The Preseason Warming of the Indian Ocean Resulting in Soybean Failure in U.S.

Li, Menghan; Li, Xichen; Zhou, Yi; Hou, Yurong

doi:10.5194/egusphere-2025-2930

Preprints

https://doi.org/10.5194/egusphere-2025-2930

Preprints

04 Jul 2025

| 04 Jul 2025

The Preseason Warming of the Indian Ocean Resulting in Soybean Failure in U.S.

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Abstract. Soybean is the most important oilseed and feed crop globally. As one of the major soybean producers in the world, soybean yield variability in the United States has garnered widespread attention. We analyze the effect of the Indian Ocean sea surface temperature (SST) on soybean yield variability. Our findings indicate that variations in Indian Ocean SST during the November–December–January (hereinafter referred to as ND(-1)J) period, approximately nine months prior to harvest, account for 16 % of the anomaly in U.S. soybean yields. Furthermore, for each standard deviation change in the Indian Ocean Basin (IOB) index, there is an estimated 4.0 % change in total soybean production in the United States. The root zone soil moisture and maximum temperature during the reproductive growth stage in summer are the key factors influencing the United States soybean yields. The warming of the Indian Ocean could cause hot and dry conditions during July-August-September (JAS) by influencing ND(-1)J soil moisture and the eastern Pacific SST, leading to substantial soybean failures in the United States. Our findings emphasize the importance of the Indian Ocean SST on soybean production in the United States and reveal the pathways of this impact, which can help predict the United States soybean failures and improve food security worldwide.

Received: 20 Jun 2025 – Discussion started: 04 Jul 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 5229 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (5229 KB)

Supplement (1962 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

05 Dec 2025

The preseason warming of the Indian Ocean resulting in soybean failure in US

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Earth Syst. Dynam., 16, 2187–2199, https://doi.org/10.5194/esd-16-2187-2025,https://doi.org/10.5194/esd-16-2187-2025, 2025

Short summary

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-2930', Anonymous Referee #1, 26 Jul 2025
This manuscript presents a compelling study of the delayed teleconnection between Indian Ocean SST anomalies and U.S. soybean yield variability. The authors argue that ND(-1)J Indian Ocean Basin (IOB) warming contributes to significant soybean yield losses the following summer by inducing winter circulation anomalies, spring soil moisture deficits, and summer heat and drought stress. The central finding that IOB SST explains 16% of yield variance with a lead time of 9 months is novel and potentially valuable for seasonal agricultural forecasting. The manuscript is well-structured, the analysis is generally sound, and the physical reasoning is plausible. The manuscript is well-structured, the methods are appropriate, and the conclusions are justified. I recommend acceptance after minor revisions to enhance clarity and contextualize limitations.
A central assumption of the study is that the IOB-yield relationship reflects a signal that is distinct from ENSO. The manuscript notes that Gram-Schmidt orthogonalization was applied to remove the influence of Niño3.4 from the IOB index, which is a reasonable approach. However, given the apparent Pacific SST anomalies shown in Fig. 6, it would be helpful to clarify the extent to which the IOB signal is statistically independent of ENSO. Was the orthogonalization applied only to the IOB index, or also to the meteorological predictors used in the regression (e.g., Tmax, SMroot)? Could the Pacific anomalies still reflect residual ENSO influence? Since ENSO and Indian Ocean warming often co-evolve, further clarification would be helpful. Aconditional correlation or partial regression analysis (yield vs IOB, controlling for ENSO) would more directly test their statistical independence. If such analyses were not feasible due to sample size or other constraints, a short note acknowledging this would suffice.

In Fig. 3, Tmax and SMroot emerge as the most influential predictors of soybean yield anomalies. Given that dry soils can lead to elevated Tmax via reduced evaporative cooling, these two variables are often physically and statistically linked. This raises the question of whether they contribute independent information to the regression model or reflect overlapping aspects of the same underlying drought process. Have the authors assessed their correlation or examined variance inflation among predictors? Even a brief note on whether these variables act jointly or additively would help clarify their interpretation within the ridge regression framework.

