Search

Search Results

• 1. [Article] Improving nitrogen management with cover crops in organic broccoli production

  Legume cover crops can serve as important sources of nitrogen (N) in sustainable agriculture and can be economically beneficial when fertilizer inputs are reduced without a yield reduction. Synchronizing ...

  Citation × Citation Citation Abstract Copy Title: Improving nitrogen management with cover crops in organic broccoli production Author: Garrett, Amy (Amy Michelle) Legume cover crops can serve as important sources of nitrogen (N) in sustainable agriculture and can be economically beneficial when fertilizer inputs are reduced without a yield reduction. Synchronizing N mineralization from organic materials with the needs of the subsequent crop is a challenge for organic growers. Predicting plant available nitrogen from cover crop residue enables N fertilizer inputs to be adjusted for optimum economic yield and reduced environmental risk. An experiment was conducted near Corvallis, OR in 2006 through 2008 to evaluate cover crop and N effects in organic broccoli production in western Oregon. The specific objectives of this experiment were to: 1) evaluate biomass production and N accumulation from selected cover crop treatments; 2) compare the effects of selected cover crops grown as sole crops and as mixtures on broccoli yield, yield components, and net economic benefit; 3) estimate the quantity of feather meal N replaced by cover crops 4) estimate plant available nitrogen from cover crop residue in an organic broccoli production system; 5) evaluate soil NO₃₋N and petiole NO₃₋N as predictors of broccoli yield; and 6) evaluate models derived from laboratory incubation of cover crop residue to predict apparent nitrogen recovery (ANR) from cover crop residue in the field. The cover crop treatments included common vetch (Vicia sativa), phacelia (Phacelia tanacetifolia), 'Monida' oats (Avena sativa L.), phacelia plus common vetch, and 'Monida' oats plus common vetch. A fallow treatment was used as the control. Prior to incorporation, cover crop samples were collected from each block and frozen for later use in laboratory aerobic incubations. After the cover crops were flail-mowed and incorporated, four N rates (0, 100, 200, and 300 kg N ha⁻¹) were randomized within each cover crop treatment in a split plot design. All cover crop treatments produced much less biomass and accumulated less N in 2008 than in 2007. Planting vetch with oats or phacelia increased biomass production and N accumulation compared to sole crops in 2007 but not in 2008. Common vetch as a sole crop or in a mixture increased broccoli yield with 0 and 100 kg N ha⁻¹ applied compared to fallow. Legume cover crop mixtures with 100 kg N ha⁻¹ produced similar net economic returns for organic compared to fallow treatments with 300 kg N ha⁻¹. Nitrogen fertilizer input can be reduced by at least 100 kg N ha⁻¹ if common vetch is in the mixture and produces more than 5000 kg ha⁻¹ of biomass (130-180 kg N ha⁻¹). Vetch as a sole crop produced higher levels of soil NO₃₋N than the fallow treatments up to 80 days after soil incorporation in 2007. Oats and phacelia as sole crops, however, reduced soil NO₃₋N compared to fallow for up to 68 days after incorporation. Vetch mixtures with oats or phacelia produced intermediate levels of soil NO₃₋N between vetch as a sole crop and the fallow. Treatment effects were similar in 2008, but differences were less due to reduced cover crop biomass compared to 2007. Broccoli petiole nitrate levels were not affected by cover crop treatment in 2007, and there was no correlation with yield. In 2008, the oat cover crop treatment reduced broccoli petiole nitrate levels compared to the fallow. Petiole nitrate levels were strongly correlated with broccoli yield, with highest yields associated with petiole NO₃₋ N greater than 10,000 ppm. In the aerobic incubations with cover crop mixtures, a quadratic model described the relationship of percent N in the mixture to the apparent nitrogen recovery (ANR) at both 4 and 10 weeks. The highest ANR (about 40 percent) was similar for both extraction days. Net mineralization occurred when the percent N of the cover crop mixture was 1.5-1.8 percent. There was a strong correlation in 2007 between the ANR predicted by the incubation-derived model and the ANR in the soil and in the aboveground broccoli biomass. The model over predicted the ANR in the field soil, however the model more accurately predicted ANR in the broccoli biomass. The incubation model correctly predicted negative ANR values for the oat and phacelia cover-crop treatments. In 2008, the laboratory-predicted ANR and the field soil ANR were correlated (r²=.45), and the laboratory model over predicted the field ANR. The incubation model gave a poor prediction of broccoli biomass ANR. Title: Improving nitrogen management with cover crops in organic broccoli production Author: Garrett, Amy (Amy Michelle) Legume cover crops can serve as important sources of nitrogen (N) in sustainable agriculture and can be economically beneficial when fertilizer inputs are reduced without a yield reduction. Synchronizing N mineralization from organic materials with the needs of the subsequent crop is a challenge for organic growers. Predicting plant available nitrogen from cover crop residue enables N fertilizer inputs to be adjusted for optimum economic yield and reduced environmental risk. An experiment was conducted near Corvallis, OR in 2006 through 2008 to evaluate cover crop and N effects in organic broccoli production in western Oregon. The specific objectives of this experiment were to: 1) evaluate biomass production and N accumulation from selected cover crop treatments; 2) compare the effects of selected cover crops grown as sole crops and as mixtures on broccoli yield, yield components, and net economic benefit; 3) estimate the quantity of feather meal N replaced by cover crops 4) estimate plant available nitrogen from cover crop residue in an organic broccoli production system; 5) evaluate soil NO₃₋N and petiole NO₃₋N as predictors of broccoli yield; and 6) evaluate models derived from laboratory incubation of cover crop residue to predict apparent nitrogen recovery (ANR) from cover crop residue in the field. The cover crop treatments included common vetch (Vicia sativa), phacelia (Phacelia tanacetifolia), 'Monida' oats (Avena sativa L.), phacelia plus common vetch, and 'Monida' oats plus common vetch. A fallow treatment was used as the control. Prior to incorporation, cover crop samples were collected from each block and frozen for later use in laboratory aerobic incubations. After the cover crops were flail-mowed and incorporated, four N rates (0, 100, 200, and 300 kg N ha⁻¹) were randomized within each cover crop treatment in a split plot design. All cover crop treatments produced much less biomass and accumulated less N in 2008 than in 2007. Planting vetch with oats or phacelia increased biomass production and N accumulation compared to sole crops in 2007 but not in 2008. Common vetch as a sole crop or in a mixture increased broccoli yield with 0 and 100 kg N ha⁻¹ applied compared to fallow. Legume cover crop mixtures with 100 kg N ha⁻¹ produced similar net economic returns for organic compared to fallow treatments with 300 kg N ha⁻¹. Nitrogen fertilizer input can be reduced by at least 100 kg N ha⁻¹ if common vetch is in the mixture and produces more than 5000 kg ha⁻¹ of biomass (130-180 kg N ha⁻¹). Vetch as a sole crop produced higher levels of soil NO₃₋N than the fallow treatments up to 80 days after soil incorporation in 2007. Oats and phacelia as sole crops, however, reduced soil NO₃₋N compared to fallow for up to 68 days after incorporation. Vetch mixtures with oats or phacelia produced intermediate levels of soil NO₃₋N between vetch as a sole crop and the fallow. Treatment effects were similar in 2008, but differences were less due to reduced cover crop biomass compared to 2007. Broccoli petiole nitrate levels were not affected by cover crop treatment in 2007, and there was no correlation with yield. In 2008, the oat cover crop treatment reduced broccoli petiole nitrate levels compared to the fallow. Petiole nitrate levels were strongly correlated with broccoli yield, with highest yields associated with petiole NO₃₋ N greater than 10,000 ppm. In the aerobic incubations with cover crop mixtures, a quadratic model described the relationship of percent N in the mixture to the apparent nitrogen recovery (ANR) at both 4 and 10 weeks. The highest ANR (about 40 percent) was similar for both extraction days. Net mineralization occurred when the percent N of the cover crop mixture was 1.5-1.8 percent. There was a strong correlation in 2007 between the ANR predicted by the incubation-derived model and the ANR in the soil and in the aboveground broccoli biomass. The model over predicted the ANR in the field soil, however the model more accurately predicted ANR in the broccoli biomass. The incubation model correctly predicted negative ANR values for the oat and phacelia cover-crop treatments. In 2008, the laboratory-predicted ANR and the field soil ANR were correlated (r²=.45), and the laboratory model over predicted the field ANR. The incubation model gave a poor prediction of broccoli biomass ANR. Close

