CORRECTED 21 JANUARY 2025; SEE BELOW

Supplementary Materials for

Global increase in the occurrence and impact of multiyear droughts

Liangzhi Chen et al.

Corresponding author: Liangzhi Chen, [Link]@[Link]

Science 387, 278 (2025)

DOI: 10.1126/science.ado4245

The PDF file includes:

Materials and Methods

Figs. S1 to S9
Tables S1 to S4
References

Correction (21 January 2025): A duplicate reference was inadvertently added to the reference list (reference 109).
That reference has now been deleted and subsequent references have been renumbered.
 Materials and Methods

Overview and workflow

The analyses conducted consist of identifying multiyear droughts (MYDs), characterizing their
climatic drivers, and screening for those MYDs that are also multiyear ecological droughts
(MYEDs) (Fig. S6). To identify MYDs, we first calculate the standardized precipitation and
potential evapotranspiration index (SPEI) (28) on an integration interval of 12 months globally,
during 1980-2018. Then, for each year, we identify droughts as contiguous patches of pixels below
a SPEI threshold. To capture MYD events, we stack these annual patches and combine patches of
consecutive years if they spatially overlap. MYDs are therefore defined as three-dimensional
shapes consisting of spatiotemporal voxels along the dimensions of latitude, longitude, and time
that are interconnected and below a given SPEI threshold. Next, we characterize the climate of
MYDs by extracting the mean daily near surface (2 m) air temperature (tas), precipitation rate (pr),
and potential evapotranspiration (pet) from those voxels that are identified to be part of a specific
MYD event. In the last step, we screen for MYEDs by ranking MYDs’ associated anomalies in the
greenness of natural vegetation, as inferred from satellite observations of the normalized difference
vegetation index (NDVI).

Detecting multiyear drought events

Calculating the standardized precipitation and evapotranspiration index. We use the

standardized precipitation and evapotranspiration index (SPEI) (88) to identify drought conditions.
SPEI is a more reliable measure of drought than precipitation anomalies alone as it additionally
considers the effect of potential evapotranspiration (28, 26, 25). Potential evapotranspiration (pet)
Correction (21amount
is the January
of2025):
water per area per time that can evaporate at the soil surface or be transpired by
plants in absence of water supply limitations. It plays an essential role in the hydrological cycle
and strongly affects drought occurrence and development (51, 89). SPEI effectively estimates for
a given location how the climatic water balance, i.e., the difference between precipitation rate (pr)
and potential evapotranspiration (pet), relates to the long-term mean. Here, we use estimates of pet
that were calculated with the Penman-Montheith approach, assuming surface conductance of a
reference crop of 12 cm height (90) (Table S1) following FAO definition, originating from
CHELSA-BIOCLIM+ database (91). Data of pr is taken from CHELSA V2.1 (92) (Table S1).
 Both data sets are monthly time series from 1980 to 2018 with global coverage at 1/24° (~5km)
spatial resolution and unit of kg m-2 month-1.

To obtain global time series of SPEI, in each pixel the frequency distribution of water balance
estimates (pr-pet) is approximated with a log-logistic probability distribution function. This
distribution function is of the form (88)
𝛼
𝐹(𝑥) = [1 + (𝑥−𝛾)𝛽 ]−1 , (1)

where 𝛼, 𝛽, and 𝛾 are scale, shape, and origin parameters that can be estimated using the L-moment
procedure (93). To obtain SPEI, 𝐹(𝑥) is then transformed to a standardized, normal random
variable with mean of zero and variance of one, using the approximation suggested by Abramowitz
and Stegun (94), as:

C + C 𝑊 + C2 𝑊 2
𝑆𝑃𝐸𝐼 = 𝑊 − 1+d0𝑊 +1d 2 , (2)
1 2𝑊 + d3 𝑊 3

where

√−2𝑙𝑛(𝑃), 𝑃 ≤ 0.5
𝑊={ (3)
√−2𝑙𝑛(1 − 𝑃), 𝑃 > 0.5

and 𝑃 = 1 − 𝐹(𝑥) . The constants are C0 = 2.515517, C1 = 0.802853, C2 = 0.010328, d1 =

1.432788, d2 = 0.189269, and d3 = 0.001308 (94). An important advantage of SPEI is that as a
standardized variable, it is comparable across space and time (88), and therefore across climate
zones.

We calculate monthly SPEI using a 12-month memory and approximate the parameters of the log-
logistic probability distribution based on the fthe time series from 1980-2018. Such a 12-month
Correction (21 January 2025):
time window is suitable to capture long-term drought events that are long enough to impose
substantial impacts on agriculture, hydrology, and ecosystems while ones at shorter time scales
would be more appropriate to detect short-term drought (89, 95–97). For each month, SPEI is
calculated considering the respective water-balance value and the values of the eleven preceding
months. We use SPEI-12 in December to represent the overall water-balance condition of the
respective year. The resulting SPEI data has been aggregated to 1/24° (~5km) spatial resolution for
computational purposes. The calculation of SPEI is done in R with the package SPEI (98).
 Identifying the spatiotemporal extent of MYDs. As SPEI is a continuous index, a classification
into drought and non-drought conditions requires a threshold value. A widely-used categorization
scheme for SPEI considers the values from 0 to -0.99 as mild drought, from -1 to -2 as moderate
to extreme drought (99). Here, we target MYDs with sustained, severe, and intense dry conditions.
To capture these conditions, we set the SPEI to -1.1 for thresholding drought conditions.

Then, we apply a mathematical morphology approach to identify spatiotemporally contiguous

MYD events globally during the period 1980-2018 (100). To this end, the binary (drought/non-
drought) grids are first stacked (in chronological order) and the resulting stack is interpreted as a
three-dimensional cuboid, with longitude, latitude, and time as dimensions, and voxels of 1/24° ×
1/24° × 12 months resolution. In addition, only drought voxels that are identified as being part of
events that last at least two consecutive years (with direct neighbours in the time dimension) are
retained. As we focus on multiyear events, we remove drought voxels from this cuboid that are not
adjacent to the neighboring drought voxels which are defined as the other 26 voxels around the
target voxel in a cubic consisting overall 27 voxels (three voxels in each dimension, 3 × 3 × 3 =
27). Moreover, we exclude voxels in regions that are classified as “hot desert” and “cold desert”
by the Köppen-Geiger climate classification system (91) (Fig. S7), as co-occurring changes in
vegetation greenness are hard to detect in these sparsely vegetated regions and SPEI anomalies are
difficult to interpret in deserts. Then, the spatiotemporal neighborhood of each drought voxel is
screened for adjacent drought voxels, and contiguous voxels are assigned to the same drought ID.
This approach corresponds to the view that a multiyear drought is an event that propagates through
space and time (50, 55). The start and end of a MYD event are given by the years of its first and
last voxel, respectively. Notably, a pixel that initially belongs to a MYD in a given year may no
longer belong to the MYD in subsequent years before it merges back into the same MYD event
Correction (21Therefore,
again. January 2025):
multiple seemingly spatiotemporally independent drought events can be part of
a single MYD event, granted that they concatenate spatially at some time point before the overall
end of a MYD event (55). The MYDs identified from this method consist of at least two voxels in
the dimension of time, representing the minimum two-year drought condition. To explore if the
identification of MYD is sensitive to spatial resolution, we performed a sensitivity analysis, using
the 20 MYD events with the largest spatiotemporal volume identified at 1/12° (~10 km), and 1/8°
(~15 km) resolution and comparing their spatiotemporal extent together with 1/24° (~5 km) (Table
 S2). The identification of contiguous events is carried out using the conncomp_3D function of the
neuroim2 (101) package in R.

