Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 5

The DOI matches its Crossref record; 95 of the 95 references it cites were found in a registry or an index.

Groundwater and Remotely Sensed Phenology Reveal Vulnerability of Riparian Trees to Drought

Mohammadi, Rose M., Dawson, Todd E., Tiedeman, Claire R., Stella, John C., Ruhí, Albert·2025·Global Change Biology
DOI 10.1111/gcb.70564PMC PMC12555873
Open this paper in CoChat — free account

Abstract

The increasing frequency and magnitude of climatic extremes are altering water availability in dryland ecosystems globally. However, riparian vulnerability to hydroclimate whiplash remains poorly understood. Here, we examined how riparian willow, cottonwood, and valley oak trees respond to groundwater fluctuations and drought through their water use patterns and phenology. To this end, we combined time‐series analysis of in situ, high‐frequency groundwater monitoring with high‐resolution PlanetScope satellite imagery of a drought‐prone and relatively pristine watershed in California (Chalone Creek, Pinnacles National Park). We found that flow regime dictates the potential for trees to access groundwater, while the identity of tree species determines the timing and magnitude of their use. Machine‐learning models revealed that at intermittent sites groundwater depth predominantly controlled vegetation greenness, represented by the Normalized Difference Vegetation Index (NDVI). In contrast, variation in photoperiod length dominated at the perennial site where water was more reliably available. During the severe 2020–2022 drought, all species experienced reduced greenness, but phenological responses differed by flow regime. While the start of the season was delayed across all sites, trees at intermittent reaches exhibited a substantially earlier end of season during drought, resulting in growing seasons shortened by as much as 37 days. These phenological shifts vastly exceed those documented across aridity classifications in global datasets from satellite observations, ground‐based monitoring networks, and experimental precipitation manipulations. Although riparian trees in drylands have been shaped by exposure to drought over evolutionary timescales, our findings show that trees in these intermittent systems may be operating close to critical groundwater thresholds, rendering them particularly vulnerable to increasingly long and severe droughts.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1111/gcb.70564

