upload new sampling data: ethass.au, ethamc.au 2014, 2015, 2016, 2017, 2018, 2019 (Years 2, 3, 4, 5, 6, 7)

[nutnetadmin]

Received from Glenda Wardle 27-July-2019

Dear Ashley, cc Elizabeth

Many thanks for this prompt to send along the data.

Good news. The data for both sites and for each of the years 2014, 2015, 2016, 2017, 2018 and 2019 are attached. Ethabuka (South Site) Ethabuka (Main Camp)

Still to complete: — the biomass for 2019. [Person is travelling and had trouble uploading the spreadsheet before they left] — Also bits of weather data for 2018-2019. This has to wait until our onsite weather station data are processed (later this year)

My sincere apologies for this lengthy delay.

Yes we have definitely been plugging away and keeping up with the field side of the experiment, but have struggled to keep up with the next stages. Some of the delay was a confusion from lab personnel about waiting for identification of unknown plant species, or following up on data entry.

Over the years we have not been able to secure any funds to support the onsite work of the two sites and so have to find time when we can. Unfortunately the ARC funding in Australia is very tight and we experienced prolonged period of low resourcing which meant very little technical support across all our projects.

It would be so good to attend the NutNet workshops, but they occur when I have a heavy teaching load. I leave for a field course tomorrow(today), and then teach everyday all through August.

I do hope to have a period of study leave in first half of 2020 (yet to be approved), and if so will be in touch to see if a visit would work out from your end.

kind regards, Glenda

[nutnetadmin]

Duplicate cover values for TRIODIA BASEDOWII arising because they record both live and dead Triodia cover, but sometimes forgot to put in the "dead" designation. Going in by hand into the .csv file and correcting these for ethass.au 2015: plots 17, 19, 21, 29, 8 (assuming 2nd observation is dead, as this is pattern for previous plots) 2016: plot 3 (assuming 2nd observation is dead, as this is pattern for previous plots)

2018: plot 29 Looks like this observation actually belonged in plot 30 & looks like duplicate cover value for Aristide contort (plot 5) belonged in plot 6. deleted duplicate 1% bare ground cover value for plot 10

[nutnetadmin]

Note that sometimes biomass will be NULL at this site, but actually biomass was 0

[nutnetadmin]

Ethabuka South complete with ethass-2014-2015-2016-2017-2018-2019.sql

[nutnetadmin]

Ethabuka Main Camp stuck because of many duplicate values of cover for Aristida contorta in 2018.Sent an email to Glenda Wardle, got an out-of-office reply.

Hi Glenda,

This is all so wonderful. Thank you for sending it! I think Eric and I will be able to use it in our papers.

I’ve just uploaded the data from the South Camp and am in the middle of uploading data from the Main Camp.

Can you check the 2018 cover data for me? Many plots (3, 7, 8, 9, 10, 12, 13, 21, 22, 24…) have two different cover values for the species Aristida_contorta. I’m assuming one of the two in each case should be “dead” Aristida contorta, but I’m not sure which is which.

Thanks so much! A

[nutnetadmin]

Went ahead and uploaded all the species/species information, and all the biomass and par data. Just need to upload the cover.

Mostly complete in import-ethamc-2014-2015-2016-2017-2018-2019.sql

[nutnetadmin]

Fixed cover data here: nutnet-data-Ethabuka-AUS_2018_submit_2019m8d5.xlsx

[nutnetadmin]

Hi Ashley,

Thanks for catching that slip up.

Yes you were correct, the two entries for Aristida contorta are because we had a value for alive and dead. I have added a new taxa Aristida_contorta_DEAD to the data and highlighted where the changes are in a new copy of the Cover spreadsheet.

So happy to finally have the data to you.

Kind regards, Glenda

[nutnetadmin]

PAW finished ethamc.au cover. There were a lot of mismatches between site reported lifespans and database lifespans that I need to check on.

[nutnetadmin]

Email exchange between Glenda Wardle and Jon Bakker revealed some biomass and cover changes that need to be attended to:

1. Biomass
   • the early data (2013, 2014) differ much more strongly between these two sites than the later data. For example, in 2013 there is a single live mass value at ethamc.aucompared to 19 at ethass.au.

1) BIOMASS - EARLY YEAR DIFFERENCES There is a clear reason for the stronger between-site differences for the early year data - FIRE.

I set up a burnt vs unburnt comparison between these two sites within our study system. See attached images taken on day of set up.
ethamc site had been burnt in a wildfire just prior to the set up of the experiment while the ethass site had been unburnt for >20 years.
There was near to zero live cover or biomass at ethamc at the start and the idea was to track responses to nutrient addition when above ground plant cover was none compared to typical cover and composition. 