• 2. [Article] Estimating the dry matter production, nitrogen requirements, and yield of organic farm-grown potatoes

  As part of a participatory research project, where farmers and Oregon State University researchers collaborated, aspects of potato (Solanum tuberosum L.) growing systems were studied. It was determined ...

  Citation × Citation Citation Abstract Copy Title: Estimating the dry matter production, nitrogen requirements, and yield of organic farm-grown potatoes Author: McQueen, John Paul Gainer As part of a participatory research project, where farmers and Oregon State University researchers collaborated, aspects of potato (Solanum tuberosum L.) growing systems were studied. It was determined through conversations with the farmers that quantification of certain growth parameters of potato was lacking, including dry matter accumulation, crop nitrogen (N) uptake, and yield. In order to better assist the farmers in making fertilization decisions a study of the systems was undertaken. An important aspect in making fertilization recommendations is estimating the contribution of N by the soil in a growing season. This contribution was estimated in two ways. The first was through a laboratory aerobic incubation. Soil was collected in spring, summer, and post-harvest from the participating farms' potato fields in 2006. The N mineralization incubation was conducted for 63 d at 22°C with subsampling to determine NO₃-N every 21 d beginning with Day 0. The second method was by using a plant bioassay approach. Plant biomass, excluding roots, was harvested three to four times during tuber bulking from plots where no current season amendments were added to the soil (zero-N plots). The concentration of N was determined for the plant residues and N uptake estimated. Petiole samples were analyzed for NO₃-N to monitor the relative N status during the season and to determine if there were N deficiencies. Growth measurements of potato were also conducted on "Intensive Farms" in plots where farmers conducted "typical" nutrient management. Weekly plant samples were removed beginning when tubers were 1 to 3 cm in diameter. From these samples dry matter accumulation, crop N uptake, and fresh tuber yield were estimated. Petiole samples were also taken to determine the relative N status of the plant. Yield was also measured for a second group of farm, "All-Farms". This sampling used an "as-is" approach that consisted of hand-digging sections of row and calculating yield per hectare using row spacing found in the sampled field. In the laboratory, net N mineralization rates were approximately 0.4 to 1.3 NO₃-N mg kg⁻¹ d⁻¹ during the 63-day incubation at 22°C for soils collected from 11 farms in the spring with a median of 0.7 NO₃-N mg kg⁻¹ d⁻¹. The N-supplying capacity of the soils is estimated at 120 to 160 kg N ha⁻¹ for 2000 degree days. The uptake of N by the potato in zero-N plots at harvest was 83 to 237 kg N ha⁻¹ confirming the high amounts of N mineralization observed in the laboratory incubation. The soils on these farms mineralize an estimated 3% of total soil N in less than 1400 degree days (base 0°C). On Intensive Farms, total nitrogen uptake was an estimated 145, 190, and 245 kg N ha⁻¹ for Farms 1, 2, and 3. Despite different levels of N uptake, fresh tuber yields on Intensive Farms were similar between farms with 53 Mg ha⁻¹ on Farm 1, 45 Mg ha⁻¹ on Farm 2, and 43 Mg ha⁻¹ on Farm 3. Higher N levels on Farm 3 did not increase yields and total N in shoots at harvest was highest at 150 kg ha⁻¹ verses Farm 1 with 40 kg ha⁻¹. Tuber bulking rates were 0.8, 0.7, and 1.0 Mg ha⁻¹ d⁻¹ for Farms 1, 2, and 3. Delaying harvest could have resulted in higher yields. For tuber yields of 50 Mg ha⁻¹ farmers can expect a total N uptake of approximately 200 kg N ha⁻¹ with 0.9 m between-row spacing. Fresh tuber yields on All-Farms varied by cultivar and location with a median of 19 Mg ha⁻¹ that reflected in-field between-row spacing. At harvest, medium tubers 85 to 227 g (3 to 8 oz) averaged 50% of fresh tuber yield. To accomplish higher yields, row spacing near 1 m is recommended. Petiole N levels were variable during the season with starting values lower than common recommendations and decreasing rapidly as time progressed. We estimate that the soil at these farms supplies approximately 120 to 160 kg N ha⁻¹ and applications of less than 100 kg of plant-available N should support current yields. Title: Estimating the dry matter production, nitrogen requirements, and yield of organic farm-grown potatoes Author: McQueen, John Paul Gainer As part of a participatory research project, where farmers and Oregon State University researchers collaborated, aspects of potato (Solanum tuberosum L.) growing systems were studied. It was determined through conversations with the farmers that quantification of certain growth parameters of potato was lacking, including dry matter accumulation, crop nitrogen (N) uptake, and yield. In order to better assist the farmers in making fertilization decisions a study of the systems was undertaken. An important aspect in making fertilization recommendations is estimating the contribution of N by the soil in a growing season. This contribution was estimated in two ways. The first was through a laboratory aerobic incubation. Soil was collected in spring, summer, and post-harvest from the participating farms' potato fields in 2006. The N mineralization incubation was conducted for 63 d at 22°C with subsampling to determine NO₃-N every 21 d beginning with Day 0. The second method was by using a plant bioassay approach. Plant biomass, excluding roots, was harvested three to four times during tuber bulking from plots where no current season amendments were added to the soil (zero-N plots). The concentration of N was determined for the plant residues and N uptake estimated. Petiole samples were analyzed for NO₃-N to monitor the relative N status during the season and to determine if there were N deficiencies. Growth measurements of potato were also conducted on "Intensive Farms" in plots where farmers conducted "typical" nutrient management. Weekly plant samples were removed beginning when tubers were 1 to 3 cm in diameter. From these samples dry matter accumulation, crop N uptake, and fresh tuber yield were estimated. Petiole samples were also taken to determine the relative N status of the plant. Yield was also measured for a second group of farm, "All-Farms". This sampling used an "as-is" approach that consisted of hand-digging sections of row and calculating yield per hectare using row spacing found in the sampled field. In the laboratory, net N mineralization rates were approximately 0.4 to 1.3 NO₃-N mg kg⁻¹ d⁻¹ during the 63-day incubation at 22°C for soils collected from 11 farms in the spring with a median of 0.7 NO₃-N mg kg⁻¹ d⁻¹. The N-supplying capacity of the soils is estimated at 120 to 160 kg N ha⁻¹ for 2000 degree days. The uptake of N by the potato in zero-N plots at harvest was 83 to 237 kg N ha⁻¹ confirming the high amounts of N mineralization observed in the laboratory incubation. The soils on these farms mineralize an estimated 3% of total soil N in less than 1400 degree days (base 0°C). On Intensive Farms, total nitrogen uptake was an estimated 145, 190, and 245 kg N ha⁻¹ for Farms 1, 2, and 3. Despite different levels of N uptake, fresh tuber yields on Intensive Farms were similar between farms with 53 Mg ha⁻¹ on Farm 1, 45 Mg ha⁻¹ on Farm 2, and 43 Mg ha⁻¹ on Farm 3. Higher N levels on Farm 3 did not increase yields and total N in shoots at harvest was highest at 150 kg ha⁻¹ verses Farm 1 with 40 kg ha⁻¹. Tuber bulking rates were 0.8, 0.7, and 1.0 Mg ha⁻¹ d⁻¹ for Farms 1, 2, and 3. Delaying harvest could have resulted in higher yields. For tuber yields of 50 Mg ha⁻¹ farmers can expect a total N uptake of approximately 200 kg N ha⁻¹ with 0.9 m between-row spacing. Fresh tuber yields on All-Farms varied by cultivar and location with a median of 19 Mg ha⁻¹ that reflected in-field between-row spacing. At harvest, medium tubers 85 to 227 g (3 to 8 oz) averaged 50% of fresh tuber yield. To accomplish higher yields, row spacing near 1 m is recommended. Petiole N levels were variable during the season with starting values lower than common recommendations and decreasing rapidly as time progressed. We estimate that the soil at these farms supplies approximately 120 to 160 kg N ha⁻¹ and applications of less than 100 kg of plant-available N should support current yields. Close