Cites

95 of 95 references verified0 citation errors · 0 not checkable

Statistics

Figures

FIGURE 1 Study site and time series of long‐term, high‐frequency groundwater data. (a) Map of the four monitoring wells (circles) on Chalone Creek (Pinnacles National Park, California) and their corresponding riparian areas of influence (delineated by color). The CHA well is in the Chalone alluvial area, an intermittent, losing reach. SC wells are in the Sandy Creek alluvial area, where SC‐5 is a perennial, gaining reach. (b) 5‐day Palmer Drought Severity Index (PDSI) and daily precipitation (mm). Color bands indicate corresponding qualitative drought classifications from the National Weather Service. (c) 3 days in September 2021 showing diel water table fluctuations and implementation of the White method to determine the amount of evapotranspiration coming from groundwater (ETgw). The daily change in storage (S) and the recovery rate (R) are used to calculate ETgw for a 24‐h period (t). (d) Water table depth (m) time series at the four monitoring wells. All wells exhibited diel water table fluctuations. Each year, groundwater declines to its deepest point at the end of the dry season and rises with the onset of precipitation. The 2020–22 period captured a historic drought, with water years 2021 and 2022 being the second and third driest on record in California (California Department of Water Resources 2021, 2022).
FIGURE 1 Study site and time series of long‐term, high‐frequency groundwater data. (a) Map of the four monitoring wells (circles) on Chalone Creek (Pinnacles National Park, California) and their corresponding riparian areas of influence (delineated by color). The CHA well is in the Chalone alluvial area, an intermittent, losing reach. SC wells are in the Sandy Creek alluvial area, where SC‐5 is a perennial, gaining reach. (b) 5‐day Palmer Drought Severity Index (PDSI) and daily precipitation (mm). Color bands indicate corresponding qualitative drought classifications from the National Weather Service. (c) 3 days in September 2021 showing diel water table fluctuations and implementation of the White method to determine the amount of evapotranspiration coming from groundwater (ETgw). The daily change in storage (S) and the recovery rate (R) are used to calculate ETgw for a 24‐h period (t). (d) Water table depth (m) time series at the four monitoring wells. All wells exhibited diel water table fluctuations. Each year, groundwater declines to its deepest point at the end of the dry season and rises with the onset of precipitation. The 2020–22 period captured a historic drought, with water years 2021 and 2022 being the second and third driest on record in California (California Department of Water Resources 2021, 2022).CC BY-NC · PubMed Central
FIGURE 2 The mainstem of Chalone Creek. Photos of Chalone Creek, 600 m downstream of the CHA‐4 well, during (a) the wet season (March) and (b) the dry season (November) of 2024. Photo credit: Rose Mohammadi.
FIGURE 2 The mainstem of Chalone Creek. Photos of Chalone Creek, 600 m downstream of the CHA‐4 well, during (a) the wet season (March) and (b) the dry season (November) of 2024. Photo credit: Rose Mohammadi.CC BY-NC · PubMed Central
FIGURE 3 Timing and magnitude of groundwater use by trees. (a–c) Wavelet diagrams illustrating the power of each frequency in depth to water table data from wells SC‐5 (perennial), CHA‐4 (intermittent), and SC‐4 (intermittent). Warmer colors at the daily scale (period = ~1) indicate when there is a strong evapotranspiration signal. (d–f) Depth to water table (m). (g–i) Estimates of evapotranspiration from groundwater (ETgw) (mm/day) from the White method are shown for the dry (yellow triangles) and wet (blue circles) seasons. Growing seasons are highlighted in gray.
FIGURE 3 Timing and magnitude of groundwater use by trees. (a–c) Wavelet diagrams illustrating the power of each frequency in depth to water table data from wells SC‐5 (perennial), CHA‐4 (intermittent), and SC‐4 (intermittent). Warmer colors at the daily scale (period = ~1) indicate when there is a strong evapotranspiration signal. (d–f) Depth to water table (m). (g–i) Estimates of evapotranspiration from groundwater (ETgw) (mm/day) from the White method are shown for the dry (yellow triangles) and wet (blue circles) seasons. Growing seasons are highlighted in gray.CC BY-NC · PubMed Central
FIGURE 4 Normalized Difference Vegetation Index (NDVI) time series and growing season division from satellite imagery. (a) NDVI time series for willow, cottonwood, and valley oak trees within each site (indicated by line color). SC‐5 is the only perennial site among the four, and SC‐4 has the deepest groundwater. The raw average values are plotted in lighter colors, with the weighted HANTS (Harmonic Analysis of NDVI Time Series) smoothed values plotted with darker, thicker lines. Growing seasons delineated using the threshold method are highlighted in gray. (b) Violin and boxplots of smoothed growing season NDVI values binned by site. Letters indicate groupings based on Tukey's HSD test: Different letters denote statistically significant differences (p < 0.05) between sites. The boxes represent the interquartile range 25th to 75th percentiles, thick lines mark the median, whiskers represent the 5th and 95th percentiles, and outlying points are plotted individually.
FIGURE 4 Normalized Difference Vegetation Index (NDVI) time series and growing season division from satellite imagery. (a) NDVI time series for willow, cottonwood, and valley oak trees within each site (indicated by line color). SC‐5 is the only perennial site among the four, and SC‐4 has the deepest groundwater. The raw average values are plotted in lighter colors, with the weighted HANTS (Harmonic Analysis of NDVI Time Series) smoothed values plotted with darker, thicker lines. Growing seasons delineated using the threshold method are highlighted in gray. (b) Violin and boxplots of smoothed growing season NDVI values binned by site. Letters indicate groupings based on Tukey's HSD test: Different letters denote statistically significant differences (p < 0.05) between sites. The boxes represent the interquartile range 25th to 75th percentiles, thick lines mark the median, whiskers represent the 5th and 95th percentiles, and outlying points are plotted individually.CC BY-NC · PubMed Central
FIGURE 5 Variable importance and variable interaction network plots from random forest models predicting vegetation greenness (Normalized Difference Vegetation Index, NDVI) for perennial and intermittent cottonwood (a, b), valley oak (c, d), and willow (e, f). Higher variable importance (Vimp) is indicated with larger and darker nodes (green). Higher variable interaction strength (Vint) is plotted with thicker and darker lines (blue).
FIGURE 5 Variable importance and variable interaction network plots from random forest models predicting vegetation greenness (Normalized Difference Vegetation Index, NDVI) for perennial and intermittent cottonwood (a, b), valley oak (c, d), and willow (e, f). Higher variable importance (Vimp) is indicated with larger and darker nodes (green). Higher variable interaction strength (Vint) is plotted with thicker and darker lines (blue).CC BY-NC · PubMed Central
FIGURE 6 Impacts of drought on growing season Normalized Difference Vegetation Index (NDVI) and phenology. (a) Violin and boxplots of smoothed growing season NDVI values (see Section 2 for details) for each tree species binned by flow regime and drought (2020–2022) and non‐drought (2019, 2023, 2024) years. Boxes represent the interquartile range from the 25th to 75th percentiles, thick lines mark the median, whiskers represent the 5th and 95th percentiles, and outlying points are plotted individually. (b) Average growing start of season (SOS) and end of season (EOS) for each tree species binned by flow regime and drought (2020–2022) and non‐drought (2019, 2023, 2024) years. Dots represent the mean, and whiskers are +/− one standard deviation from the mean.
FIGURE 6 Impacts of drought on growing season Normalized Difference Vegetation Index (NDVI) and phenology. (a) Violin and boxplots of smoothed growing season NDVI values (see Section 2 for details) for each tree species binned by flow regime and drought (2020–2022) and non‐drought (2019, 2023, 2024) years. Boxes represent the interquartile range from the 25th to 75th percentiles, thick lines mark the median, whiskers represent the 5th and 95th percentiles, and outlying points are plotted individually. (b) Average growing start of season (SOS) and end of season (EOS) for each tree species binned by flow regime and drought (2020–2022) and non‐drought (2019, 2023, 2024) years. Dots represent the mean, and whiskers are +/− one standard deviation from the mean.CC BY-NC · PubMed Central
FIGURE 7 Comparison of drought‐induced phenological shifts across study types and ecosystems. We compared our values on tree phenological shifts in response to drought to observational and experimental values from the literature (n = 7,077,557). Box and violin plots show changes in (a) start of season, (b) end of season, and (c) growing season length, expressed as shifts in days. Colors represent different datasets: intermittent (orange) and perennial (blue) vegetation from this study; long‐term satellite observations from the Global Inventory Modeling and Mapping Studies (GIMMS) NDVI 3g dataset (1982–2015) (green; Liu et al. 2025); ground‐based observational data (purple), including the Pan European Phenology Network (PEP725) (1945–2016), the Russian “Chronicles of Nature” Network (RCNN) (1901–2017), and the China Phenological Observation Network (CPON) (1963–2014), as compiled in Liu et al. (2025); and experimental precipitation manipulations (magenta; Lu et al. 2023). Boxplots represent interquartile ranges, medians, and 5th and 95th percentiles. The dashed vertical line marks a zero shift (i.e., no change). Sample sizes (n) indicate the number of observations per group.
FIGURE 7 Comparison of drought‐induced phenological shifts across study types and ecosystems. We compared our values on tree phenological shifts in response to drought to observational and experimental values from the literature (n = 7,077,557). Box and violin plots show changes in (a) start of season, (b) end of season, and (c) growing season length, expressed as shifts in days. Colors represent different datasets: intermittent (orange) and perennial (blue) vegetation from this study; long‐term satellite observations from the Global Inventory Modeling and Mapping Studies (GIMMS) NDVI 3g dataset (1982–2015) (green; Liu et al. 2025); ground‐based observational data (purple), including the Pan European Phenology Network (PEP725) (1945–2016), the Russian “Chronicles of Nature” Network (RCNN) (1901–2017), and the China Phenological Observation Network (CPON) (1963–2014), as compiled in Liu et al. (2025); and experimental precipitation manipulations (magenta; Lu et al. 2023). Boxplots represent interquartile ranges, medians, and 5th and 95th percentiles. The dashed vertical line marks a zero shift (i.e., no change). Sample sizes (n) indicate the number of observations per group.CC BY-NC · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 16

