N water chemistry from small Arctic streams in relation to vegetation cover


In the Arctic, little information is available, especially in terms of N availability and composition (i.e., nitrate, ammonium, and dissolved organic nitrogen) from small, flowing waters. This data set aims to quantify N concentrations across small Arctic streams and explore the link between terrestrial vegetation and stream water N concentration. The data set is the result of a literature study where data on N water chemistry was collected and combined from peer-reviewed, published articles and data sets selected by specific criteria. 20 articles met the selected criteria along with four datasets from databases resulting in a total of 2381 observations on N water chemistry from Arctic flowing waters from 1996 to 2021. Bioclimate subzones, NDVI and phytomass describe vegetation. Data on dissolved organic nitrogen (DON) was scarce: only 161 of the 2381 observations contained DON data. We found that nitrate (NO3-), ammonium (NH4+) and DON ranged undetectable to 1155, 547 and 1587 µg N/l, respectively. We found that sparsely vegetated areas had higher stream water N-concentrations, while barren areas and higher vegetated areas had lower stream water N-concentrations.

Data was collected through a systematic search on the Web of Science (WOS) search engine to find relevant, scientific peer-reviewed articles on N water chemistry from Arctic flowing waters. A "topic" search performed on 4 November 2021 yielded 175 articles: (((stream OR river OR watercourse OR tributary OR creek OR brook) AND (Arctic OR Greenland OR Svalbard OR Siberia) AND (nitrate OR ammonium OR nitrogen) NOT (sea OR ocean OR lake OR pond))). The articles from this search were combined with data obtained from The Greenland Ecosystem Monitoring (GEM) database (Greenland Ecosystem Monitoring, 2020b; doi:10.1016/j.rse.2006.03.011, 2020a; doi:10.17897/ZJK8-0B45 ), the National Science Foundation's Long Term Ecological Research (LTER) database (Bowden, 2021),and published articles that were not initially found through the main keyword search. This resulted in 215 articles in total. This was also combined with 21 unpublished data points of our group.After retrieval, all articles and data were systematically scanned for data adhering to the following criteria: (a) data on N water concentration (NO3-, NH4+, DON and/or TDN) for a particular year from Arctic flowing waters; and (b) not to include streams that was impacted by human nutrient enrichment. The initial search did not yield any data regarding particulate N and, therefore, we did not assess this N type. In this study, the Arctic is defined as everything above the treeline (i.e. the northernmost latitude at which trees can grow and sustain themselves (CAVM, 2003). Data from figures were extracted using the WebPlotDigitizer. In order to also include sites with solute concentrations below detection limit we dealt with these low values in the following way: if the detection limit was specified in the study, the values below the detection limit were substituted for the limit of detection divided by 2. If there was no information on the limit of detection (i.e., if only the abbreviation was presented), we excluded the values from the analyses (See dataset in Supporting Information.)Bioclimate subzonesWe divided the Arctic into bioclimate subzones A-E, which is based on vegetation height and July temperature (Walker, 2005, doi:10.1111/j.1654-1103.2005.tb02365.x). The bioclimate subzone GeoTiff dataset was downloaded from the Circumpolar Arctic Vegetation Mapping Project (https://www.geobotany.uaf.edu/cavm/data/index.html) created by the Alaska Geobotany Center (Fig. 1A). Within this classification, temperature and vegetation height increase from north to south corresponding to Subzone A to Subzone E (Walker, 2005, doi:10.1111/j.1654-1103.2005.tb02365.x).Out of the 215 published articles, 20 articles met our criteria along with four datasets from databases (Holmboe et al., 2024). Combined with the 21 unpublished data points from our group, we obtained 2381 observations on N