Ranking MYDs by potential severity

The average SPEI, the extent of the affected area, and duration of a MYD jointly determine the
severity of a MYD impact in non-trivial ways, making it challenging to come up with a priori
assumptions about the effects and the relative importance of these factors. We therefore use an
empirical approach to estimate the potential severity of the MYDs identified, which is then used to
rank them. To do so, we have compiled a global inventory of 33 reported multiyear droughts that
occurred between 1980 and 2018 and have had substantial impacts on nature and humans, as
reported in literature (29, 39) (Table S3). For each of these reported droughts, we draw polygons
of their approximate spatial extent in Google Earth, based on the information available, and infer
their temporal extent. Then, we intersect these spatiotemporal shapes with the identified MYDs
and calculate for each MYD the total area-years it intersects with reported multiyear droughts. We
interpret these area-years of intersection with reported droughts as a proxy for MYDs’ potential
severity (response variable) and model it as a function of event-average SPEI, area affected, and
duration of MYDs (predictors). To do so, we employ a generalized linear model (102), assuming
errors to follow a Poisson distribution and fitting second-order polynomials for each predictor.
Thus, the equation is:

𝑃. 𝑠𝑒𝑣𝑒𝑟𝑖𝑡𝑦 = 𝛽0 + 𝛽1 SPEI + 𝛽2 SPEI2 + 𝛽3 area + 𝛽4 area2 + 𝛽5 time + 𝛽6 time2 , (4)

where SPEI refers to average SPEI, area refers to annual average of area affected, and time refers
to the duration of the MYD. The coefficients 𝛽0 - 𝛽6 are optimized by the GLM. Finally, for each
Correction (21 January 2025):
MYD, we use this fitted model (4) to predict potential severity and rank the MYDs accordingly.
For a summary of the resulting model see Table S4 & Fig. S7. Fitting and predicting of the
generalized linear model is done in the R environment.

Characterizing the climate of MYD events

We compile high-resolution global climate data from CHELSA V.2.1 and CHELSA-Bioclim+ (91)
to characterize the MYDs. The climate variables include mean monthly near-surface (2 m) air
 temperature (tas, °C), annual precipitation rates (pr, kg m-2 year-1), and potential evapotranspiration
(pet, kg m-2 year-1), and are available as monthly time series with 30 arcsec spatial resolution for
the period 1980-2018 (Table S1). All the variables are first aggregated by average to annual
temporal and 1/24° (~5km) spatial resolution. Then, we calculate long-term means and annual
anomalies for each variable for the period 1980-2018, with anomalies being calculated both in
absolute terms (difference from the long-term mean which is symbolized by Δ) and in relative terms
(fraction of the long-term mean which is symbolized by Q with unit of %). Finally, all climate
statistics are intersected with the spatiotemporal volume of the identified MYDs, and the values of
intersecting pixels are summarized with area-weighted averages to characterize the climatic
conditions of a given MYD event.

Quantifying the vegetation response during MYD events

We estimate vegetation responses during the identified MYD events based on satellite-sensed
anomalies in vegetation greenness. The normalized difference vegetation index is a remotely
sensed proxy for vegetation greenness that is strongly correlated with leaf area index,
photosynthetic capacity (103, 104), and productivity (105, 106), and therefore can represent
important ecosystem responses to droughts (40, 107). Regardless of development of modern sensor
quality and method of correcting atmospheric effect, the saturation problem remains in NDVI
products which become insensitive to monitor the changes in densly-vegetated areas. We thus use
the recently developed kernel normalized difference vegetation index (48) (kNDVI) to capture the
anomalies in vegetation greenness. The kNDVI is designed to capture NDVI’s nonlinear sensitivity
to vegetation density and thus reduce the saturation effects of NDVI. kNDVI has been
demonstrated to perform better than NDVI and NIRv in many applications, including forests,
Correction (21 January 2025):
biomes and climatic zones (48, 108). The formula for calculating kNDVI is:

𝑁𝐼𝑅−𝑅𝑒𝑑 2
𝑘𝑁𝐷𝑉𝐼 = tanh (( ) ), (5)
2𝛿

where 𝛿 is a length-scale parameter to be specified to represent the sensitivity of the index to

sparsely/densely vegetated regions, Red and NIR are red and near-infrared radiations, respectively.
As recommended by authors (48), we set 𝛿 to 0.5 × (NIR + red), which leads to a simplified
equation as expressed:
 𝑘𝑁𝐷𝑉𝐼 = tanh(𝑁𝐷𝑉𝐼 2 ) (6)

We calculate annual kNDVI as equation 6 by compiling monthly gridded NDVI data from the PKU
GIMMS Normalized Difference Vegetation Index product (PKU GIMMS NDVI, version 1.2)
(87) covering the globe at 1/12° spatial resolution from 1982 to 2018 to investigate the co-
occurrence of kNDVI anomalies and MYDs (Table S1). The PKU GIMMS NDVI is the longest-
lasting, global NDVI time series, and Version 1.2 is the latest version with substantial
improvements in resolving distortions form orbital drifts, degradation of sensors, and effects of
atmospheric components by being comprehensively calibrated with MODIS and Landsat products
(109). Nevertheless, beyond drought, remotely-sensed vegetation anomalies can have various
causes, including agricultural activities such as irrigation or harvesting, detection artifacts due to,
e.g., the presence of snow, windthrows, or logging activities (110). To filter out the most
important of these confounding effects, we restricted the analysis to pixels that are neither
defined as croplands nor frequently covered with snow. We considered pixels of class 12 of
LC_Type1 in MCD12Q1.061 MODIS land cover type product (111) as cropland after being
upscaled from the original 500 m to 1/12° (~10 km) resolution and identified areas with frequent
snow cover based on the Köppen-Geiger classes 30 (polar tundra) and 31 (polar frost) based on
the gridded product from CHELSA database (92).

The kNDVI anomalies (QkNDVI) over the years are calculated as the annual value relative to the
long-term mean for the period 1982-2018, which results in the negative (QkNDVI <1) or positive
(QkNDVI >1) vegetation greenness response of the regarding year.

Nominating multiyear ecological droughts (MYEDs) by accumulated anomalies in vegetation

Correction (21 January 2025):
greenness
We define multiyear ecological droughts (MYEDs) as MYDs that are associated with strong
negative anomalies in vegetation greenness. We thus rank the MYDs by the scores calculated as
follows:

𝑠𝑐𝑜𝑟𝑒 = ∑𝑛𝑖=1( 𝑄𝑘𝑁𝐷𝑉𝐼𝑖 − 1) ∗ 𝑎𝑟𝑒𝑎𝑖 , (7)

where n is the total number of pixels of an MYD, excluding croplands and snow-affected areas
(see above), and area represents the actual area of the pixel. Thus, the lower the score calculated
 from equation 7, the more substantial the decline in ecosystem greenness (negative response) co-
occurring with the MYD. Accordingly, we nominate 10 MYEDs that have the most significant
kNDVI decrease. We combine the climatic characterization and vegetation response of MYDs into
a combined severity index (SSPEI+kNDVI) that classifies MYDs with large impacts that occurred
between 1980 and 2018. The SSPEI+kNDVI is based on SPEI and kNDVI anomalies, after aggregating
them during the period that MYEDs occurred. First, aggregated SPEI and kNDVI anomalies are
individually classified into individual severity index (SSPEI and SkNDVI, respectively) from 0 to 5
corresponding to 0.9, 0.7,0.5,0.3,0.1 percentiles, respectively, with a higher index indicating a
higher impact. Then, the SSPEI+kNDVI is calculated as the sum of SSPEI and SkNDVI, resulting in a range
of 0-10.

