Readme.txt written by: Catherine Wiltshire
Written on [2020-03-21]

This readme is related to the following data files:

Loch Davan n-alkane compound-specific d13C data.csv
Loch Davan n-alkane concentration data.csv
Loch Davan nlfa compound-specific d13C data.csv
Loch Davan nlfa concentration data.csv
Loch Davan percentage carbon and nitrogen data.csv

===== Variables ===== 
All data files: 
Land Use (arable, forest, moorland, pasture, bed sediment, suspended sediment)
Sample ID

Loch Davan n-alkane compound-specific d13C data.csv:
n-alkanes δ13C (C23-C31)(‰ [per mille])

Loch Davan n-alkane concentration data.csv:
n-alkanes concentration (C23-C31) (nano moles per g soil)

Loch Davan nlfa compound-specific d13C data.csv
Neutral Lipid Fatty Acid (NLFA) δ13C (C23-C31)(‰ [per mille])

Loch Davan nlfa concentration data.csv
NLFA concentration (C23-C31) (nano moles ester per g soil)

Loch Davan percentage carbon and nitrogen data.csv
C (%, w/w)
d13C (δ13C) (‰ [per mille])
N (%, w/w)
d15N (δ15N) (‰ [per mille])

Code: N/A
Definition: No data available

===== Sample collection =====

Soil samples (arable, forest, moorland, pasture):
Date/date range of data collection: 2019-06-18 to 2019-06-19
Geographic location of data collection, Loch Davan catchment, Aberdeenshire, Scotland

Details:
Replicate soil samples were taken to characterise each of the four land uses arable (n=16), forest (n=16), moorland (n=18) and pasture (n=19). Sampling sites were chosen on the basis of likely hydrological connectivity and were stratified by land use and soil type. For each sampling point, three replicates were chosen at random within a 2 m radius. Each sample was taken with a steel cylinder (6cm depth and 6cm diameter) and litter was removed before taking the sample. All samples were georeferenced by using a GPS device (horizontal accuracy sub-meter real-time), stored in plastic bags and freeze-dried on return to the laboratory. The samples were then passed through a 2 mm sieve to remove stones and larger organic material before being ground. A composite sample was formed for each site by adding an equal weight of each of the three finely ground samples. Samples were stored in sealed containers at room temperature until required for analysis. 	

Streambed samples:
Date/date range of data collection: 2019-06-18 to 2019-06-19
Geographic location of data collection, Loch Davan catchment, Aberdeenshire, Scotland

Details:
Bed sediment samples were taken at 3 locations (Figure 1b), representing two tributaries and a joint outlet. The locations were carefully chosen above their joint junction in the stream network so the contributions from each tributary could be assessed. Logie Burn originates in two main headwaters (Figure 1) with the northern most branch (BS1) supporting similar cover of pasture (30%), forest (29%) and moorland (28%) with around 10% arable land. The western branch (BS2) predominantly passes though moorland (78%) with around 14% of the land use being pasture, less than 5% forest and no arable land. A third site (BS3) was located close to the outlet of Logie Burn to Loch Davan integrating input from the whole catchment. At each site three samples of bed sediments were taken with a steel cylinder (6 cm depth and 6 cm diameter) along a transect across the streambed and composited. All samples were georeferenced using GPS, stored in plastic bags and freeze-dried on return to the laboratory. The samples were then passed through a 2 mm sieve to remove stones and larger organic material before being ground and stored at room temperature until further analysis.

Suspended sediment samples:
Date/date range of data collection: 2019-06-19 to 2020-11-17
Samples collected on the following dates:
2019-08-27, 2019-10-23, 2019-12-17, 2020-02-11, 2020-04-14, 2020-06-04, 2020-07-29, 2020-09-22, 2020-11-17
Geographic location of data collection, Loch Davan catchment, Aberdeenshire, Scotland

Details:
Suspended sediments (SS) were collected bi-monthly at 3 locations over a period of eighteen months (June 2019 – Dec 2020). Time integrated mass samplers (Phillips et al. 2000) were used to collect SS at all sampling locations. Unfortunately, due to high flows at Site 1, the sampler was lost twice, and therefore no SS were collected between August 2019 and January 2020. Suspended sediment from the traps was collected and placed in clean food grade plastic buckets, left to settle for 5 days before the supernatant was removed and the remaining water left to evaporate at room temperature for up to 6 weeks. The dried sediment was then decanted and freeze dried. Samples collected after March 2020 were not freeze dried due to Covid-19 lockdown restrictions and were only air dried. Samples were weighed and passed through a 2 mm sieve before being ground and stored at room temperature until required for analysis.