While this may not always suit nutnet study designs, wildfire is a key driver of the dynamics of these hummock grasslands. Fire return intervals are typically 26 years. 
A couple of our papers on fire in the study system.

Verhoeven E. M., Murray B.R., Dickman C.R., Wardle G.M., and Greenville A.C. (2020) Fire and rain are one: extreme rainfall events predict wildfire extent in an arid grassland. International Journal of Wildland Fire https://doi.org/10.1071/WF19087
Greenville A.C., Dickman C.R., Wardle G.M., Letnic, M. (2009) The fire history of an arid grassland: the influence of antecedent rainfall and ENSO. International Journal of Wildland Fire 18: 631-639.    

I also wanted to verify that all biomass values were in g/m^2, as they are so much lower in the early years at ethamc.au than at that site in later years or at ethass.au in the same years.

BIOMASS - units The units are grams per SAMPLE. Values submitted will need to be multiplied by 5 for per m^2.
I dug back from the raw field data sheets through the process the lab technician was using to record the adjustments for bag weights were made, and it does not appear that the conversion from sample area (20cm x 100cm) to m^2 happened. 
My apologies for this. I had asked for all conversions so this is a big surprise and I am so glad you raised the query. We were at least consistently doing the wrong thing.
I will follow up with Nutnet HQ so that the biomass data are updated to the correct units for all years.

-no biomass measurements in 2017?

The biomass data for 2017 were not available at the time of submitting. We definitely took samples and have the data. Will get back to you with this.

2. Cover -In a few years (2013, 2018, 2019), there are taxa that are coded as dead. Do these reflect plants that grew in this growing season? In your opinion, should these be excluded from compositional analyses? The taxa I'm referring to are: -OTHER ARISTIDA CONTORTA (DEAD): in 2013 at both sites and in 2018 at ethamc.au -OTHER SALSOLA KALI (DEAD): only in 2013 at ethass.au -OTHER TRIODIA BASEDOWII (DEAD): in 2013 and 2019 at both sites, and in 2015-2018 at ethass.au

For ARISTIDA CONTORTA or ARISTIDA HOLATHERA the dead biomass will be plants that grew and died within the year.

These are short-lived grasses and often they have hayed off by the time of harvest so would be in the compositional analyses.

SALSOLA KALI (DEAD) - These plants are NOT current year growth.

TRIODIA BASEDOWII (DEAD) - See attached photo panel of Tridia basedowii.

These plants are NOT current year growth but can regrow in future years.
This is a perennial grass with an unusual growth form that can live for decades and form rings. 
The plant grows out from the centre. It starts as a small mound that grows to a large mound around 1m in height and 1 m in diameter, after that the plant dies back from the middle  flattens into a circular form with an area of persistent dead stems flattened in the centre, that then collapses further and the centre becomes bare sand as the ring grows outwards.

I've attached a summary of the cover data for each site. Each summary is a tally where rows are taxa, columns are years, and the data are the number of values for that taxon in that year. Rows are sorted alphabetically by species within family. Would you mind reviewing these for taxonomic issues? In addition to the above question about dead taxa, there are two other adjustments I wanted to ask about:

-At both sites, should Aristida holathera and Aristida contorta be combined into a single taxon? A. contorta is only reported in 2018 at ethamc.au (and A. holathera is not reported in that year at that site). At ethass.au, A. contorta is only reported in 2013 and 2018 (and A. holathera is not reported in either year), but both species are reported in 2019.

See Excel workbook attached with Taxa list submitted with updated notes.
Yes A. contorta and A. holathera should be combined into a single taxon for nutnet. They are difficult to identify in the field without all reproductive features. 
Earlier years we called it all these plants A. contorta but after another study started, we now use both taxa. 
My own impression is that this is still a bit of a mess for field identification, and for ecological studies it is more robust to treat as a single taxon.

-At ethass.au, should Portulaca intraterranea and Portulaca sp. be combined into a single taxon? Yes do combine these into a single taxon.

[nutnetadmin]

My proposed database changes:

1) Scale biomass values up to g m-2 by multiplying by five for both sites 2) Enter 2017 data for both sites 3) Combine Aristada contorta (dead), A. contorta (live), and A. holathera into a single taxon (A. contorta) in the cover data for both sites in all years 4) Combine Portulaca sp. and P. intraterrana into a single taxon (P. intraterrana)

[nutnetadmin]

2017 data in 'to import -> 2017' folder on Google Drive.

nutnet-data-Ethabuka-AUS_2017_biomass updated

[nutnetadmin]

All changes are made in script 'correct-ethabuka-biomass-cover-issues.sql'

High biomass value in plot 13 of ethass.au. Waiting to hear back from Glenda Wardle before closing this issue.