Statistical analyses
We analyze the trends of areas affected by MYDs and their climatic characteristics (Fig. 2) by
relating averaged metrics (area, SPEI, pr, and pet anomalies) of MYDs to time (years) with
ordinary least square regression. The colored bands indicate the 95%-confidence intervals of
respective linear regressions. Figure 2 is based on the analyses of the top 500 MYDs.

We detrend kNDVI time series by subtracting the slope of linear regression from 1982 to 2018 to
remove the effect of elevated CO2 on vegetation greenness (a positive trend in kNDVI) during the
period. As studies suggest that CO2 fertilization dominates global greenness (112, 75), we
exclusively detrend the series in which the temporal linear regression is significantly positive (slope
> 0 and p < 0.05), and leave the series as original if the linear regression is negative or insignificant.
Detrended kNDVI time series were used to estimate anomalies per MYD (Fig. 4 and 5).
Correction (21 January 2025):
 Fig. S1. Variations of land area affected by top 10 multiyear droughts (MYDs).

Correction (21 January 2025):

Correction (21 January 2025):
 Fig. S2. Sensitivity of multiyear ecological drought identification on the spatial resolution. Two
maps show the top10 MYDs identified with the same procedure as that in Figure 1, except for the
spatial resolution. These two maps are at 1/12° (~10km) (a) and 1/8° (~15km) (b) resolutions,
respectively.

Correction (21 January 2025):

 Fig. S3. Climatic characteristics of global top 500 MYD events. First, the annual pet and pr
anomalies (in %) were calculated with a reference period of 1980-2018, and the anomalies for each
pixel were calculated as the arithmetic means over the years during which the multiyear droughts
occurred. Then, each pixel is classified and colored respectively against the magnitude of pet and
pr anomalies, respectively (a). Sub-bar plots show the climatic characteristics of MYDs, calculated
as the area-weighed average of anomalies across the affected regions. b. The averaged ratio of
|Qpr/Qpet|of the top 500 MYDs along the latitude from the Arctic to the Southern Hemisphere.

Correction (21 January 2025):

 Fig. S4. The nominated 10 multiyear ecological droughts ranked by the event-accumulated kNDVI
anomalies (QkNDVI) over all pixels of each drought event (see Methods).

Correction (21 January 2025):

 Fig. S5. Anomalies in kNDVI of top ten multiyear droughts (MYDs) ranked by potential severity.
Areas in green and yellow show drought events with relatively low anomalies in the kernel
Normalized Difference Vegetation index (kNDVI), while areas in blue and red show high kNDVI
anomalies. Anomalies in kNDVI are shown relative to the mean over the 1980-2018 period. The
bar plots show the mean and the standard deviation of SPEI and kNDVI anomalies during the
respective events with numbered ranks above, calculated as the area-weighed mean across all
drought-affected pixels. Grey-colored pixels indicate croplands or snow-affected areas that are not
included in the analysis as they could potentially confound the signal of kNDVI (see Methods).
Correction (21 January 2025):
 Fig. S6. Workflow to identify multiyear droughts (MYDs) and quantify their impacts on
vegetation. Arrows indicate the flow of information. SPEI is calculated based on monthly
precipitation rates and potential evapotranspiration (pet) calculated from mean daily near-surface
(2 m) air temperatures (tas), downwelling shortwave solar radiation (rsds) from CHELSA V2.1
(92) and vapor pressure deficit and near-surface (10 m) wind speed (sfcWind) from CHELSA
Bioclim+ (91). The vegetation response is then quantified for drought conditions, using the
normalized difference vegetation index (NDVI). Numbered black squares indicate the data
transformation and calculations. The details of the datasets used are listed in Table S1.

Correction (21 January 2025):

 Fig. S7. The global extent of the study region which excludes oceans, “hot desert” (Köppen
Geiger class BWh) and “cold desert” (Köppen Geiger class BWk) indicated in white color. The
classification follows the Köppen Geiger climate classification (kg0) provided by the CHELSA
V.2.1 Bioclim+ database (91).

Correction (21 January 2025):

 Fig. S8. The response curves of GLM to predict the intersecting areas between identified MYDs
and documented extreme droughts (see Methods and SI table 3).

Correction (21 January 2025):

 Fig. S9 | Occurrence of five multiyear drought (MYD) events occurred in the western US
since 1999. The subplots show the spatiotemporal distribution of the MYD events, with the hue of
the color indicating the number of years during which a pixel was affected by the respective MYD
event. Numbers indicate their rank in terms of severity among the 13,176 MYD events identified,
which were inferred empirically as a function of affected area, duration, and SPEI anomalies (see
Methods). The periods during which these MYD events occurred are indicated by their respective
colors in the legend.

The identified MYDs across the western US overlap well with the commonly recognized
Correction (21 January
megadrought in 2025):
the western US from 2000 till the present (6, 113) (Fig. S9). The reason we are
unable to completely identify the exact megadrought in the western US as mapped in the previous
studies is that the morphology approach we applied explicitly identifies the spatiotemporally
contiguous areas as individual events, which are naturally separated by any gaps in space and time.
Therefore, megadrought in the western US is largely separated into a couple of MYDs which
overall started in 1999 and ended in 2016 with substantial spatial overlaps. As a consequence, each
MYDs identified in the western US are not top-ranked but at # 61 to #166 among all events.
 Table S1: Datasets used in the study
Variables Resolution Scale Products Sources

tas (°C), 1/24°; monthly 1980- CHELSA v2.1 [Link]

2018; 017.122
pr (mm/year)
globe

pet 1/24°; monthly 1980- CHELSA v2.1, [Link]

(mm/year) 2018; CHELSA- 2017.122
BIOCLIM+
globe [Link]
2022-212

NDVI 1/12°; 1982- PKU GIMMS [Link]

annually 2018; NDVI, version 3971 [Link]
1.2 4181-2023
globe

Correction (21 January 2025):

 Table S2: Summary of sensitivity analysis of MYD identification to spatial resolutions. The
column - “ID of MYD” is the ID of the top20 MYD events with the largest spatiotemporal volume
identified at 1/24°, 1/12°, and 1/8°. The SPEI value of -1.1 is used to identify listed multiyear
ecological drought events (see Methods). The MYDs identified at 1/24° are the references to other
spatial resolutions. Therefore, if the respective MYD is not identified at the resolutions of 0.10/0.15
° as identified at 0.05°, the cell is signed NA in red color.
Size in total (million* km2) temporal duration
ID of MYD
0.05° 0.10° 0.15° 0.05° 0.10° 0.15°