===== Sample Analysis =====

N-alkane extraction:
To isolate the hydrocarbon fraction of the samples for analysis, total lipid extraction was followed by lipid fractionation (Dove and Mayes 2006). For quality and quantification control purposes, 50 µl of alkane standard solution (docosane (C22) and tetratriacontane (C34) in decane) was added to the samples prior to extraction. First 3ml of 1M Ethanolic KOH solution was added to each sample in a tube before they were capped and heated for 16 hours at 90°C in a dry-block heater. The following steps were then repeated twice: 3ml N-heptane was added to each tube which were capped and swirled before 1ml of deionised water was added and the tubes re-capped and shaken vigorously; after separation into two liquid layers, the top (non-aqueous) layer was transferred to a new glass tube. The resulting solution was evaporated to dryness on a dry-block heater fitted with a sample concentrator blowing nitrogen (N2) into the tube. The resultant was re-dissolved in 0.3ml heptane with warming before transferring the sample to SPE-Si cartridge, adding 2ml heptane and collecting the elution in a 1.5mL autosampler vial. Solution in the vial was then evaporated to dryness.

Neutral lipid fatty acid (NLFA) extraction
The samples (5 – 10 g) were analysed by lipid extraction with a single phase chloroform mixture before fractionation on a SPE Si column and mild methanolysis. The solvents chloroform and methanol (1:2) were used in lipid extraction and, in addition, 0.15 M of citrate buffer (0.8:1:2 of citrate buffer: chloroform: methanol (Bligh and Dyer (B&D) solvent ratio)).  15 - 20 ml B&D solvent was added to the sample in a glass media bottle (closed using PTFE lined cap) which was then sonicated for 30 minutes (in ultrasonic bath). This was followed by centrifuge at 700 RCF for 10 minutes before the upper layer was poured into a clean glass media bottle. Chloroform (4ml) and citrate buffer (4ml) were added before centrifuging at 700 RCF for 10 minutes. Successful separation was indicated by both layers appearing clear and, using a vacuum pump, the upper (aqueous) layer was removed and discarded leaving the bottom organic layer. The glass bottle was then placed in an evaporator at 37°C and dried under N2.The neutral lipids were then separated from the lipid extract by fractionation (Solid Phase Extraction). Commercially prepared SPE columns were loaded onto vacuum manifold and prepared with ~0.5 g sodium sulphate (Na2SO4) and chloroform before the lipid extracts were added along with 1ml chloroform. 5ml chloroform was used to elute the neutral lipids (sterols). These were collected in a clean glass bottle placed in an evaporator at 37°C and dried under N2 for 4-5 hours. To quantify the FAMES against an internal standard, 60 µL of C19:0 methyl ester (Methyl nonadecanoate in methanol: 25mg/L) was added to the dried phospholipid fraction after SPE and evaporated to dryness under N2.
The samples were then methylated by adding 1ml of toluene:methanol (1 : 1) (stored on Na2SO4) and 1ml of 0.56g potassium hydroxide in 50 ml methanol to the neutral lipid fraction (NLF). This was swirled and incubated @37°C for 30 minutes. Subsequently, 0.25ml of acetic acid 59 ml L-1, 5ml of hexane:chloroform (4 : 1)(v/v) and 3ml deionised water were added and the NLF glass bottle was centrifuged at 700 RCF for 10 minutes. The upper organic phase was collected in a Gilson pipettor and the lower aqueous phase discarded. The resultant liquid was dispensed through 10ml pipette tip packed with glass-wool and Na2SO4 into clean glass bottle and rinsed with a few ml of hexane. The glass bottle was placed in an evaporator at 20-25°C and dried under N2. The dried FAMEs were then stored in a freezer at -20°C

Analysis of n-alkanes and FAMES by GC-C-IRMS 
Individual n-alkane and FAMEs were quantified and their δ13C values determined by GC-C-IRMS using a Trace GC Ultra gas chromatograph (Thermo Finnigan, Bremen, Germany) equipped with a GC PAL autosampler (CTC Analytics AG, Zwingen, Switzerland) following the method described in Thornton et al., (2011). 

Total C, total N, δ13C and δ15N:
All suspended sediment and terrestrial soil samples were analysed for carbon and nitrogen concentrations (%, w/w) using a Flash EA 1112 Series Elemental Analyser connected via a Conflo III to a DeltaPlus XP isotope ratio mass spectrometer (all Thermo Finnigan, Bremen, Germany). USGS40 was used as a reference material for both C and N concentrations, measured using the area output of the mass spectrometer. Long term precisions for a quality control standard (dried milled topsoil) were: total C 3.80 ± 0.15 % and total N 0.28 ± 0.02 % (mean ± SD). Data processing was performed using Isodat 2.0 (Thermo Fisher Scientific, Bremen, Germany).