• 3. [Article] Nitrogen management and variety selection for dryland production of hard red winter wheat in northeastern Oregon

  Proper nitrogen (N) management and variety selection are important for profitable hard red winter (HRW) wheat production in the dryland growing regions of northeastern Oregon. In these dryland systems, ...

  Citation × Citation Citation Abstract Copy Title: Nitrogen management and variety selection for dryland production of hard red winter wheat in northeastern Oregon Author: Jepsen, Daniel R. (Daniel Robert) Proper nitrogen (N) management and variety selection are important for profitable hard red winter (HRW) wheat production in the dryland growing regions of northeastern Oregon. In these dryland systems, N management for grain yield and grain protein concentration (GPC) is challenging due to climatic and year-to-year variation in production environments. However, current fertilizer guides make little distinction between locations and incorporate relatively little data from HRW production. Identifying adequate N management practices and scenarios suitable for HRW production will help producers reduce risk and enhance profits. This study investigates the effects of fertilizer N rate, N application timing, variety and location over six site-years in northeastern Oregon from 2007-2009. Whole plant tissue nitrogen (TN) concentration at Zadoks growth stage (GS) 30 and flag leaf nitrogen (FLN) were also evaluated as decision making tools for N management in this region. Three sites representing low and intermediate precipitation zones were chosen for this study. A site at Pendleton, Oregon represented an intermediate precipitation zone (420 mm), while sites at Lexington and Arlington, Oregon were in a low precipitation (250-300 mm) zone. Study sites were minimally responsive to N treatments in terms of yield. Spring N was less detrimental to yield than fall application when N was excessive at Lexington and Arlington. Grain protein concentration response to fertilizer N was significant and varied by site-year. Some site-years proved favorable for efficient production of high GPC HRW wheat, whereas acceptable GPC was very difficult to achieve in others, underscoring the difficulty of consistently producing high GPC HRW wheat in these regions. Fertilizer N use efficiency was 18-39% at Pendleton, but generally less than 20% at Lexington and Arlington, dropping to zero in some circumstances. At all sites the soil N pool was used more efficiently than fertilizer N, indicating that HRW production is best suited where only minimal fertilizer N is required to complement crop N requirements. Spring N application improved GPC one year at Pendleton following above average late spring rainfall, and may therefore be a useful N management strategy in that environment. In contrast, spring N had a neutral or negative impact on GPC at Lexington and Arlington. Overall, current recommendations did not adequately describe N requirements observed in this study. However, requirements for achieving target GPC were generally lower and more stable at Pendleton, indicating that this and similar environments may be more suitable for HRW production than low yield, high stress environments such as Lexington and Arlington. Varieties showed similar response to N treatments regardless of site. Grain yield of HRW varieties were generally competitive with the soft white winter (SWW) variety ‘Stephens’. Among tested HRW varieties, ‘Norwest 553’ expressed the best combination of yield and GPC performance. The relationship of tissue N (TN) concentration at Zadoks growth stage 30 to GPC was stable across site-years. A critical TN level of 41 g kg-1 corresponded to 126 g kg-1 GPC. This level could be used to indicate when additional N is required to achieve desired GPC, but it remains uncertain how useful this test would be at high stress, low rainfall sites considering the poor response to spring N at Lexington and Arlington. Flag leaf N also showed promise for predicting GPC, but additional research is necessary to clarify this relationship. Title: Nitrogen management and variety selection for dryland production of hard red winter wheat in northeastern Oregon Author: Jepsen, Daniel R. (Daniel Robert) Proper nitrogen (N) management and variety selection are important for profitable hard red winter (HRW) wheat production in the dryland growing regions of northeastern Oregon. In these dryland systems, N management for grain yield and grain protein concentration (GPC) is challenging due to climatic and year-to-year variation in production environments. However, current fertilizer guides make little distinction between locations and incorporate relatively little data from HRW production. Identifying adequate N management practices and scenarios suitable for HRW production will help producers reduce risk and enhance profits. This study investigates the effects of fertilizer N rate, N application timing, variety and location over six site-years in northeastern Oregon from 2007-2009. Whole plant tissue nitrogen (TN) concentration at Zadoks growth stage (GS) 30 and flag leaf nitrogen (FLN) were also evaluated as decision making tools for N management in this region. Three sites representing low and intermediate precipitation zones were chosen for this study. A site at Pendleton, Oregon represented an intermediate precipitation zone (420 mm), while sites at Lexington and Arlington, Oregon were in a low precipitation (250-300 mm) zone. Study sites were minimally responsive to N treatments in terms of yield. Spring N was less detrimental to yield than fall application when N was excessive at Lexington and Arlington. Grain protein concentration response to fertilizer N was significant and varied by site-year. Some site-years proved favorable for efficient production of high GPC HRW wheat, whereas acceptable GPC was very difficult to achieve in others, underscoring the difficulty of consistently producing high GPC HRW wheat in these regions. Fertilizer N use efficiency was 18-39% at Pendleton, but generally less than 20% at Lexington and Arlington, dropping to zero in some circumstances. At all sites the soil N pool was used more efficiently than fertilizer N, indicating that HRW production is best suited where only minimal fertilizer N is required to complement crop N requirements. Spring N application improved GPC one year at Pendleton following above average late spring rainfall, and may therefore be a useful N management strategy in that environment. In contrast, spring N had a neutral or negative impact on GPC at Lexington and Arlington. Overall, current recommendations did not adequately describe N requirements observed in this study. However, requirements for achieving target GPC were generally lower and more stable at Pendleton, indicating that this and similar environments may be more suitable for HRW production than low yield, high stress environments such as Lexington and Arlington. Varieties showed similar response to N treatments regardless of site. Grain yield of HRW varieties were generally competitive with the soft white winter (SWW) variety ‘Stephens’. Among tested HRW varieties, ‘Norwest 553’ expressed the best combination of yield and GPC performance. The relationship of tissue N (TN) concentration at Zadoks growth stage 30 to GPC was stable across site-years. A critical TN level of 41 g kg-1 corresponded to 126 g kg-1 GPC. This level could be used to indicate when additional N is required to achieve desired GPC, but it remains uncertain how useful this test would be at high stress, low rainfall sites considering the poor response to spring N at Lexington and Arlington. Flag leaf N also showed promise for predicting GPC, but additional research is necessary to clarify this relationship. Close

• 4. [Article] Uptake and Partitioning of Nutrients in Blackberry and Raspberry and Evaluating Plant Nutrient Status for Accurate Assessment of Fertilizer Requirements

  To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by the American ...

  Citation × Citation Citation Abstract Copy Title: Uptake and Partitioning of Nutrients in Blackberry and Raspberry and Evaluating Plant Nutrient Status for Accurate Assessment of Fertilizer Requirements Author: Strik, Bernadine C., Bryla, David R. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by the American Society for Horticultural Science and can be found at: http://horttech.ashspublications.org/ Title: Uptake and Partitioning of Nutrients in Blackberry and Raspberry and Evaluating Plant Nutrient Status for Accurate Assessment of Fertilizer Requirements Author: Strik, Bernadine C., Bryla, David R. To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by the American Society for Horticultural Science and can be found at: http://horttech.ashspublications.org/ Close

• 5. [Article] Simulating the global distribution of nitrogen isotopes in the ocean

  We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth system climate model designed for millennial timescale simulations. The model includes ...