The persistence of soil moisture from winter to summer is a key element of the proposed mechanism. Supplementary Fig. S3 appears to illustrate this, but the discussion could benefit from making more use of it. Would the authors consider highlighting which regions show the strongest ND(-1)J-JAS soil moisture correlation? A short mention in the main text would help readers better understand the spatial aspects of this memory effect.

The use of a 5-year running mean to detrend soybean yield is a standard choice to remove technological and management-related trends. However, this method may also suppress low-frequency climate variability, such as decadal SST modes, and reduce the number of effective degrees of freedom. Was the sensitivity of the results to this detrending method evaluated? For example, how do key correlations or regression outcomes compare when using a linear detrending approach instead? A brief justification for selecting the 5-year running mean, or a short note on whether this choice meaningfully affects the results, would help readers assess the robustness of the teleconnection signal.

Pearson correlation is used extensively throughout the manuscript to assess relationships among SST indices, meteorological variables, and anomalies in soybean yield. While this is a standard approach, Pearson correlation assumes linearity and normality, and can be sensitive to outliers. Were these assumptions checked in the analysis? For key relationships such as IOB-yield or SMroot-yield, would the results be consistent if Spearman rank correlation were used instead? Even a brief mention of this in the methods or supplement would help confirm the robustness of the reported associations.

The manuscript describes a compelling multi-step pathway: IOB warming leads to changes in atmospheric circulation, reduced soil moisture, increased summer heat and drought, and ultimately, yield loss. Would the authors consider adding a simple schematic to summarize this mechanism? This could help readers from interdisciplinary fields quickly grasp the whole story.

Specific comments
L59. “food securety” → “food security”.
L66. “Political units” could be ambiguous to international readers. Please specify that this refers to U.S. states.
L97. The Gram-Schmidt procedure is mentioned but not described in detail. Clarify whether Niño3.4 was regressed from IOB or vice versa and consider including a short equation or citing a standard reference.
L126. When selecting ND(-1)J as the optimal window, indicate whether a formal selection criterion (e.g., max correlation, statistical threshold) or multiple testing adjustment was applied.
L201-209. Please update figure references to follow the standard format:
L201. “Fig. 4(b) and 4(c)” → “Figs. 4(b) and 4(c)”
L208. “Fig. 4(b) and 4(e)” → “Figs. 4(b) and 4(e)”
L209. “Fig. 4(c)-(e)” → “Figs. 4(c)- (e)”
L255. The text refers to “Fig. 6(g)”, but the panels go only to (f). This should be corrected.
Units and labeling: Colorbars in Figs. 3-6 should include clear units (e.g., “% per σ” or “°C per σ”). Consistent labeling will improve readability.
Citation: https://doi.org/10.5194/egusphere-2025-2930-RC1
- AC1: 'Reply on RC1', Menghan Li, 05 Sep 2025
  
  We sincerely thank the reviewer for the detailed comments and valuable suggestions. We hope that our responses have satisfactorily addressed all concerns. A complete set of replies to the reviewer’s comments is provided in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2930-AC1
RC2:
'Comment on egusphere-2025-2930', Anonymous Referee #2, 14 Aug 2025
Review “The Preseason Warming of the Indian Ocean Resulting in Soybean Failure in U.S”
The study analyses the relation between the Indian Ocean Basin (IOB) index and soybean production/yield in the US. This comes from a clear research gap: while the role of ENSO on soybean production has been explored before, other climate oscillations have not, despite evidence of their potential influence on soybean yields. And the authors look precisely at that in this study. The approach to do so, looking for soybean yield variability among different climate indices and at different time lags, is adequate and leads to interesting results. They detected the IOB index as the other main pattern other than ENSO for soybeans in the US. The other results are also meaningful, with a 16% explanation of the yield variability due to the IOB index and a satisfactory explanation of the chain of effects between IOB index variability to soybean yield variability. I think this study has practical implications to seasonal forecasts and other planning tools, and fits the journal scope. However, I do have some comments on different aspects of the study, which follow bellow.
Major comments:
Methods:
More details are needed on the Gram-Schmidt orthogonalisation method, it is too vague for now. Among questions and points that I would like to see explained: Can you explain what it does and how you applied it? Can you expand the motivation for this (or what would happen without this step)? Could you be missing signal or information by doing that (special attention to ENSO here)? And if this is a common approach / which other studies have done this before?

