originally as the temperature dependence of their solubilities is not available in Rolf Sander's compilation of Henry's law data.

For key monoterpenes, Copolovici, Lucian O. and Niinemets, "Ulo () have measured Henry's law constants at various temperatures. These need converting to the correct units for input to the scheme as follows:

parameter	Copolovici	compounds.dat	Conversion factor
Henry solubility	Hpc: Pa m3 mol−1	H =M atm−1	H = 101.325/Hpc
Temperature dependence	Hvol: KJ mol−1	-solnH/R: K	-solnH/R = Hvolx103/8.314472

For isoprene there is surprisingly little data. No temperature dependence of solubility appears anywhere in the literature as far as I can tell (after pretty much a whole day's worth of searching) and there is some disagreement in the value of the solubility under standard conditions between different studies. The commonly measured value is 1.3x10-2 M/atm but the most recent study cited in Rolf Sander's Henry's law compilation is 2.8x10-2 so investigation of the latter as a possible upper bound is probably useful. I assume the temperature dependence is similar to that of 1,3-butadiene (isoprene being 2-methyl-1,3-butadiene), the value of which is given as 4500 K in Sander.

new lines for compounds.dat:

	nicename	mw	KH	tVar	C	H	N	O	S	Br	Cl	F	I	Se 	db	tb	rings	Vb
isoprene	Isoprene	68.12	1.3e-02	4500	5	8	0	0	0	0	0	0	0	0	2	0	0	0
isoprene_upper	Isoprene*	68.12	2.8e-02	4500	5	8	0	0	0	0	0	0	0	0	2	0	0	0
aPinene	alpha,-Pinene	136.24	7.47e-03	4440	10	16	0	0	0	0	0	0	0	0	1	0	1	0
bPinene	beta,-Pinene	136.24	1.47e-02	4450	10	16	0	0	0	0	0	0	0	0	1	0	1	0
aPhellandrene	alpha,-Phellandrene	136.24	1.82e-02	4450	10	16	0	0	0	0	0	0	0	0	2	0	1	0
bPhellandrene	beta,-Phellandrene	136.24	1.82e-02	5090	10	16	0	0	0	0	0	0	0	0	2	0	1	0
aTerpinene	alpha,-Terpinene	136.24	2.90e-02	4790	10	16	0	0	0	0	0	0	0	0	2	0	1	0
yTerpinene	gamma,-Terpinene	136.24	3.87e-02	4775	10	16	0	0	0	0	0	0	0	0	2	0	1	0
aTerpinolene	alpha,-Terpinonlene	136.24	3.81e-02	5450	10	16	0	0	0	0	0	0	0	0	2	0	1	0
aTerpinolene	alpha,-Terpinonlene	136.24	3.81e-02	5450	10	16	0	0	0	0	0	0	0	0	2	0	1	0
Limonene	Limonene	136.24	3.78e-02	4510	10	16	0	0	0	0	0	0	0	0	2	0	1	0
aTerpineol	alpha,-Terpineol	154.25	448	2165	10	18	0	1	0	0	0	0	0	0	1	0	1	0

Atkinson, Roger, Aschmann, Sara M and Hasegawa, David (1990). Kinetics of the atmospherically important reactions of dimethyl selenide. Environ. Sci. Technol. 24, 1326–1332.

Copolovici, Lucian O. and Niinemets, "Ulo (). Temperature dependencies of Henry's law constants and octanol/water partition coefficients for key plant volatile monoterpenoids. Chemosphere 61, 1390—1400.

Conference presentations / posters

Sun, 24 Apr 2011 10:20:00 GMT

Until Mendeley gets a nice entry for this stuff in the profile info, I'm going to start accumulating these here.

M.T. Johnson, P.S. Liss, T.G. Bell, C. Hughes and J. Woeltjen, Transfer velocities for a suite of trace gases of emerging biogeochemical importance: Liss and Slater (1974) revisited, at 6th international symposium on gas transfer at water surfaces, Kyoto, May 17th - 21st 2010.
Hare, J.H, D. Jackson, M.T. Johnson, T.G. Bell and G. Wick, Development of global air-sea gas transfer products at 17th AMS conference on air-sea interaction, Annapolis, MD, USA, September 2010.
M.T. Johnson, Constraining uncertainty in estimates of ocean-atmosphere trace gas fluxes,at SOFI Marine Trace Gas Workshop, Plymouth, 18th-19th April 2011

SMARTBuoyTNanalysis.220311

Thu, 24 Mar 2011 12:20:00 GMT

Sample volume for standard additions: 5ml. Standard: TNTCMixedstock1. Std add. volumes: 0, 15,30,45,60,75,150 uL.

CRMs: LCW10.09, DSR05.10

Tracer: 5ml LCW05.04 + 300uL TN standard

Samples run: Dowsing 10(EDU sampler), 11(2)

raw data

NH4_240311

Thu, 24 Mar 2011 12:16:00 GMT

Ammonium analysis of Dowsing deployments 11(2) and re-runs of some questionable looking data from NH4_140311.

Standard additions of NH4ClStd50u.001 to 1ml of sample, using Hamilton syringes - in an attempt to reduce volume requirement.

data

SMARTBuoyTNanalysis.150311

Tue, 15 Mar 2011 13:41:00 GMT

SMARTBuoyTNanalysis.140311

Tue, 15 Mar 2011 13:16:00 GMT

NH4_140311

Tue, 15 Mar 2011 13:14:00 GMT

Ammonium analysis of Dowsing deployments other than Dowsing 11 (see NH4_100311).

Also comparison of Test 50uM stock NH4Cl and NH4ClStd50u.001

Standard additions to 5ml of sample

data