  Citation × Citation Citation Abstract Copy Title: Simulating the global distribution of nitrogen isotopes in the ocean Author: Somes, Christopher J., Lehmann, Moritz F., Bourbonnais, Annie, Eby, Michael, Schmittner, Andreas, Altabet, Mark A., Galbraith, Eric D., Mix, Alan C., Montoya, Joseph P., Letelier, Ricardo M. We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth system climate model designed for millennial timescale simulations. The model includes prognostic tracers for the two stable nitrogen isotopes, 14N and 15N, in the nitrate (NO3−), phytoplankton, zooplankton, and detritus variables of the marine ecosystem model. The isotope effects of algal NO3− uptake, nitrogen fixation, water column denitrification, and zooplankton excretion are considered as well as the removal of NO3− by sedimentary denitrification. A global database of δ15NO3− observations is compiled from previous studies and compared to the model results on a regional basis where sufficient observations exist. The model is able to qualitatively and quantitatively reproduce many of the observed patterns such as high subsurface values in water column denitrification zones and the meridional and vertical gradients in the Southern Ocean. The observed pronounced subsurface minimum in the Atlantic is underestimated by the model presumably owing to too little simulated nitrogen fixation there. Sensitivity experiments reveal that algal NO3− uptake, nitrogen fixation, and water column denitrification have the strongest effects on the simulated distribution of nitrogen isotopes, whereas the effect from zooplankton excretion is weaker. Both water column and sedimentary denitrification also have important indirect effects on the nitrogen isotope distribution by reducing the fixed nitrogen inventory, which creates an ecological niche for nitrogen fixers and, thus, stimulates additional N2 fixation in the model. Important model deficiencies are identified, and strategies for future improvement and possibilities for model application are outlined. Title: Simulating the global distribution of nitrogen isotopes in the ocean Author: Somes, Christopher J., Lehmann, Moritz F., Bourbonnais, Annie, Eby, Michael, Schmittner, Andreas, Altabet, Mark A., Galbraith, Eric D., Mix, Alan C., Montoya, Joseph P., Letelier, Ricardo M. We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth system climate model designed for millennial timescale simulations. The model includes prognostic tracers for the two stable nitrogen isotopes, 14N and 15N, in the nitrate (NO3−), phytoplankton, zooplankton, and detritus variables of the marine ecosystem model. The isotope effects of algal NO3− uptake, nitrogen fixation, water column denitrification, and zooplankton excretion are considered as well as the removal of NO3− by sedimentary denitrification. A global database of δ15NO3− observations is compiled from previous studies and compared to the model results on a regional basis where sufficient observations exist. The model is able to qualitatively and quantitatively reproduce many of the observed patterns such as high subsurface values in water column denitrification zones and the meridional and vertical gradients in the Southern Ocean. The observed pronounced subsurface minimum in the Atlantic is underestimated by the model presumably owing to too little simulated nitrogen fixation there. Sensitivity experiments reveal that algal NO3− uptake, nitrogen fixation, and water column denitrification have the strongest effects on the simulated distribution of nitrogen isotopes, whereas the effect from zooplankton excretion is weaker. Both water column and sedimentary denitrification also have important indirect effects on the nitrogen isotope distribution by reducing the fixed nitrogen inventory, which creates an ecological niche for nitrogen fixers and, thus, stimulates additional N2 fixation in the model. Important model deficiencies are identified, and strategies for future improvement and possibilities for model application are outlined. Close

• 6. [Article] Reconnecting aquatic habitats : validating historical habitat use by anadromous fishes using telemetry and stable isotope analysis above barriers

  I conducted a study to identify potential spawning habitat for anadromous salmonids above a 60-year-old hydropower dam in the headwaters of the North Umpqua River in Oregon. Like many other historical ...

  Citation × Citation Citation Abstract Copy Title: Reconnecting aquatic habitats : validating historical habitat use by anadromous fishes using telemetry and stable isotope analysis above barriers Author: Lewis, Sierra Koch I conducted a study to identify potential spawning habitat for anadromous salmonids above a 60-year-old hydropower dam in the headwaters of the North Umpqua River in Oregon. Like many other historical salmonid-bearing rivers, little documentation exists for anadromous fish presence above potential natural obstacles upstream of Soda Springs Dam. My prediction was that if migratory salmonids are allowed to move upstream of the dam using a new fish ladder they would utilize available upstream spawning habitat. I captured, radio-tagged, and transplanted wild adult summer steelhead, O. mykiss, above Soda Springs Dam and tracked daily movements for 1 year to observe habitat preferences. I also investigated evidence of historical anadromy throughout the North Umpqua River basin using stable isotope analysis of salmon-derived nitrogen (¹⁵N) in foliar samples from Douglas fir trees, P. menziesii, (>100 