An official website of the United States government
This content will become publicly available on December 1, 2026
Snow drought to hydrologic drought progression using machine learning and probabilistic analysis
- Award ID(s):
- 1856054
- PAR ID:
- 10629356
- Publisher / Repository:
- Nature, Scientific Reports
- Date Published:
- Journal Name:
- Scientific Reports
- Volume:
- 15
- Issue:
- 1
- ISSN:
- 2045-2322
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Drought stress is an important crop yield limiting factor worldwide. Plant physiological responses to drought stress are driven by changes in gene expression. While drought-responsive genes (DRGs) have been identified in maize, regulation patterns of gene expression during progressive water deficits remain to be elucidated. In this study, we generated time-series transcriptomic data from the maize inbred line B73 under well-watered and drought conditions. Comparisons between the two conditions identified 8,626 DRGs and the stages (early, middle, and late drought) at which DRGs occurred. Different functional groups of genes were regulated at the three stages. Specifically, early and middle DRGs display higher copy number variation among diverse Zea mays lines, and they exhibited stronger associations with drought tolerance as compared to late DRGs. In addition, correlation of expression between small RNAs (sRNAs) and DRGs from the same samples identified 201 negatively sRNA/DRG correlated pairs, including genes showing high levels of association with drought tolerance, such as two glutamine synthetase genes, gln2 and gln6 . The characterization of dynamic gene responses to progressive drought stresses indicates important adaptive roles of early and middle DRGs, as well as roles played by sRNAs in gene expression regulation upon drought stress.more » « less
-
Abstract Grasslands are subject to climate change, such as severe drought, and an important aspect of their functioning is temporal stability in response to extreme climate events. Previous research has explored the impacts of extreme drought and post‐drought periods on grassland stability, yet the mechanistic pathways behind these changes have rarely been studied.Here, we implemented an experiment with 4 years of drought and 3 years of recovery to assess the effects of drought and post‐drought on the temporal stability of above‐ground net primary productivity (ANPP) and its underlying mechanisms. To do so, we measured community‐weighted mean (CWM) of six plant growth and nine seed traits, functional diversity, population stability and species asynchrony across two cold, semiarid grasslands in northern China. We also performed piecewise structural equation models (SEMs) to assess the relationships between ANPP stability and its underlying mechanisms and how drought and post‐drought periods alter the relative contribution of these mechanisms to ANPP stability.We found that temporal stability of ANPP was not reduced during drought due to grasses maintaining productivity, which compensated for increased variation of forb productivity. Moreover, ANPP recovered rapidly after drought, and both grasses and forbs contributed to community stability during the post‐drought period. Overall, ANPP stability decreased during the combined drought and post‐drought periods because of rapid changes in ANPP from drought to post‐drought. SEMs revealed that the temporal stability of ANPP during drought and post‐drought periods was modulated by functional diversity and community‐weighted mean traits directly and indirectly by altering species asynchrony and population stability. Specifically, the temporal stability of ANPP was positively correlated with functional divergence of plant communities. CWMs of seed traits (e.g. seed width and thickness), rather than plant growth traits (e.g. specific leaf area and leaf nutrient content), stabilized grassland ANPP. Productivity of plant communities with large and thick seeds was less sensitive to precipitation changes over time.These results emphasize the importance of considering both the functional trait distribution among species and seed traits of dominant species since their combined effects can stabilize ecosystem functions under global climate change scenarios. Read the freePlain Language Summaryfor this article on the Journal blog.more » « less
-
Increases in the severity and frequency of drought in a warming climate may negatively impact agricultural production and food security. Unlike previous studies that have estimated agricultural impacts of climate condition using single-crop yield distributions, we develop a multivariate probabilistic model that uses projected climatic conditions (e.g., precipitation amount or soil moisture) throughout a growing season to estimate the probability distribution of crop yields. We demonstrate the model by an analysis of the historical period 1980–2012, including the Millennium Drought in Australia (2001–2009). We find that precipitation and soil moisture deficit in dry growing seasons reduced the average annual yield of the five largest crops in Australia (wheat, broad beans, canola, lupine, and barley) by 25–45% relative to the wet growing seasons. Our model can thus produce region- and crop-specific agricultural sensitivities to climate conditions and variability. Probabilistic estimates of yield may help decision-makers in government and business to quantitatively assess the vulnerability of agriculture to climate variations. We develop a multivariate probabilistic model that uses precipitation to estimate the probability distribution of crop yields. The proposed model shows how the probability distribution of crop yield changes in response to droughts. During Australia's Millennium Drought precipitation and soil moisture deficit reduced the average annual yield of the five largest crops.more » « less