1 5.02 5.02 5.02 10-18 11-18 11-18

2 2.93 2.95 2.64 08-14 08-14 09-14

3 2.46 2.46 2.46 84-87 84-87 84-87

4 2.41 2.44 2.48 98-05 98-05 98-05

5 1.70 1.82 1.95 06-14 06-14 06-14

6 2.47 2.49 2.50 92-96 92-96 92-96

7 3.13 3.22 3.05 81-87 81-92 81-92

8 1.87 1.87 1.87 10-18 10-18 10-18

9 1.41 1.43 1.44 14-17 14-17 14-17

10 1.43 1.43 1.43 07-12 07-12 07-12

11 1.79 1.78 1.78 87-90 87-90 87-90

12 1.41 1.41 1.41 09-17 09-17 09-17

13 1.55 1.54 1.55 90-98 90-98 90-98

14 1.10 1.10 1.11 04-13 04-13 04-14

Correction (21
15January 2025):
1.02 1.02 1.03 84-87 84-87 84-87

16 0.72 0.76 0.79 11-14 11-14 11-14

17 1.41 1.41 1.40 02-07 02-07 02-07

18 1.88 1.88 1.88 87-91 87-91 87-91

19 1.09 1.40 1.50 90-01 90-01 90-01

20 1.27 1.27 1.27 99-03 99-03 99-03

 Table S3: The extreme droughts reported during 1980-2018 from the drought inventories of
Spinoni et al. 2019 (Table 6) and Hessl et al. 2018 used to fit GLM for empirically ranking
identified multiyear droughts (MYDs).
# extreme droughts Start year End year Main references

1 Sahel drought 1983 1988 Henricksen, 1986; Hulme, 2001

2 Conterminous United States 1985 1990 Andreadis et al., 2005 ; Peters et al., 2002(40)

3 Patagonia 1988 1989 Rivera and Penalba, 2014

4 Greece 1989 1991 Tselepidaki et al., 1992

5 Northeastern Brazil 1992 1993 Rao et al., 1995

6 Southern Africa 1992 1993 Munro, 2006; Eldridge, 2002

7 Western U.S., Mexico 1995 1996 Hayes et al., 1999

8 Central Amazonia 1997 1998 Williamson et al., 2000

9 Central Amazonia 2000 2001 Nepstad et al., 2004

10 Central Asia, China 1999 2003 Barlow et al., 2016; Zhang and Zhou,
2015

11 Western U.S. and Canada 2000 2004 Seager, 2007

Schwalm et al., 2012
Correction (21 January 2025):

12 Balkans, Greece 2000 2002 Stagge et al., 2013

13 India 2002 2003 Dutta et al., 2015

14 Central Africa 2002 2005 Calow et al., 2010 ; Zhou et al., 2014

15 Europe 2003 2004 Ciais et al., 2005; Rebetez et al., 2006

 16 Southwestern China 2004 2010 He et al., 2011

17 Great Plains and Canada 2006 2007 Dong et al., 2011

18 Millennium Drought 2001 2009 Van Dijk et al., 2013; Heberger, 2012

19 Turkey 2007 2008 Simsek and Cakmak, 2010

20 Argentina and Chile 2008 2010 Müller et al., 2014

21 Horn of Africa 2008 2010 Masih et al., 2014

22 Central Asia and India 2008 2010 Cook et al., 2016 ; Neena et al., 2011

23 Russia 2010 2011 Russo et al., 2015; Fasullo, 2012

24 Central U.S. 2011 2013 Hoerling et al., 2014

25 Central-Southern Europ 2011 2013 Bissolli et al., 2012

26 Southern U.S. and Mexico 2011 2013 Seager et al., 2014

27 Southern China 2011 2013 Lu et al., 2014

28 Western U.S. 2011 2014 Otkin et al., 2016; Griffin and

Anchukaitis, 2014

29 East Australia 2013–16 2013 2016 Aus Gov (BOM), 2015

Correction (21 January 2025):

30 South Africa 2015 2018 Archer et al., 2017

31 Mediterranean 2015 2017 Van Lanen et al., 2016

32 Amazonia 2015 2017 Jiménez-Muñoz et al., 2016; Reliefweb,

2017
 33 Mongolia 2000 2011 Hessl et al. 2018

Correction (21 January 2025):

 Table S4: Summary of statistical significance of the relationships between the characteristics
(predictors) of identified MYDs and area intersected (response variable) with the reported extreme
drought (see Methods and Table S3).
Term P-value

area <0.001

area² <0.001

SPEI <0.001

SPEI² <0.001

time <0.001

time² <0.001

Model <0.001

The adjusted R² of the model is 0.71, indicate the high capability of explain deviance.

Correction (21 January 2025):