years old). I tested the hypothesis that I could identify undocumented, historical salmon spawning reaches above natural and anthropogenic obstacles based solely upon δ¹⁵N foliar deposition patterns. While tracking radio-tagged steelhead, I documented holding locations, the timing of spawning activities, and the outmigration of kelts. Tagged fish did not show any extensive movement in the river or tributaries above the dam. Most of the tagged adults showed incremental daily movements near the release site or downstream into the hydropower reservoir but did not travel further than 1 kilometer upstream in the main river channel. I recorded movements and localized activities that suggested some of the fish spawned and subsequently moved downstream towards the Pacific Ocean. My observations indicated that spawning behavior of tagged fish above Soda Springs Dam was delayed several weeks relative to fish spawning below the dam. There was no evidence from telemetry that any fish moved upstream in Fish Creek, the newly accessible habitat of interest in this project. My data suggest that steelhead will use restored habitat above dams, but that re-colonization activities are variable and may be affected by altered flow regimes within the restored habitat upstream of the dam and intra-specific density dependence. My stable isotope data indicated that the foliar δ¹⁵N deposition patterns were confounded with elevation. I documented potential "salmon-derived" false positives on the Umpqua National Forest above impassable waterfalls. Overall I found that foliar δ¹⁵N deposition patterns were highly variable and unexpectedly indicated a statistically significant negative correlation of foliar δ¹⁵N values with historical salmon presence. My linear mixed-effects modeling suggests that the presence of salmon is the most important indicator of foliar δ¹⁵N values, rather than the proximity of a sampled tree to stream flow above or below migratory barriers. I was not able to use the mixed effects model to identify previously undocumented salmon spawning habitat. My results from the foliar technique suggest that the method may not be universally applicable as has been previously suggested in the literature. Title: Reconnecting aquatic habitats : validating historical habitat use by anadromous fishes using telemetry and stable isotope analysis above barriers Author: Lewis, Sierra Koch I conducted a study to identify potential spawning habitat for anadromous salmonids above a 60-year-old hydropower dam in the headwaters of the North Umpqua River in Oregon. Like many other historical salmonid-bearing rivers, little documentation exists for anadromous fish presence above potential natural obstacles upstream of Soda Springs Dam. My prediction was that if migratory salmonids are allowed to move upstream of the dam using a new fish ladder they would utilize available upstream spawning habitat. I captured, radio-tagged, and transplanted wild adult summer steelhead, O. mykiss, above Soda Springs Dam and tracked daily movements for 1 year to observe habitat preferences. I also investigated evidence of historical anadromy throughout the North Umpqua River basin using stable isotope analysis of salmon-derived nitrogen (¹⁵N) in foliar samples from Douglas fir trees, P. menziesii, (>100 years old). I tested the hypothesis that I could identify undocumented, historical salmon spawning reaches above natural and anthropogenic obstacles based solely upon δ¹⁵N foliar deposition patterns. While tracking radio-tagged steelhead, I documented holding locations, the timing of spawning activities, and the outmigration of kelts. Tagged fish did not show any extensive movement in the river or tributaries above the dam. Most of the tagged adults showed incremental daily movements near the release site or downstream into the hydropower reservoir but did not travel further than 1 kilometer upstream in the main river channel. I recorded movements and localized activities that suggested some of the fish spawned and subsequently moved downstream towards the Pacific Ocean. My observations indicated that spawning behavior of tagged fish above Soda Springs Dam was delayed several weeks relative to fish spawning below the dam. There was no evidence from telemetry that any fish moved upstream in Fish Creek, the newly accessible habitat of interest in this project. My data suggest that steelhead will use restored habitat above dams, but that re-colonization activities are variable and may be affected by altered flow regimes within the restored habitat upstream of the dam and intra-specific density dependence. My stable isotope data indicated that the foliar δ¹⁵N deposition patterns were confounded with elevation. I documented potential "salmon-derived" false positives on the Umpqua National Forest above impassable waterfalls. Overall I found that foliar δ¹⁵N deposition patterns were highly variable and unexpectedly indicated a statistically significant negative correlation of foliar δ¹⁵N values with historical salmon presence. My linear mixed-effects modeling suggests that the presence of salmon is the most important indicator of foliar δ¹⁵N values, rather than the proximity of a sampled tree to stream flow above or below migratory barriers. I was not able to use the mixed effects model to identify previously undocumented salmon spawning habitat. My results from the foliar technique suggest that the method may not be universally applicable as has been previously suggested in the literature. Close

