Field: Five resin bags (~8mL resin/bag), except when it is specified, were deployed at each location, and the bags were left in the field for ~3 weeks each time. Bags were deployed at ~10cm deep soil, except for snowmelt bags which were placed at the bottom of snow pack in early May then pushed into thawed moss layer as the soil melted. All bags were kept at ~5-degrees C until extraction. Lab: Five bags were pooled, then 2N KCl (500 mL for NO3 and 400mL for NH4) was used to completely replace NO3 or NH4 accumulated on the resins. NO3 and NH4 concentrations were determined (see below) then d15N values were determined using LINX2 diffusion method (see below). NO3: Cd reduction followed by Griess-Ilosvay method for NO2 NH4: Berthelot method d15N for NO3: 1. Measure out the sample volume to be boiled down and add to a clean, acid-washed beaker or flask. 2. Add about 5g of ashed NaCl (for ionic balance, want final concentration after boil down to be about 50 g/L) and about 3 g ashed MgO (to raise pH) to sample (add proportionately more NaCl if boiling down smaller volumes of sample to achieve a 50 g/L concentration after boil-down, add same amount of MgO regardless of sample size to boil down). 3. Add stir bar and place on hot plate with stirring capability. Heat until volume is reduced to roughly 100 ml (doesn’t have to be exact, 75-125 mLs should be okay). Aim for a simmering boil, but if just below this then that is fine as well. Stir during the boil. It takes about 3-4 hours of gentle boiling to reduce 1 L to about 100 mL. 4. After cooling somewhat, pour the boiled down sample into an acid-washed 250 mL HDPE bottle (suggest rigid rectangular bottles). These can be stored in refrigerator until you have an entire series (stream) ready for next step (reduction of nitrate to NH3 and diffusion of NH3 into headspace). 5. Add about 0.5 g of ashed MgO and then 0.5 g of Devardas alloy to boiled-down sample in 250 mL bottle. 6. Immediately after adding Devardas alloy, place filter pack (see subsection on steps for constructing filter packs below) carefully on surface of water, and cap very tightly. 7. Place bottles in oven at 60oC for 48 hours (be careful not to pack too closely in a tray since the bottles will swell slightly). 8. Remove from oven and place bottles on shaker and shake gently for 7 days at room temperature. 9. Open bottles and remove filter pack. Blot water droplets from filter pack and place in labeled scintillation vial and into dessicator. Also place an open vial of 2.5 M KHSO4 (to absorb any ammonium in air) in dessicator. 10. Let filters dry in dessicator for 4 days or so. Remove and cap the scintilation vials containing filter packs very tightly (use parafilm if necessary) and store until ready to encapsulate in tins. 11. Encapsulating filters: Remove filter pack from scintillation vial on clean surface (use alcohol to clean). Using cleaned forceps, open filter pack and remove small glass fiber filter. Place filter in tin capsule and fold opening of tin down once and compress. Crimp sides of tin to form small packet (all dimensions < 2 mm). Place tin packet into a well in the well tray recording the well location and sample ID (station, time). Place well cap strip over wells as soon as possible to minimize any further exposure of encapsulated filter to air. It is recommended that you use a separate well plate for the samples that will be only minimally enriched if at all (PRE, POST, and deionized water samples) and the enriched samples (collected during the 15N addition). Reference: Sigman, D. M., M. A. Altabet, R. Michener, D. C. McCorkle, B. Fry, and R. M. Holmes. 1997. Natural abundance-level measurement of nitrogen isotopic composition of oceanic nitrate: an adaptation of the ammonia diffusion method. Marine Chemistry 57:227-242. (Note that the methods outlined here is developed for LINX II project and based on the Sigman et al. paper, but are not exactly the same. In particular, the LINX II protocol does not go through extra incubation steps to reduce blank effects caused by breakdown of DON. Instead, DON blank effects are minimized by running diffusions at a lower incubation temperature than is used in the Sigman et al. paper.) d15N for NH4: 1. Measure 100mL of sample to an acid-washed 250 mL HDPE bottle. 2. Add about 0.5 g of ashed MgO and filter packet, cap tightly, then incubate at 40oC for 7 days. Follow NO3 method (#9).