 References and Notes
1. S. I. Seneviratne, X. Zhang, M. Adnan, W. Badi, C. Dereczynski, A. Di Luca, S. Ghosh, I.
Iskander, J. Kossin, S. Lewis, F. Otto, I. Pinto, M. Satoh, S. M. Vicente-Serrano, M.
Wehner, B. Zhou, “Chapter 11: Weather and Climate Extreme Events in a Changing
Climate” in Climate Change 2021: The Physical Science Basis, V. Masson-Delmotte, P.
Zhai, A. Pirani, S. L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M. I.
Gomis, M. Huang, K. Leitzell, E. Lonnoy, J. B. R. Matthews, T. K. Maycock, T.
Waterfield, K. Yelekçi, R. Yu, B. Zhu, Eds. (Intergovernmental Panel on Climate Change,
2021); [Link]
2. R. S. Cottrell, K. L. Nash, B. S. Halpern, T. A. Remenyi, S. P. Corney, A. Fleming, E. A.
Fulton, S. Hornborg, A. Johne, R. A. Watson, J. L. Blanchard, Food production shocks
across land and sea. Nat. Sustain. 2, 130–137 (2019).
3. S. Fauset, T. R. Baker, S. L. Lewis, T. R. Feldpausch, K. Affum-Baffoe, E. G. Foli, K. C.
Hamer, M. D. Swaine, Drought-induced shifts in the floristic and functional composition
of tropical forests in Ghana. Ecol. Lett. 15, 1120–1129 (2012).
4. W. R. L. Anderegg, C. Schwalm, F. Biondi, J. J. Camarero, G. Koch, M. Litvak, K. Ogle, J. D.
Shaw, E. Shevliakova, A. P. Williams, A. Wolf, E. Ziaco, S. Pacala, FOREST
ECOLOGY. Pervasive drought legacies in forest ecosystems and their implications for
carbon cycle models. Science 349, 528–532 (2015).
5. L. Zhou, Y. Tian, R. B. Myneni, P. Ciais, S. Saatchi, Y. Y. Liu, S. Piao, H. Chen, E. F.
Vermote, C. Song, T. Hwang, Widespread decline of Congo rainforest greenness in the
past decade. Nature 509, 86–90 (2014).
6. B. I. Cook, J. E. Smerdon, E. R. Cook, A. P. Williams, K. J. Anchukaitis, J. S. Mankin, K.
Allen, L. Andreu-Hayles, T. R. Ault, S. Belmecheri, S. Coats, B. Coulthard, B. Fosu, P.
Grierson, D. Griffin, D. A. Herrera, M. Ionita, F. Lehner, C. Leland, K. Marvel, M. S.
Morales, V. Mishra, J. Ngoma, H. T. T. Nguyen, A. O’Donnell, J. Palmer, M. P. Rao, M.
Rodriguez-Caton, R. Seager, D. W. Stahle, S. Stevenson, U. K. Thapa, A. M. Varuolo-
Clarke, E. K. Wise, Megadroughts in the Common Era and the Anthropocene. Nat. Rev.
Earth Environ. 3, 741–757 (2022).
7. A. P. Williams, B. I. Cook, J. E. Smerdon, Rapid intensification of the emerging southwestern
North American megadrought in 2020–2021. Nat. Clim. Chang. 12, 232–234 (2022).
8. M. I. Brunner, L. M. Tallaksen, Proneness of European Catchments to Multiyear Streamflow
Correction (21 Droughts.
January 2025):
Water Resour. Res. 55, 8881–8894 (2019).
9. J. B. Basara, J. N. Maybourn, C. M. Peirano, J. E. Tate, P. J. Brown, J. D. Hoey, B. R. Smith,
Drought and Associated Impacts in the Great Plains of the United States—A Review. Int.
J. Geosci. 4, 72–81 (2013).
10. P. J. van Mantgem, N. L. Stephenson, J. C. Byrne, L. D. Daniels, J. F. Franklin, P. Z. Fulé,
M. E. Harmon, A. J. Larson, J. M. Smith, A. H. Taylor, T. T. Veblen, Widespread
increase of tree mortality rates in the western United States. Science 323, 521–524 (2009).
11. W. M. Hammond, A. P. Williams, J. T. Abatzoglou, H. D. Adams, T. Klein, R. López, C.
Sáenz-Romero, H. Hartmann, D. D. Breshears, C. D. Allen, Global field observations of
 tree die-off reveal hotter-drought fingerprint for Earth’s forests. Nat. Commun. 13, 1761
(2022).
12. M. L. Goulden, R. C. Bales, California forest die-off linked to multi-year deep soil drying in
2012–2015 drought. Nat. Geosci. 12, 632–637 (2019).
13. H. B. Zeff, J. D. Herman, P. M. Reed, G. W. Characklis, Cooperative drought adaptation:
Integrating infrastructure development, conservation, and water transfers into adaptive
policy pathways. Water Resour. Res. 52, 7327–7346 (2016).
14. G. Watts, B. von Christierson, J. Hannaford, K. Lonsdale, Testing the resilience of water
supply systems to long droughts. J. Hydrol. 414–415, 255–267 (2012).
15. T. J. Peterson, M. Saft, M. C. Peel, A. John, Watersheds may not recover from drought.
Science 372, 745–749 (2021).
16. J. P. Boisier, R. Rondanelli, R. D. Garreaud, F. Muñoz, Anthropogenic and natural
contributions to the Southeast Pacific precipitation decline and recent megadrought in
central Chile. Geophys. Res. Lett. 43, 413–421 (2016).
17. E. R. Cook, R. Seager, R. R. Heim Jr., R. S. Vose, C. Herweijer, C. Woodhouse,
Megadroughts in North America: Placing IPCC projections of hydroclimatic change in a
long-term palaeoclimate context. J. Quaternary Sci. 25, 48–61 (2010).
18. A. I. J. M. van Dijk, H. E. Beck, R. S. Crosbie, R. A. M. de Jeu, Y. Y. Liu, G. M. Podger, B.
Timbal, N. R. Viney, The Millennium Drought in southeast Australia (2001–2009):
Natural and human causes and implications for water resources, ecosystems, economy,
and society. Water Resour. Res. 49, 1040–1057 (2013).
19. M. Garcia, D. Yu, S. Park, P. Yousefi Bahambari, B. Mohajer Iravanloo, M. Sivapalan,
Weathering water extremes and cognitive biases in a changing climate. Water Secur. 15,
100110 (2022).
20. V. Fanta, M. Šálek, P. Sklenicka, How long do floods throughout the millennium remain in
the collective memory? Nat. Commun. 10, 1105 (2019).
21. A. Dai, T. Zhao, J. Chen, Climate Change and Drought: A Precipitation and Evaporation
Perspective. Curr. Clim. Change Rep. 4, 301–312 (2018).
22. X. He, L. Estes, M. Konar, D. Tian, D. Anghileri, K. Baylis, T. P. Evans, J. Sheffield,
Integrated approaches to understanding and reducing drought impact on food security
across scales. Curr. Opin. Environ. Sustain. 40, 43–54 (2019).
Correction (21 January 2025):
23. M. Trimble, T. Olivier, L. A. P. Anjos, N. Dias Tadeu, G. Giordano, L. Mac Donnell, R.
Laura, F. J. Salvadores, I. M. Santana Chaves, P. H. C. Torres, M. A. Pascual, P. R.
Jacobi, N. Mazzeo, C. Zurbriggen, L. Garrido, E. G. Jobbagy, C. Pahl Wostl, How do
basin committees deal with water crises? Reflections for adaptive water governance from
South America. Ecol. Soc. 27, 42 (2022).
24. M. J. Hayes, M. D. Svoboda, D. A. Wilhite, O. V. Vanyarkho, Monitoring the 1996 Drought
Using the Standardized Precipitation Index. Bull. Am. Meteorol. Soc. 80, 429–438 (1999).
25. S. Eslamian, K. Ostad-Ali-Askari, V. P. Singh, N. R. Dalezios, M. Ghane, Y. Yihdego, M.
Matouq, A Review of Drought Indices. Int. J. Constr. Res. Civ. Eng. 3, 48–66 (2017).
 26. S. M. Vicente-Serrano, S. Beguería, J. Lorenzo-Lacruz, J. J. Camarero, J. I. López-Moreno,