water chemistry from' Arctic flowing waters from 1996 to 2021 (see Excel sheet in Supporting Information). The data included water samples collected during summer, primarily July and August. We tested for temporal trends that could have been caused by climate change since 1996 by dividing the data into the first and second half sample period (i.e.1996-2008 (early) and 2009-2021 (late). However, in an initial test of patterns found in the early period versus the patterns found in late period we did not find any noticeable differences (Fig. S2 & S3 in Supporting Information). We therefore did the analyses with all years lumped together within the single period 1996-2021.NDVI and PhytomassTo assess the effect of vegetation more directly, we used two other vegetation parameters; Normalized difference vegetation index (NDVI) and Phytomass (Raynolds et al., 2006; doi:10.1016/j.rse.2006.02.016). Both GeoTiff datasets come from the Circumpolar Arctic Vegetation Mapping Project (https://www.geobotany.uaf.edu/cavm/data/index.html) created by the Alaska Geobotany Center, and information about the data can be found in Raynolds et al. (2006; doi:10.1016/j.rse.2006.02.016). The NDVI measures the relative greenness of an area and is derived from satellite images containing information about the spectral reflectance in the near-infrared (NIR) and red (R) spectral area that can be extracted and used to calculate NDVI. NDVI can be calculated as: NDVI = (NIR - R) / (NIR + R). Values vary from -1 to +1, where values closer to 1 indicate more vegetated areas. We focussed on the maximum NDVI in an area, divided into six categories (Raynolds et al., 2006): NDVI [< 0.03], [0.03 - 0.14], [0.15 - 0.26], [0.27 - 0.38], [0.39 - 0.50] and [0.51 - 0.56]. In general, NDVI has been shown to correlate with the bioclimate subzones and decreases from south to north (Raynolds et al., 2006). Phytomass is the above-ground plant biomass measured in g/m² and is calculated from a regression relationship between NDVI and field measured phytomass derived from clip harvest data collected on the North Slope of Alaska, as described in detail in Raynolds et al. (2006) and Walker et al. (2003). Phytomass is divided into six categories; [ 0.03 and 7 corresponds to NDVI 0.51-0.56. For Phytomass, 1 corresponds to < 30 g/m2 and 6 corresponds to 850-1300 g/m².Interpreting the vegetation parametersTo be able to interpret the vegetation parameters more precisely, we have divided the sites into three categories according to the following criteria; "Barren" (Bioclimate subzones A and B, NDVI < 0.14, phytomass < 70 g/m²), "Sparsely vegetated" (Bioclimate subzones C, D and E, 0.15 < NDVI < 0.5, 70 g/m² < phytomass < 850 g/m²) and "Moderately vegetated" (Bioclimate Subzone E, 0.5 < NDVI < 0.56, 850 g/m² < phytomass < 1300 g/m²) (Julien et al., 2006 (doi:10.1016/j.rse.2006.03.011); Raynolds et al., 2006 (doi:10.1016/j.rse.2006.02.016); Sobrino & Raissouni, 2000 (doi:10.1080/014311600210876)). The data in this study was not located in areas with dense vegetation, where NDVI reaches values above 0.6 (Corrales et al., 2018, doi:10.1007/978-3-030-04447-3_7).Climatic variablesSeveral climatic variables were used to support the vegetation parameters and test the relationship with N concentrations in streams. Annual mean air temperature (°C, Air temp.), Annual precipitation (mm, Precipitation), Maximum temperature of Warmest Month (°C, Max air temp.) and Minimum temperature of Coldest Month (°C, Min. air temp.) were extracted from the WorldClim database of Bioclimatic variables (version 2, 2020) in 1 km² resolution (Fick & Hijmans, 2017, doi:10.1002/joc.5086). These values are an average for the years 1970-2000. Annual mean soil temperature (°C, Soil temp.) data were extracted from the Global Soil Bioclimate variables based on data from 1979-2013 in two depth intervals: 0-5 cm and 5-15 cm (Lembrechts et al., 2021; doi:10.1111/gcb.16060). The soil organic carbon content (SOCC, kg C/m² at 30 and 100 cm depth was extracted from the Northern Circumpolar Soil Carbon Database version 2 (NCSCDv2) (https://bolin.su.se/data/ncscd/).