Can you explain and justify the initial choice of meteorological variables? Is this based on previous studies, do similar studies select the same variables? It reads a bit unclear and arbitrary right now.

It is not clear in the text to me how root zone soil moisture is obtained or calculated. You refer to the ERA5 dataset, but as far as I am aware, this variable is not available on the ERA5 repository.

When comparing IOB with meteorological variables, you extract SLP from CRU but geopotential height at 200 hPa, and wind components at 925 hPa from ERA5. ERA5 also has SLP, so is there a reason for this? I would argue that having all variables from the same source would guarantee consistency. If you decide to keep SLP from CRU, it should be shown how similar it behaves between the two sources.

The last paragraph of the section 2.2 is confusing. On line 104, number (1), you distinguish between meteorological factors and atmospheric circulation patterns? What exactly do you refer to when you mention atmospheric circulation patterns, this has not been introduced before. Would this be the SLP, GPH200 and the wind components? If so, SLP is not an atmospheric circulation variable, and needs to be corrected. If not, then it would need to be better explained or rewritten to improve clarity.

Results & Discussion:
The results section combines both actual results and contextualisation aspects that should go into the discussion. And as a consequence, the discussion section is rather small and underdeveloped, looking more like a conclusion than a discussion. Based on that, I would suggest to have the discussion considerably expanded, with the main findings properly contextualised there. For example, the authors find DTR to be important for soybean yield using the ridge regression, which is a statistical approach. I’d like to see potential physical explanations for that (after all, DTR is the difference between two other variables, which could mean many things). Also, have other studies found similar or diverging relations between DTR and soybean yields in the area of study? These aspects should be properly discussed (you could move some of the small contextualisation points from the results to the discussion and expand them there into a coherent text).

I also missed the theoretical implications of the findings: what does it mean to have IOB index influencing soybean variability (beyond the practical point of using it to monitor it in advance)? For instance, can it have any interactions with other major climate phenomena, such as climate change? While this is not the focus of the paper, it could still be briefly discussed. Ex: What are the future projections for the IOB index? What are the future projections for soybean production in the US? Could we see a compounding interaction between both of them? These could be part of a “future work recommendation” section of what could be done next from these findings.

Finally, I would suggest for the code to be made openly available.

Minor comments:
Line 26: According to FAOSTAT, Brazil has been the main soybean producer for the past years.
Line 56: there are different verbal tenses on the same paragraph (past and present), I recommend sticking to one for consistency.
Line 58: a matter of personal taste, but I find adjectives like “valuable” unnecessary in a scientific article.
Line 59 "food securety"
Line 84: this is a matter of personal preference, but it’s more common to define precipitation as “Pr” or “Precip” than “Pre”
Line 128: can you explain explicitly in the text the logical jump (coefficient of determination (R²)) between the -0.41 corr and the 16% variability?
Figure 2: Y axis “Values” is not informative enough.
Line 213: Can you improve the clarity of the correlation sentence?
Citation: https://doi.org/10.5194/egusphere-2025-2930-RC2
- AC2: 'Reply on RC2', Menghan Li, 05 Sep 2025
  