• 7. [Article] Four centuries of soil carbon and nitrogen change after severe fire in a Western Cascades forest landscape

  Fire is a major disturbance process in many forests. Long-term studies of the biogeochemical effects of fires, especially on soils, are very rare. Consequently, long-term effects of fire on soils are often ...

  Citation × Citation Citation Abstract Copy Title: Four centuries of soil carbon and nitrogen change after severe fire in a Western Cascades forest landscape Author: Giesen, Thomas William Fire is a major disturbance process in many forests. Long-term studies of the biogeochemical effects of fires, especially on soils, are very rare. Consequently, long-term effects of fire on soils are often hypothesized from short-term effects. In a chronosequence study, I studied 24 western Cascades (Oregon) forest stands thought to have been initiated in fire. Twelve of those burned about 150 years ago (“young” sites), and the other 12 burned an average of 550 years ago (“old” sites). I hypothesized that young stands would have less carbon (C) and nitrogen (N) in forest floor and in 0 -10 cm mineral soil than old stands. I found that forest floor N pools of old sites (average = 1,823 kg/ha ± s.e. = 132 kg/ha) were significantly greater than young sites (1,450 ± 98 kg/ha). Similarly, forest floor C pools of old sites (62,980 ± 5,403 kg/ha) were significantly greater than young sites (49,032 ± 2,965 kg/ha). Greater N and C pools in forest floor of old sites resulted from greater forest floor mass in old sites; concentrations of both N and C, and C:N ratios, did not differ significantly by forest age class. In mineral soil, neither concentrations nor pools of N and C differed between young and old sites. Despite overall similarity of C:N ratios in young versus old sites, potential N mineralization rates were twice as high in forest floor of old sites (average = 60 ± 7.3 mg N / g soil) than young sites (26 ± 3.5 mg N / g soil), . Nitrate accounted for only 2% or less of total N mineralized in forest floor samples. In mineral soil, potential net N mineralization did not differ by forest age class. The pattern of high net N mineralization and low nitrification in old forests is consistent with other studies of fire-prone forests, yet contrasts with many studies of forests that lack fire, and suggests that ammonium is not the sole control over nitrification in fire-prone ecosystems. Overall, fire appears to impart a longterm legacy of reduced forest floor N and C pools in this western Oregon Cascades landscape, which suggests that current fire-suppression activities in the region may increase forest floor N and C storage over historical conditions within several centuries. The differences in forest floor and soil N cycling processes that I observed by forest age class raise the further possibility that fire exclusion in these forests may change the relative abundance of soil inorganic N forms to favor ammonium over nitrate. Such changes may have unknown consequences for relative competitive abilities of plant and microbial species that rely preferentially on different N-forms to meet N nutrition requirements. While forest floor N and C pools increase from young to old stands, forest floor and soil N and C pools are not different, or decline, between 450 year old stands and the oldest stands at 800+ years, That, and other, anomalous changes in values from ~450 to 800+ years, suggest possible changes in ecosystem functions, and may indicate that this landscape could be a fruitful study area for examinations of a mature, steady-state ecosystem Title: Four centuries of soil carbon and nitrogen change after severe fire in a Western Cascades forest landscape Author: Giesen, Thomas William Fire is a major disturbance process in many forests. Long-term studies of the biogeochemical effects of fires, especially on soils, are very rare. Consequently, long-term effects of fire on soils are often hypothesized from short-term effects. In a chronosequence study, I studied 24 western Cascades (Oregon) forest stands thought to have been initiated in fire. Twelve of those burned about 150 years ago (“young” sites), and the other 12 burned an average of 550 years ago (“old” sites). I hypothesized that young stands would have less carbon (C) and nitrogen (N) in forest floor and in 0 -10 cm mineral soil than old stands. I found that forest floor N pools of old sites (average = 1,823 kg/ha ± s.e. = 132 kg/ha) were significantly greater than young sites (1,450 ± 98 kg/ha). Similarly, forest floor C pools of old sites (62,980 ± 5,403 kg/ha) were significantly greater than young sites (49,032 ± 2,965 kg/ha). Greater N and C pools in forest floor of old sites resulted from greater forest floor mass in old sites; concentrations of both N and C, and C:N ratios, did not differ significantly by forest age class. In mineral soil, neither concentrations nor pools of N and C differed between young and old sites. Despite overall similarity of C:N ratios in young versus old sites, potential N mineralization rates were twice as high in forest floor of old sites (average = 60 ± 7.3 mg N / g soil) than young sites (26 ± 3.5 mg N / g soil), . Nitrate accounted for only 2% or less of total N mineralized in forest floor samples. In mineral soil, potential net N mineralization did not differ by forest age class. The pattern of high net N mineralization and low nitrification in old forests is consistent with other studies of fire-prone forests, yet contrasts with many studies of forests that lack fire, and suggests that ammonium is not the sole control over nitrification in fire-prone ecosystems. Overall, fire appears to impart a longterm legacy of reduced forest floor N and C pools in this western Oregon Cascades landscape, which suggests that current fire-suppression activities in the region may increase forest floor N and C storage over historical conditions within several centuries. The differences in forest floor and soil N cycling processes that I observed by forest age class raise the further possibility that fire exclusion in these forests may change the relative abundance of soil inorganic N forms to favor ammonium over nitrate. Such changes may have unknown consequences for relative competitive abilities of plant and microbial species that rely preferentially on different N-forms to meet N nutrition requirements. While forest floor N and C pools increase from young to old stands, forest floor and soil N and C pools are not different, or decline, between 450 year old stands and the oldest stands at 800+ years, That, and other, anomalous changes in values from ~450 to 800+ years, suggest possible changes in ecosystem functions, and may indicate that this landscape could be a fruitful study area for examinations of a mature, steady-state ecosystem Close

• 8. [Article] Interactions between ecosystem nitrogen and bedrock control long-term calcium sources in Oregon Coast Range forests

  Ecosystem nitrogen (N) supply strongly influences the availability and cycling of other essential nutrients in temperate forests, especially calcium (Ca). Short-term additions of N that exceed ecosystem ...