DOI https://doi.org/10.1594/PANGAEA.965140
Related Identifier IsSupplementTo https://doi.org/10.1029/2023GB007840
Related Identifier IsDerivedFrom https://doi.org/10.17897/ZJK8-0B45
Related Identifier IsDerivedFrom https://doi.org/10.17897/1GTF-SX86
Related Identifier IsDerivedFrom https://doi.org/10.1007/s10533-012-9761-9
Related Identifier IsDerivedFrom https://doi.org/10.1111/j.1365-2427.2004.01319.x
Related Identifier IsDerivedFrom https://doi.org/10.1002/rra.2706
Related Identifier IsDerivedFrom https://doi.org/10.1029/2007JG000470
Related Identifier IsDerivedFrom https://doi.org/10.6073/PASTA/3FAACD18B63B3BACC5A0DBD6F09660E1
Related Identifier IsDerivedFrom https://doi.org/10.1080/17518369.2018.1440107
Related Identifier IsDerivedFrom https://doi.org/10.1046/j.1365-2427.2001.00857.x
Related Identifier IsDerivedFrom https://doi.org/10.1029/2020GB006567
Related Identifier IsDerivedFrom https://doi.org/10.1029/2019JG005161
Related Identifier IsDerivedFrom https://doi.org/10.3389/fmicb.2020.00669
Related Identifier IsDerivedFrom https://doi.org/10.1029/2020GB006661
Related Identifier IsDerivedFrom https://doi.org/10.1007/s00248-019-01330-w
Related Identifier IsDerivedFrom https://doi.org/10.1007/s00300-021-02955-9
Related Identifier IsDerivedFrom https://doi.org/10.1002/hyp.14369
Related Identifier IsDerivedFrom https://doi.org/10.3389/fmars.2017.00176
Related Identifier IsDerivedFrom https://doi.org/10.1007/s10533-020-00719-x
Related Identifier IsDerivedFrom https://doi.org/10.1002/2013WR014317
Related Identifier IsDerivedFrom https://doi.org/10.1007/s11629-014-3034-0
Related Identifier IsDerivedFrom https://doi.org/10.1016/S0016-7061(02)00335-X
Related Identifier IsDerivedFrom https://doi.org/10.1046/j.1365-2427.2003.01094.x
Related Identifier IsDerivedFrom https://doi.org/10.1016/j.chemgeo.2007.06.008
Related Identifier References https://doi.org/10.1002/joc.5086
Related Identifier References https://doi.org/10.1007/978-3-030-04447-3_7
Related Identifier References https://doi.org/10.1016/j.rse.2006.03.011
Related Identifier References https://doi.org/10.1111/gcb.16060
Related Identifier References https://doi.org/10.1016/j.rse.2006.02.016
Related Identifier References https://doi.org/10.1080/014311600210876
Related Identifier References https://doi.org/10.1029/2001JD000986
Related Identifier References https://doi.org/10.1111/j.1654-1103.2005.tb02365.x
Metadata Access https://ws.pangaea.de/oai/provider?verb=GetRecord&metadataPrefix=datacite4&identifier=oai:pangaea.de:doi:10.1594/PANGAEA.965140
Creator Holmboe, Cecilie Marie Hartvig ORCID logo; Riis, Tenna ORCID logo; Pastor, Ada (ORCID: 0000-0002-7114-770X)
Publisher PANGAEA
Publication Year 2024
Funding Reference Aarhus University https://doi.org/10.13039/100007605 Crossref Funder ID 36063-26101
Rights Creative Commons Attribution 4.0 International; https://creativecommons.org/licenses/by/4.0/
OpenAccess true
Resource Type Dataset
Format text/tab-separated-values
Size 65302 data points
Discipline Earth System Research
Spatial Coverage (-149.726W, 66.118S, 66.090E, 78.946N)