  We sincerely thank the reviewer for the detailed comments and valuable suggestions. We hope that our responses have satisfactorily addressed all concerns. A complete set of replies to the reviewer’s comments is provided in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2930-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-2930', Anonymous Referee #1, 26 Jul 2025
This manuscript presents a compelling study of the delayed teleconnection between Indian Ocean SST anomalies and U.S. soybean yield variability. The authors argue that ND(-1)J Indian Ocean Basin (IOB) warming contributes to significant soybean yield losses the following summer by inducing winter circulation anomalies, spring soil moisture deficits, and summer heat and drought stress. The central finding that IOB SST explains 16% of yield variance with a lead time of 9 months is novel and potentially valuable for seasonal agricultural forecasting. The manuscript is well-structured, the analysis is generally sound, and the physical reasoning is plausible. The manuscript is well-structured, the methods are appropriate, and the conclusions are justified. I recommend acceptance after minor revisions to enhance clarity and contextualize limitations.
A central assumption of the study is that the IOB-yield relationship reflects a signal that is distinct from ENSO. The manuscript notes that Gram-Schmidt orthogonalization was applied to remove the influence of Niño3.4 from the IOB index, which is a reasonable approach. However, given the apparent Pacific SST anomalies shown in Fig. 6, it would be helpful to clarify the extent to which the IOB signal is statistically independent of ENSO. Was the orthogonalization applied only to the IOB index, or also to the meteorological predictors used in the regression (e.g., Tmax, SMroot)? Could the Pacific anomalies still reflect residual ENSO influence? Since ENSO and Indian Ocean warming often co-evolve, further clarification would be helpful. Aconditional correlation or partial regression analysis (yield vs IOB, controlling for ENSO) would more directly test their statistical independence. If such analyses were not feasible due to sample size or other constraints, a short note acknowledging this would suffice.

In Fig. 3, Tmax and SMroot emerge as the most influential predictors of soybean yield anomalies. Given that dry soils can lead to elevated Tmax via reduced evaporative cooling, these two variables are often physically and statistically linked. This raises the question of whether they contribute independent information to the regression model or reflect overlapping aspects of the same underlying drought process. Have the authors assessed their correlation or examined variance inflation among predictors? Even a brief note on whether these variables act jointly or additively would help clarify their interpretation within the ridge regression framework.

The persistence of soil moisture from winter to summer is a key element of the proposed mechanism. Supplementary Fig. S3 appears to illustrate this, but the discussion could benefit from making more use of it. Would the authors consider highlighting which regions show the strongest ND(-1)J-JAS soil moisture correlation? A short mention in the main text would help readers better understand the spatial aspects of this memory effect.

The use of a 5-year running mean to detrend soybean yield is a standard choice to remove technological and management-related trends. However, this method may also suppress low-frequency climate variability, such as decadal SST modes, and reduce the number of effective degrees of freedom. Was the sensitivity of the results to this detrending method evaluated? For example, how do key correlations or regression outcomes compare when using a linear detrending approach instead? A brief justification for selecting the 5-year running mean, or a short note on whether this choice meaningfully affects the results, would help readers assess the robustness of the teleconnection signal.

Pearson correlation is used extensively throughout the manuscript to assess relationships among SST indices, meteorological variables, and anomalies in soybean yield. While this is a standard approach, Pearson correlation assumes linearity and normality, and can be sensitive to outliers. Were these assumptions checked in the analysis? For key relationships such as IOB-yield or SMroot-yield, would the results be consistent if Spearman rank correlation were used instead? Even a brief mention of this in the methods or supplement would help confirm the robustness of the reported associations.

The manuscript describes a compelling multi-step pathway: IOB warming leads to changes in atmospheric circulation, reduced soil moisture, increased summer heat and drought, and ultimately, yield loss. Would the authors consider adding a simple schematic to summarize this mechanism? This could help readers from interdisciplinary fields quickly grasp the whole story.

Specific comments
L59. “food securety” → “food security”.
L66. “Political units” could be ambiguous to international readers. Please specify that this refers to U.S. states.
L97. The Gram-Schmidt procedure is mentioned but not described in detail. Clarify whether Niño3.4 was regressed from IOB or vice versa and consider including a short equation or citing a standard reference.
L126. When selecting ND(-1)J as the optimal window, indicate whether a formal selection criterion (e.g., max correlation, statistical threshold) or multiple testing adjustment was applied.
L201-209. Please update figure references to follow the standard format:
L201. “Fig. 4(b) and 4(c)” → “Figs. 4(b) and 4(c)”
L208. “Fig. 4(b) and 4(e)” → “Figs. 4(b) and 4(e)”
L209. “Fig. 4(c)-(e)” → “Figs. 4(c)- (e)”
L255. The text refers to “Fig. 6(g)”, but the panels go only to (f). This should be corrected.
Units and labeling: Colorbars in Figs. 3-6 should include clear units (e.g., “% per σ” or “°C per σ”). Consistent labeling will improve readability.
Citation: https://doi.org/10.5194/egusphere-2025-2930-RC1
- AC1: 'Reply on RC1', Menghan Li, 05 Sep 2025
  