  Citation × Citation Citation Abstract Copy Title: Interactions between ecosystem nitrogen and bedrock control long-term calcium sources in Oregon Coast Range forests Author: Hynicka, Justin D. Ecosystem nitrogen (N) supply strongly influences the availability and cycling of other essential nutrients in temperate forests, especially calcium (Ca). Short-term additions of N that exceed ecosystem demands often increase dissolved nitrate fluxes and decrease soil pH, which can stimulate soil Ca loss. However, the long-term effects of high N supply on ecosystem Ca availability are more difficult to determine, and may depend on the Ca content of bedrock and mineral soils. To address this, we examined major and trace element concentrations and ⁸⁷Sr/⁸⁶Sr ratios that trace Ca sources in precipitation, foliage, soil pools, and bedrock at 24 forested sites in the Oregon Coast Range having a wide, natural range of soil N (0.16 - 0.97 % N, 0-10 cm) on contrasting basaltic and sedimentary bedrock. Using a suite of 17 site properties, we also evaluated whether soil N variation across sites was related to the five major state-factors of soil and ecosystem development: climate, organisms, topography, parent material, and time. We found that as soil N increased across sites, its ¹⁵N/¹⁴N ratio declined towards atmospheric values, suggesting that soil N variation reflects a biotic legacy of symbiotic N fixation inputs. In contrast, soil N variation was unrelated to 17 other metrics of soil forming factors that represented climate (mean annual precipitation, mean annual temperature, and distance from the coast), topography (slope, soil depth, and abundance of coarse rock fragments), parent material (within bedrock type bulk and 1 M HNO₃ leachable rock Ca chemistry), and proxies of soil age (Hurst's redness rating, effective cation exchange capacity, Ca in non-exchangeable soil residues, chemical index of alteration, weathering index of Parker, Ca in coarse soil fragments, and soil Ca loss relative to bedrock). These analyses highlight symbiotic N-fixing red alder as a keystone organismal state-factor that produces a wide range of soil N accumulation in coastal Oregon forests. Strontium isotopes (⁸⁷Sr/⁸⁶Sr) and other geochemical analyses indicate that long-term Ca sources in foliage and exchangeable soil pools in Oregon Coast Range forests depend on an interactive effect between N availability and bedrock. Basaltic rocks contained nearly 20-times more Ca than sedimentary rocks across our sites, and this difference was reflected in Sr-isotope partitioning of base cation sources. Atmospheric sources dominated plant and soil pools in forests overlying Ca-poor sedimentary rock, regardless of variation in soil N, indicating extremely limited capacity of weathering to support forest Ca demands. In contrast, forests overlying basaltic rock obtained as much as 80% of Ca from rock weathering in low N sites, yet relied to a greater extent on atmospheric Ca as soil N increased, with less than 10% of Ca from rock weathering at sites with the highest soil N. Surprisingly, differences in fresh rock Ca content and base cation sources between sedimentary and basaltic sites was not reflected in ecosystem Ca availability, and instead increasing soil N caused similar declines in foliar and exchangeable Ca across both rock types. This illustrates that nutrient pool sizes do not necessarily reflect long-term nutrient supply, and highlights how coupled biogeochemical cycles within ecosystems can regulate nutrient loss and supply to biota. Broadly, our results highlight how interactions between biological and geologic factors can influence base cation sources in forest ecosystems. The sustainability of base cation supplies to forests may therefore depend greatly on variation in bedrock weathering at low N sites, yet converge to depend on atmospheric inputs in sites that receive high N loading from biological fixation or anthropogenic deposition. Title: Interactions between ecosystem nitrogen and bedrock control long-term calcium sources in Oregon Coast Range forests Author: Hynicka, Justin D. Ecosystem nitrogen (N) supply strongly influences the availability and cycling of other essential nutrients in temperate forests, especially calcium (Ca). Short-term additions of N that exceed ecosystem demands often increase dissolved nitrate fluxes and decrease soil pH, which can stimulate soil Ca loss. However, the long-term effects of high N supply on ecosystem Ca availability are more difficult to determine, and may depend on the Ca content of bedrock and mineral soils. To address this, we examined major and trace element concentrations and ⁸⁷Sr/⁸⁶Sr ratios that trace Ca sources in precipitation, foliage, soil pools, and bedrock at 24 forested sites in the Oregon Coast Range having a wide, natural range of soil N (0.16 - 0.97 % N, 0-10 cm) on contrasting basaltic and sedimentary bedrock. Using a suite of 17 site properties, we also evaluated whether soil N variation across sites was related to the five major state-factors of soil and ecosystem development: climate, organisms, topography, parent material, and time. We found that as soil N increased across sites, its ¹⁵N/¹⁴N ratio declined towards atmospheric values, suggesting that soil N variation reflects a biotic legacy of symbiotic N fixation inputs. In contrast, soil N variation was unrelated to 17 other metrics of soil forming factors that represented climate (mean annual precipitation, mean annual temperature, and distance from the coast), topography (slope, soil depth, and abundance of coarse rock fragments), parent material (within bedrock type bulk and 1 M HNO₃ leachable rock Ca chemistry), and proxies of soil age (Hurst's redness rating, effective cation exchange capacity, Ca in non-exchangeable soil residues, chemical index of alteration, weathering index of Parker, Ca in coarse soil fragments, and soil Ca loss relative to bedrock). These analyses highlight symbiotic N-fixing red alder as a keystone organismal state-factor that produces a wide range of soil N accumulation in coastal Oregon forests. Strontium isotopes (⁸⁷Sr/⁸⁶Sr) and other geochemical analyses indicate that long-term Ca sources in foliage and exchangeable soil pools in Oregon Coast Range forests depend on an interactive effect between N availability and bedrock. Basaltic rocks contained nearly 20-times more Ca than sedimentary rocks across our sites, and this difference was reflected in Sr-isotope partitioning of base cation sources. Atmospheric sources dominated plant and soil pools in forests overlying Ca-poor sedimentary rock, regardless of variation in soil N, indicating extremely limited capacity of weathering to support forest Ca demands. In contrast, forests overlying basaltic rock obtained as much as 80% of Ca from rock weathering in low N sites, yet relied to a greater extent on atmospheric Ca as soil N increased, with less than 10% of Ca from rock weathering at sites with the highest soil N. Surprisingly, differences in fresh rock Ca content and base cation sources between sedimentary and basaltic sites was not reflected in ecosystem Ca availability, and instead increasing soil N caused similar declines in foliar and exchangeable Ca across both rock types. This illustrates that nutrient pool sizes do not necessarily reflect long-term nutrient supply, and highlights how coupled biogeochemical cycles within ecosystems can regulate nutrient loss and supply to biota. Broadly, our results highlight how interactions between biological and geologic factors can influence base cation sources in forest ecosystems. The sustainability of base cation supplies to forests may therefore depend greatly on variation in bedrock weathering at low N sites, yet converge to depend on atmospheric inputs in sites that receive high N loading from biological fixation or anthropogenic deposition. Close