  1. Astronomical Applications Department, 2025, Duration of Daylight/Darkness Table for One Year
  2. California Department of Water Resources, 2021, Water Year 2021: An Extreme Year
  3. California Department of Water Resources, 2022, Water Year 2022: The Drought Continues
  4. 10.1007/s00484‐011‐0494‐6 · DOI
  5. 10.32614/rj‐2023‐054 · DOI
  6. Levick L. R., 2008, The Ecological and Hydrological Significance of Ephemeral and Intermittent Streams in the Arid and Semi‐Arid American Southwest
  7. Liaw A., 2002, Classification and Regression by randomForest, R News, 2, 18
  8. 10.1111/j.1466‐8238.2006.00247.x · DOI · Paper page
  9. R Core Team, 2021, R: A Language and Environment for Statistical Computing
  10. 10.1088/1748‐9326/ac1294 · DOI
  11. Shafroth P. B., 2000, Woody Riparian Vegetation Response to Different Alluvial Water Table Regimes, Western North American Naturalist, 60, 66
  12. 10.1175/1520‐0477(1998)079<0061:APGTWA>2.0.CO;2 · doi:10.1175/1520-0477(1998)079
  13. 10.1111/j.1469‐8137.2007.02275.x · DOI · Paper page
  14. Verhoef W., 1996, Fourier Analysis of Temporal NDVI in the Southern African and American Continents, Report 108, 19
  15. White W. N., 1932, A Method of Estimating Ground‐Water Supplies Based on Discharge by Plants and Evaporation From Soil: Results of Investigations in Escalante Valley, Utah
  16. Zeileis A., 2002, Diagnostic Checking in Regression Relationships, R News, 2, 7

Cited by

See all 5

Metadata

Authors
Mohammadi, R. M., Dawson, T. E., Tiedeman, C. R., Stella, J. C., & Ruhí, A.
Affiliations
  • Department of Environmental Science, Policy, and Management University of California, Berkeley Berkeley California USA
  • Retired Hydrologist Sunnyvale CA USA
  • Department of Sustainable Resources Management State University of New York College of Environmental Science and Forestry Syracuse New York USA
ORCID
Venue
Global Change Biology
Publisher
Wiley
Published
2025-10
Volume
31
Issue
10
ISSN
  • 1354-1013
  • 1365-2486
DOI
10.1111/gcb.70564
PMC
PMC12555873
Cited by
5
References
95
Language
en
Licence
Funders
  • Adolph C. and Mary Sprague Miller Institute for Basic Research in Science, University of California Berkeley
  • Division of Environmental Biology (DEB‐2047324)
  • National Science Foundation Graduate Research Fellowship Program

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.