  We sincerely thank the reviewer for the detailed comments and valuable suggestions. We hope that our responses have satisfactorily addressed all concerns. A complete set of replies to the reviewer’s comments is provided in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2930-AC1
RC2:
'Comment on egusphere-2025-2930', Anonymous Referee #2, 14 Aug 2025
Review “The Preseason Warming of the Indian Ocean Resulting in Soybean Failure in U.S”
The study analyses the relation between the Indian Ocean Basin (IOB) index and soybean production/yield in the US. This comes from a clear research gap: while the role of ENSO on soybean production has been explored before, other climate oscillations have not, despite evidence of their potential influence on soybean yields. And the authors look precisely at that in this study. The approach to do so, looking for soybean yield variability among different climate indices and at different time lags, is adequate and leads to interesting results. They detected the IOB index as the other main pattern other than ENSO for soybeans in the US. The other results are also meaningful, with a 16% explanation of the yield variability due to the IOB index and a satisfactory explanation of the chain of effects between IOB index variability to soybean yield variability. I think this study has practical implications to seasonal forecasts and other planning tools, and fits the journal scope. However, I do have some comments on different aspects of the study, which follow bellow.
Major comments:
Methods:
More details are needed on the Gram-Schmidt orthogonalisation method, it is too vague for now. Among questions and points that I would like to see explained: Can you explain what it does and how you applied it? Can you expand the motivation for this (or what would happen without this step)? Could you be missing signal or information by doing that (special attention to ENSO here)? And if this is a common approach / which other studies have done this before?

Can you explain and justify the initial choice of meteorological variables? Is this based on previous studies, do similar studies select the same variables? It reads a bit unclear and arbitrary right now.

It is not clear in the text to me how root zone soil moisture is obtained or calculated. You refer to the ERA5 dataset, but as far as I am aware, this variable is not available on the ERA5 repository.

When comparing IOB with meteorological variables, you extract SLP from CRU but geopotential height at 200 hPa, and wind components at 925 hPa from ERA5. ERA5 also has SLP, so is there a reason for this? I would argue that having all variables from the same source would guarantee consistency. If you decide to keep SLP from CRU, it should be shown how similar it behaves between the two sources.

The last paragraph of the section 2.2 is confusing. On line 104, number (1), you distinguish between meteorological factors and atmospheric circulation patterns? What exactly do you refer to when you mention atmospheric circulation patterns, this has not been introduced before. Would this be the SLP, GPH200 and the wind components? If so, SLP is not an atmospheric circulation variable, and needs to be corrected. If not, then it would need to be better explained or rewritten to improve clarity.

Results & Discussion:
The results section combines both actual results and contextualisation aspects that should go into the discussion. And as a consequence, the discussion section is rather small and underdeveloped, looking more like a conclusion than a discussion. Based on that, I would suggest to have the discussion considerably expanded, with the main findings properly contextualised there. For example, the authors find DTR to be important for soybean yield using the ridge regression, which is a statistical approach. I’d like to see potential physical explanations for that (after all, DTR is the difference between two other variables, which could mean many things). Also, have other studies found similar or diverging relations between DTR and soybean yields in the area of study? These aspects should be properly discussed (you could move some of the small contextualisation points from the results to the discussion and expand them there into a coherent text).

I also missed the theoretical implications of the findings: what does it mean to have IOB index influencing soybean variability (beyond the practical point of using it to monitor it in advance)? For instance, can it have any interactions with other major climate phenomena, such as climate change? While this is not the focus of the paper, it could still be briefly discussed. Ex: What are the future projections for the IOB index? What are the future projections for soybean production in the US? Could we see a compounding interaction between both of them? These could be part of a “future work recommendation” section of what could be done next from these findings.