C. Azorin-Molina, J. Revuelto, E. Morán-Tejeda, A. Sanchez-Lorenzo, Performance of
Drought Indices for Ecological, Agricultural, and Hydrological Applications. Earth
Interact. 16, 1–27 (2012).
27. A. W. R. Seddon, M. Macias-Fauria, P. R. Long, D. Benz, K. J. Willis, Sensitivity of global
terrestrial ecosystems to climate variability. Nature 531, 229–232 (2016).
28. S. Beguería, S. M. Vicente-Serrano, F. Reig, B. Latorre, Standardized precipitation
evapotranspiration index (SPEI) revisited: Parameter fitting, evapotranspiration models,
tools, datasets and drought monitoring. Int. J. Climatol. 34, 3001–3023 (2014).
29. J. Spinoni, P. Barbosa, A. De Jager, N. McCormick, G. Naumann, J. V. Vogt, D. Magni, D.
Masante, M. Mazzeschi, A new global database of meteorological drought events from
1951 to 2016. J. Hydrol. Reg. Stud. 22, 100593 (2019).
30. G. E. Ponce Campos, M. S. Moran, A. Huete, Y. Zhang, C. Bresloff, T. E. Huxman, D.
Eamus, D. D. Bosch, A. R. Buda, S. A. Gunter, T. H. Scalley, S. G. Kitchen, M. P.
McClaran, W. H. McNab, D. S. Montoya, J. A. Morgan, D. P. C. Peters, E. J. Sadler, M.
S. Seyfried, P. J. Starks, Ecosystem resilience despite large-scale altered hydroclimatic
conditions. Nature 494, 349–352 (2013).
31. G. E. Maurer, A. J. Hallmark, R. F. Brown, O. E. Sala, S. L. Collins, Sensitivity of primary
production to precipitation across the United States. Ecol. Lett. 23, 527–536 (2020).
32. T. Park, S. Ganguly, H. Tømmervik, E. S. Euskirchen, K.-A. Høgda, S. R. Karlsen, V.
Brovkin, R. R. Nemani, R. B. Myneni, Changes in growing season duration and
productivity of northern vegetation inferred from long-term remote sensing data. Environ.
Res. Lett. 11, 084001 (2016).
33. P. R. L. Bittencourt, R. S. Oliveira, A. C. L. da Costa, A. L. Giles, I. Coughlin, P. B. Costa,
D. C. Bartholomew, L. V. Ferreira, S. S. Vasconcelos, F. V. Barros, J. A. S. Junior, A. A.
R. Oliveira, M. Mencuccini, P. Meir, L. Rowland, Amazonia trees have limited capacity
to acclimate plant hydraulic properties in response to long-term drought. Glob. Change
Biol. 26, 3569–3584 (2020).
34. Y. Zhang, T. F. Keenan, S. Zhou, Exacerbated drought impacts on global ecosystems due to
structural overshoot. Nat. Ecol. Evol. 5, 1490–1498 (2021).
35. S. M. Vicente-Serrano, C. Gouveia, J. J. Camarero, S. Beguería, R. Trigo, J. I. López-
Moreno, C. Azorín-Molina, E. Pasho, J. Lorenzo-Lacruz, J. Revuelto, E. Morán-Tejeda,
Correction (21 A.
January 2025):
Sanchez-Lorenzo, Response of vegetation to drought time-scales across global land
biomes. Proc. Natl. Acad. Sci. U.S.A. 110, 52–57 (2013).
36. B. Choat, T. J. Brodribb, C. R. Brodersen, R. A. Duursma, R. López, B. E. Medlyn, Triggers
of tree mortality under drought. Nature 558, 531–539 (2018).
37. S. D. Crausbay, J. Betancourt, J. Bradford, J. Cartwright, W. C. Dennison, J. Dunham, C. A.
F. Enquist, A. G. Frazier, K. R. Hall, J. S. Littell, C. H. Luce, R. Palmer, A. R. Ramirez, I.
Rangwala, L. Thompson, B. M. Walsh, S. Carter, Unfamiliar Territory: Emerging Themes
for Ecological Drought Research and Management. One Earth 3, 337–353 (2020).
38. A. K. Mishra, V. P. Singh, A review of drought concepts. J. Hydrol. 391, 202–216 (2010).
 39. A. E. Hessl, K. J. Anchukaitis, C. Jelsema, B. Cook, O. Byambasuren, C. Leland, B. Nachin,
N. Pederson, H. Tian, L. A. Hayles, Past and future drought in Mongolia. Sci. Adv. 4,
e1701832 (2018).
40. A. Buras, A. Rammig, C. S. Zang, Quantifying impacts of the 2018 drought on European
ecosystems in comparison to 2003. Biogeosciences 17, 1655–1672 (2020).
41. P. Ciais, M. Reichstein, N. Viovy, A. Granier, J. Ogée, V. Allard, M. Aubinet, N. Buchmann,
C. Bernhofer, A. Carrara, F. Chevallier, N. De Noblet, A. D. Friend, P. Friedlingstein, T.
Grünwald, B. Heinesch, P. Keronen, A. Knohl, G. Krinner, D. Loustau, G. Manca, G.
Matteucci, F. Miglietta, J. M. Ourcival, D. Papale, K. Pilegaard, S. Rambal, G. Seufert, J.
F. Soussana, M. J. Sanz, E. D. Schulze, T. Vesala, R. Valentini, Europe-wide reduction in
primary productivity caused by the heat and drought in 2003. Nature 437, 529–533
(2005).
42. A. Karnieli, N. Agam, R. T. Pinker, M. Anderson, M. L. Imhoff, G. G. Gutman, N. Panov, A.
Goldberg, Use of NDVI and Land Surface Temperature for Drought Assessment: Merits
and Limitations. J. Clim. 23, 618–633 (2010).
43. A. J. Peters, E. A. Walter-Shea, L. Ji, A. Viña, M. Hayes, M. D. Svoboda, Drought
monitoring with NDVI-based Standardized Vegetation Index. Photogramm. Eng. Remote
Sensing 68, 71–75 (2002).
44. E. Rimkus, E. Stonevicius, J. Kilpys, V. Maciulyte, D. Valiukas, Drought identification in the
eastern Baltic region using NDVI. Earth Syst. Dyn. 8, 627–637 (2017).
45. G. P. Asner, J. M. O. Scurlock, J. A. Hicke, Global synthesis of leaf area index observations:
Implications for ecological and remote sensing studies. Glob. Ecol. Biogeogr. 12, 191–
205 (2003).
46. G. Asrar, M. Fuchs, E. T. Kanemasu, J. L. Hatfield, Estimating Absorbed Photosynthetic
Radiation and Leaf Area Index from Spectral Reflectance in Wheat1. Agron. J. 76, 300–
306 (1984).
47. L. S. Galvão, J. R. dos Santos, R. D. da Silva, C. V. da Silva, Y. M. Moura, F. M. Breunig,
Following a site-specific secondary succession in the Amazon using the Landsat CDR
product and field inventory data. Int. J. Remote Sens. 36, 574–596 (2015).
48. G. Camps-Valls, M. Campos-Taberner, Á. Moreno-Martínez, S. Walther, G. Duveiller, A.
Cescatti, M. D. Mahecha, J. Muñoz-Marí, F. J. García-Haro, L. Guanter, M. Jung, J. A.
Gamon, M. Reichstein, S. W. Running, A unified vegetation index for quantifying the
Correction (21 terrestrial
January 2025):
biosphere. Sci. Adv. 7, eabc7447 (2021).
49. B. Lloyd-Hughes, A spatio-temporal structure-based approach to drought characterisation.
Int. J. Climatol. 32, 406–418 (2012).
50. J. Zscheischler, M. D. Mahecha, S. Harmeling, M. Reichstein, Detection and attribution of
large spatiotemporal extreme events in Earth observation data. Ecol. Inform. 15, 66–73
(2013).
51. A. J. Teuling, A. F. Van Loon, S. I. Seneviratne, I. Lehner, M. Aubinet, B. Heinesch, C.
Bernhofer, T. Grünwald, H. Prasse, U. Spank, Evapotranspiration amplifies European
summer drought. Geophys. Res. Lett. 40, 2071–2075 (2013).
 52. A. Berg, K. Findell, B. Lintner, A. Giannini, S. I. Seneviratne, B. van den Hurk, R. Lorenz,