• 9. [Article] Nitrogen isotopes in a global ocean biogeochemical model : constraints on the coupling between denitrification and nitrogen fixation

  We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth System Climate Model designed for millennial timescale simulations. The model includes ...

  Citation × Citation Citation Abstract Copy Title: Nitrogen isotopes in a global ocean biogeochemical model : constraints on the coupling between denitrification and nitrogen fixation Author: Somes, Christopher J. We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth System Climate Model designed for millennial timescale simulations. The model includes prognostic tracers for the stable nitrogen isotopes, ¹⁴N and ¹⁵N, in the nitrate (NO₃ˉ), phytoplankton, zooplankton, and detritus variables of the marine ecosystem model. The isotope effects of algal NO₃ˉ assimilation, water column denitrification, and zooplankton excretion are considered as well as the input of newly fixed nitrogen by diazotrophs. A global database of δ¹⁵NO₃ˉ observations is compiled from previous studies and compared to the model results on a regional basis where sufficient observations exist. The model is able to qualitatively and quantitatively reproduce the observed patterns such as high subsurface values in denitrification zones, the meridional and vertical gradients in the Southern Ocean, and the meridional gradient in the Central Equatorial Pacific. The observed subsurface minimum in the Atlantic is underestimated presumably owing to too little nitrogen fixation there. Sensitivity experiments show that algal NO₃ˉ assimilation, nitrogen fixation and water column denitrification have strong effects on the simulated distribution of nitrogen isotopes, whereas the effect from zooplankton excretion is weaker. Both water column and sedimentary denitrification also have important indirect effects on the nitrogen isotopes distribution by reducing the fixed nitrogen inventory, which creates an ecological niche for diazotrophs and stimulates additional nitrogen fixation. Water column denitrification has a strong but rather localized effect on the nitrogen isotope distribution in model versions without iron limitation of diazotrophy, in which a tight coupling of nitrogen fixation exists. However, including iron limitation of diazotrophy inhibits a tight coupling between water column denitrification and nitrogen fixation in the Eastern Pacific and shifts the main location of nitrogen fixation from the Eastern Tropical Pacific to the Western Tropical Pacific, which results in a better agreement with N' = NO₃ˉ−16PO₄³ˉ and δ¹⁵NO₃ˉ observations. Thus, our model results suggest that iron limitation of diazotrophy can modulate the feedback between denitrification and nitrogen fixation in the ocean. We speculate that a feedback response time on the centennial to millennial time scale may exist between denitrification and nitrogen fixation, producing imbalances in the global oceanic fixed nitrogen cycle, which may well have contributed to past changes of atmospheric CO₂ via the biological pump. Title: Nitrogen isotopes in a global ocean biogeochemical model : constraints on the coupling between denitrification and nitrogen fixation Author: Somes, Christopher J. We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth System Climate Model designed for millennial timescale simulations. The model includes prognostic tracers for the stable nitrogen isotopes, ¹⁴N and ¹⁵N, in the nitrate (NO₃ˉ), phytoplankton, zooplankton, and detritus variables of the marine ecosystem model. The isotope effects of algal NO₃ˉ assimilation, water column denitrification, and zooplankton excretion are considered as well as the input of newly fixed nitrogen by diazotrophs. A global database of δ¹⁵NO₃ˉ observations is compiled from previous studies and compared to the model results on a regional basis where sufficient observations exist. The model is able to qualitatively and quantitatively reproduce the observed patterns such as high subsurface values in denitrification zones, the meridional and vertical gradients in the Southern Ocean, and the meridional gradient in the Central Equatorial Pacific. The observed subsurface minimum in the Atlantic is underestimated presumably owing to too little nitrogen fixation there. Sensitivity experiments show that algal NO₃ˉ assimilation, nitrogen fixation and water column denitrification have strong effects on the simulated distribution of nitrogen isotopes, whereas the effect from zooplankton excretion is weaker. Both water column and sedimentary denitrification also have important indirect effects on the nitrogen isotopes distribution by reducing the fixed nitrogen inventory, which creates an ecological niche for diazotrophs and stimulates additional nitrogen fixation. Water column denitrification has a strong but rather localized effect on the nitrogen isotope distribution in model versions without iron limitation of diazotrophy, in which a tight coupling of nitrogen fixation exists. However, including iron limitation of diazotrophy inhibits a tight coupling between water column denitrification and nitrogen fixation in the Eastern Pacific and shifts the main location of nitrogen fixation from the Eastern Tropical Pacific to the Western Tropical Pacific, which results in a better agreement with N' = NO₃ˉ−16PO₄³ˉ and δ¹⁵NO₃ˉ observations. Thus, our model results suggest that iron limitation of diazotrophy can modulate the feedback between denitrification and nitrogen fixation in the ocean. We speculate that a feedback response time on the centennial to millennial time scale may exist between denitrification and nitrogen fixation, producing imbalances in the global oceanic fixed nitrogen cycle, which may well have contributed to past changes of atmospheric CO₂ via the biological pump. Close