Finally, I would suggest for the code to be made openly available.

Minor comments:
Line 26: According to FAOSTAT, Brazil has been the main soybean producer for the past years.
Line 56: there are different verbal tenses on the same paragraph (past and present), I recommend sticking to one for consistency.
Line 58: a matter of personal taste, but I find adjectives like “valuable” unnecessary in a scientific article.
Line 59 "food securety"
Line 84: this is a matter of personal preference, but it’s more common to define precipitation as “Pr” or “Precip” than “Pre”
Line 128: can you explain explicitly in the text the logical jump (coefficient of determination (R²)) between the -0.41 corr and the 16% variability?
Figure 2: Y axis “Values” is not informative enough.
Line 213: Can you improve the clarity of the correlation sentence?
Citation: https://doi.org/10.5194/egusphere-2025-2930-RC2
- AC2: 'Reply on RC2', Menghan Li, 05 Sep 2025
  
  We sincerely thank the reviewer for the detailed comments and valuable suggestions. We hope that our responses have satisfactorily addressed all concerns. A complete set of replies to the reviewer’s comments is provided in the attached PDF.
  
  Citation: https://doi.org/10.5194/egusphere-2025-2930-AC2

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

ED: Reconsider after major revisions (12 Sep 2025) by Karin van der Wiel

AR by Menghan Li on behalf of the Authors (01 Oct 2025)

EF by Anna Mirena Feist-Polner (08 Oct 2025) Manuscript Author's response Author's tracked changes

ED: Referee Nomination & Report Request started (10 Nov 2025) by Karin van der Wiel

RR by Anonymous Referee #1 (10 Nov 2025)

RR by Anonymous Referee #2 (18 Nov 2025)

ED: Publish subject to technical corrections (20 Nov 2025) by Karin van der Wiel

AR by Menghan Li on behalf of the Authors (21 Nov 2025) Manuscript

Journal article(s) based on this preprint

05 Dec 2025

The preseason warming of the Indian Ocean resulting in soybean failure in US

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Earth Syst. Dynam., 16, 2187–2199, https://doi.org/10.5194/esd-16-2187-2025,https://doi.org/10.5194/esd-16-2187-2025, 2025

Short summary

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Supplement

https://doi.org/10.5194/egusphere-2025-2930-supplement

Menghan Li, Xichen Li, Yi Zhou, and Yurong Hou

Viewed

Total article views: 877 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
755	104	18	877	46	28	34

HTML: 755
PDF: 104
XML: 18
Total: 877
Supplement: 46
BibTeX: 28
EndNote: 34

Views and downloads (calculated since 04 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	109	40	8	157
Aug 2025	205	10	1	216
Sep 2025	385	11	2	398
Oct 2025	26	10	3	39
Nov 2025	25	33	4	62
Dec 2025	5	0	5

Cumulative views and downloads (calculated since 04 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	109	40	8	157
Aug 2025	205	10	1	216
Sep 2025	385	11	2	398
Oct 2025	26	10	3	39
Nov 2025	25	33	4	62
Dec 2025	5	0	5

Viewed (geographical distribution)

Total article views: 885 (including HTML, PDF, and XML) Thereof 885 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 05 Dec 2025

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (5229 KB)
Metadata XML

Short summary

Warming in the Indian Ocean can cause hotter, drier U.S. summers and reduce soybean yields. Our study reveals that temperature changes in the Indian Ocean during the preceding winter explain 16 % of variations in U.S. soybean yields. This happens because warmer ocean waters disrupt atmospheric patterns, lowering soil moisture and increasing summer heat during critical growth stages, stressing soybean plants. Understanding this link helps improve crop risk forecasts and protect food security.

The Preseason Warming of the Indian Ocean Resulting in Soybean Failure in U.S.

Journal article(s) based on this preprint

Interactive discussion

Interactive discussion

Peer review completion

Suggestions for revision or reasons for rejection

Suggestions for revision or reasons for rejection

Journal article(s) based on this preprint

Supplement

Viewed

Viewed (geographical distribution)


Total:	0
HTML:	0
PDF:	0
XML:	0