A. Pitman, S. Hagemann, A. Meier, F. Cheruy, A. Ducharne, S. Malyshev, P. C. D. Milly,
Land–atmosphere feedbacks amplify aridity increase over land under global warming.
Nat. Clim. Chang. 6, 869–874 (2016).
53. S. Seneviratne, N. Nicholls, D. Easterling, C. Goodess, S. Kanae, J. Kossin, Y. Luo, J.
Marengo, K. McInnes, M. Rahimi, M. Reichstein, A. Sorteberg, C. Vera, X. Zhang, L. V.
Alexander, S. Allen, G. Benito, T. Cavazos, J. Clague, D. Conway, P. M. Della-Marta, M.
Gerber, S. Gong, B. N. Goswami, M. Hemer, C. Huggel, B. van den Hurk, V. V. Kharin,
A. Kitoh, A. M. G. Klein Tank, G. Li, S. J. Mason, W. McGuire, G. J. van Oldenborgh, B.
Orlowsky, S. Smith, W. Thiaw, A. Velegrakis, P. Yiou, T. Zhang, T. Zhou, F. W. Zwiers,
“Changes in climate extremes and their impacts on the natural physical environment,”
chap. 3 in Managing the Risks of Extreme Events and Disasters to Advance Climate
Change Adaptation, C. B. Field, , V. Barros, T. F. Stocker, D. Qin, D. J. Dokken, K. L.
Ebi, M. D. Mastrandrea, K. J. Mach, G.-K. Plattner, S. K. Allen, M. Tignor, P. M.
Midgley, Eds. (Intergovernmental Panel on Climate Change, 2012);
[Link]
54. S. I. Seneviratne, T. Corti, E. L. Davin, M. Hirschi, E. B. Jaeger, I. Lehner, B. Orlowsky, A.
J. Teuling, Investigating soil moisture–climate interactions in a changing climate: A
review. Earth Sci. Rev. 99, 125–161 (2010).
55. D. L. Schumacher, J. Keune, P. Dirmeyer, D. G. Miralles, Drought self-propagation in
drylands due to land-atmosphere feedbacks. Nat. Geosci. 15, 262–268 (2022).
56. A. Staal, B. M. Flores, A. P. D. Aguiar, J. H. C. Bosmans, I. Fetzer, O. A. Tuinenburg,
Feedback between drought and deforestation in the Amazon. Environ. Res. Lett. 15,
044024 (2020).
57. A. Dai, T. Zhao, Uncertainties in historical changes and future projections of drought. Part I:
Estimates of historical drought changes. Clim. Change 144, 519–533 (2017).
58. J. Huang, Y. Li, C. Fu, F. Chen, Q. Fu, A. Dai, M. Shinoda, Z. Ma, W. Guo, Z. Li, L. Zhang,
Y. Liu, H. Yu, Y. He, Y. Xie, X. Guan, M. Ji, L. Lin, S. Wang, H. Yan, G. Wang,
Dryland climate change: Recent progress and challenges. Rev. Geophys. 55, 719–778
(2017).
59. F. V. Barros, P. R. L. Bittencourt, M. Brum, N. Restrepo-Coupe, L. Pereira, G. S. Teodoro, S.
R. Saleska, L. S. Borma, B. O. Christoffersen, D. Penha, L. F. Alves, A. J. N. Lima, V. M.
C. Carneiro, P. Gentine, J.-E. Lee, L. E. O. C. Aragão, V. Ivanov, L. S. M. Leal, A. C.
Correction (21 January 2025):
Araujo, R. S. Oliveira, Hydraulic traits explain differential responses of Amazonian
forests to the 2015 El Niño-induced drought. New Phytol. 223, 1253–1266 (2019).
60. C. A. S. Coelho, L. Goddard, El Niño–Induced Tropical Droughts in Climate Change
Projections. J. Clim. 22, 6456–6476 (2009).
61. F. R. C. Costa, J. Schietti, S. C. Stark, M. N. Smith, The other side of tropical forest drought:
Do shallow water table regions of Amazonia act as large-scale hydrological refugia from
drought? New Phytol. 237, 714–733 (2023).
62. K. Guan, M. Pan, H. Li, A. Wolf, J. Wu, D. Medvigy, K. K. Caylor, J. Sheffield, E. F. Wood,
Y. Malhi, M. Liang, J. S. Kimball, S. R. Saleska, J. Berry, J. Joiner, A. I. Lyapustin,
 Photosynthetic seasonality of global tropical forests constrained by hydroclimate. Nat.
Geosci. 8, 284–289 (2015).
63. O. L. Phillips, L. E. O. C. Aragão, S. L. Lewis, J. B. Fisher, J. Lloyd, G. López-González, Y.
Malhi, A. Monteagudo, J. Peacock, C. A. Quesada, G. van der Heijden, S. Almeida, I.
Amaral, L. Arroyo, G. Aymard, T. R. Baker, O. Bánki, L. Blanc, D. Bonal, P. Brando, J.
Chave, Á. C. A. de Oliveira, N. D. Cardozo, C. I. Czimczik, T. R. Feldpausch, M. A.
Freitas, E. Gloor, N. Higuchi, E. Jiménez, G. Lloyd, P. Meir, C. Mendoza, A. Morel, D.
A. Neill, D. Nepstad, S. Patiño, M. C. Peñuela, A. Prieto, F. Ramírez, M. Schwarz, J.
Silva, M. Silveira, A. S. Thomas, H. ter Steege, J. Stropp, R. Vásquez, P. Zelazowski, E.
Alvarez Dávila, S. Andelman, A. Andrade, K.-J. Chao, T. Erwin, A. Di Fiore, E. Honorio
C, H. Keeling, T. J. Killeen, W. F. Laurance, A. Peña Cruz, N. C. Pitman, P. Núñez
Vargas, H. Ramírez-Angulo, A. Rudas, R. Salamão, N. Silva, J. Terborgh, A. Torres-
Lezama, Drought sensitivity of the Amazon rainforest. Science 323, 1344–1347 (2009).
64. S. R. Saleska, K. Didan, A. R. Huete, H. R. da Rocha, Amazon forests green-up during 2005
drought. Science 318, 612–612 (2007).
65. P. M. Brando, S. J. Goetz, A. Baccini, D. C. Nepstad, P. S. A. Beck, M. C. Christman,
Seasonal and interannual variability of climate and vegetation indices across the Amazon.
Proc. Natl. Acad. Sci. U.S.A. 107, 14685–14690 (2010).
66. E. A. Graham, S. S. Mulkey, K. Kitajima, N. G. Phillips, S. J. Wright, Cloud cover limits net
CO2 uptake and growth of a rainforest tree during tropical rainy seasons. Proc. Natl.
Acad. Sci. U.S.A. 100, 572–576 (2003).
67. L. Wang-Erlandsson, R. J. V. D. Ent, A. Staal, P. Keys, D. C. Zemp, I. Fetzer, M. Taniguchi,
L. Gordon, Dry Periods Amplify the Amazon and Congo Forests’ Rainfall Self-Reliance.
ESS Open Archive [Preprint] (2023).
[Link]
68. P. B. Duffy, P. Brando, G. P. Asner, C. B. Field, Projections of future meteorological drought
and wet periods in the Amazon. Proc. Natl. Acad. Sci. U.S.A. 112, 13172–13177 (2015).
69. D. L. Hoover, A. K. Knapp, M. D. Smith, Resistance and resilience of a grassland ecosystem
to climate extremes. Ecology 95, 2646–2656 (2014).
70. A. J. Felton, A. K. Knapp, M. D. Smith, Precipitation-productivity relationships and the
duration of precipitation anomalies: An underappreciated dimension of climate change.
Glob. Change Biol. 27, 1127–1140 (2021).
Correction (21
71. J. January
Zou, 2025):
J. Ding, S. Huang, B. Liu, Ecosystem Resistance and Resilience after Dry and Wet
Events across Central Asia Based on Remote Sensing Data. Remote Sens. 15, 3165
(2023).
72. W. Luo, T. O. Muraina, R. J. Griffin-Nolan, W. Ma, L. Song, W. Fu, Q. Yu, A. K. Knapp, Z.
Wang, X. Han, S. L. Collins, Responses of a semiarid grassland to recurrent drought are
linked to community functional composition. Ecology 104, e3920 (2023).
73. E. Stuart-Haëntjens, H. J. De Boeck, N. P. Lemoine, P. Mänd, G. Kröel-Dulay, I. K. Schmidt,
A. Jentsch, A. Stampfli, W. R. L. Anderegg, M. Bahn, J. Kreyling, T. Wohlgemuth, F.
Lloret, A. T. Classen, C. M. Gough, M. D. Smith, Mean annual precipitation predicts
primary production resistance and resilience to extreme drought. Sci. Total Environ. 636,
360–366 (2018).
 74. R. J. Griffin-Nolan, D. M. Blumenthal, S. L. Collins, T. E. Farkas, A. M. Hoffman, K. E.
Mueller, T. W. Ocheltree, M. D. Smith, K. D. Whitney, A. K. Knapp, Shifts in plant
functional composition following long-term drought in grasslands. J. Ecol. 107, 2133–
2148 (2019).
75. S. Piao, X. Wang, T. Park, C. Chen, X. Lian, Y. He, J. W. Bjerke, A. Chen, P. Ciais, H.
Tømmervik, R. R. Nemani, R. B. Myneni, Characteristics, drivers and feedbacks of global
greening. Nat. Rev. Earth Environ. 1, 14–27 (2020).
76. M. Macias-Fauria, B. C. Forbes, P. Zetterberg, T. Kumpula, Eurasian Arctic greening reveals
teleconnections and the potential for structurally novel ecosystems. Nat. Clim. Chang. 2,
613–618 (2012).
77. W. Jiao, L. Wang, W. K. Smith, Q. Chang, H. Wang, P. D’Odorico, Observed increasing
water constraint on vegetation growth over the last three decades. Nat. Commun. 12, 3777
(2021).
78. J. Pulliainen, K. Luojus, C. Derksen, L. Mudryk, J. Lemmetyinen, M. Salminen, J. Ikonen, M.
Takala, J. Cohen, T. Smolander, J. Norberg, Patterns and trends of Northern Hemisphere
snow mass from 1980 to 2018. Nature 581, 294–298 (2020).
79. R. Bintanja, O. Andry, Towards a rain-dominated Arctic. Nat. Clim. Chang. 7, 263–267
(2017).
80. R. Bintanja, F. M. Selten, Future increases in Arctic precipitation linked to local evaporation
and sea-ice retreat. Nature 509, 479–482 (2014).
81. C. R. Schwalm, W. R. L. Anderegg, A. M. Michalak, J. B. Fisher, F. Biondi, G. Koch, M.
Litvak, K. Ogle, J. D. Shaw, A. Wolf, D. N. Huntzinger, K. Schaefer, R. Cook, Y. Wei,
Y. Fang, D. Hayes, M. Huang, A. Jain, H. Tian, Global patterns of drought recovery.
Nature 548, 202–205 (2017).
82. G. W. Moore, C. B. Edgar, J. G. Vogel, R. A. Washington-Allen, R. G. March, R. Zehnder,
Tree mortality from an exceptional drought spanning mesic to semiarid ecoregions. Ecol.
Appl. 26, 602–611 (2016).
83. Y. Yan, S. Piao, W. M. Hammond, A. Chen, S. Hong, H. Xu, S. M. Munson, R. B. Myneni,
C. D. Allen, Climate-induced tree-mortality pulses are obscured by broad-scale and long-
term greening. Nat. Ecol. Evol. 8, 912–923 (2024).
84. M. N. Nguyen, P. T. B. Nguyen, T. P. D. Van, V. H. Phan, B. T. Nguyen, V. T. Pham, T. H.
Correction (21 Nguyen,
January 2025):
An understanding of water governance systems in responding to extreme
droughts in the Vietnamese Mekong Delta. Int. J. Water Resour. Dev. 37, 256–277
(2021).
85. D. M. Olson, E. Dinerstein, E. D. Wikramanayake, N. D. Burgess, G. V. N. Powell, E. C.
Underwood, J. A. D’amico, I. Itoua, H. E. Strand, J. C. Morrison, C. J. Loucks, T. F.
Allnutt, T. H. Ricketts, Y. Kura, J. F. Lamoreux, W. W. Wettengel, P. Hedao, K. R.
Kassem, Terrestrial Ecoregions of the World: A New Map of Life on Earth: A new global
map of terrestrial ecoregions provides an innovative tool for conserving biodiversity.
Bioscience 51, 933–938 (2001).
 86. L. Chen, P. Brun, P. Buri, S. Fatichi, A. Gessler, M. J. McCarthy, F. Pellicciotti, B. Stocker,
D. N. Karger, High resolution global standardized drought indices, EnviDat (2024);
[Link]
87. M. Li, S. Cao, Z. Zhu, Z. Wang, R. B. Myneni, S. Piao, Spatiotemporally Consistent Global
Dataset of the GIMMS Normalized Difference Vegetation Index (PKU GIMMS NDVI)
from 1982 to 2022, Version V1.2, Zenodo (2023);
[Link]
88. S. M. Vicente-Serrano, S. Beguería, J. I. López-Moreno, A Multiscalar Drought Index
Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index. J.
Clim. 23, 1696–1718 (2010).
89. A. F. Van Loon, Hydrological drought explained. WIREs. Water 2, 359–392 (2015).
90. R. Allan, L. Pereira, M. Smith, Crop Evapotranspiration-Guidelines for Computing Crop
Water Requirements-FAO Irrigation and Drainage Paper 56 (United Nations Food and
Agriculture Organization, 1998); [Link]
91. P. Brun, N. E. Zimmermann, C. Hari, L. Pellissier, D. N. Karger, Global climate-related
predictors at kilometer resolution for the past and future. Earth Syst. Sci. Data 14, 5573–
5603 (2022).
92. D. N. Karger, O. Conrad, J. Böhner, T. Kawohl, H. Kreft, R. W. Soria-Auza, N. E.
Zimmermann, H. P. Linder, M. Kessler, Climatologies at high resolution for the earth’s
land surface areas. Sci. Data 4, 170122 (2017).
93. M. I. Ahmad, C. D. Sinclair, A. Werritty, Log-logistic flood frequency analysis. J. Hydrol.
98, 205–224 (1988).
94. M. Abramowitz, I. A. Stegun, Handbook of Mathematical Functions with Formulas, Graphs,
and Mathematical Tables (US Government Printing Office, 1968).
95. A. Zargar, R. Sadiq, B. Naser, F. I. Khan, A review of drought indices. Environ. Rev. 19
(NA), 333–349 (2011).
96. I. J. Slette, A. K. Post, M. Awad, T. Even, A. Punzalan, S. Williams, M. D. Smith, A. K.
Knapp, How ecologists define drought, and why we should do better. Glob. Change Biol.
25, 3193–3200 (2019).
97. J. I. Christian, J. B. Basara, J. A. Otkin, E. D. Hunt, R. A. Wakefield, P. X. Flanagan, X.
Xiao, A Methodology for Flash Drought Identification: Application of Flash Drought
Correction (21 Frequency
January 2025):
across the United States. J. Hydrometeorol. 20, 833–846 (2019).
98. S. Beguería, V.-S. Sergio, SPEI: Calculation of the Standardized Precipitation-
Evapotranspiration Index (R package version 1.8.1, 2022); [Link]
[Link]/web/packages/SPEI/[Link].
99. S. Bachmair, C. Svensson, J. Hannaford, L. J. Barker, K. Stahl, A quantitative analysis to
objectively appraise drought indicators and model drought impacts. Hydrol. Earth Syst.
Sci. 20, 2589–2609 (2016).
100. H. Vernieuwe, B. De Baets, N. E. C. Verhoest, A mathematical morphology approach for a
qualitative exploration of drought events in space and time. Int. J. Climatol. 40, 530–543
(2020).
 101. B. Buchsbaum, Neuroim2: Data Structures and Handling for Neuroimaging Data (R package
version 0.1.0, 2022); [Link]
102. P. McCullagh, Generalized Linear Models (Routledge, 1989).
103. J. A. Gamon, C. B. Field, M. L. Goulden, K. L. Griffin, A. E. Hartley, G. Joel, J. Penuelas,
R. Valentini, Relationships Between NDVI, Canopy Structure, and Photosynthesis in
Three Californian Vegetation Types. Ecol. Appl. 5, 28–41 (1995).
104. J. Xue, B. Su, Significant Remote Sensing Vegetation Indices: A Review of Developments
and Applications. J. Sens. 2017, 1353691 (2017).
105. A. Bannari, D. Morin, F. Bonn, A. R. Huete, A review of vegetation indices. Remote Sens.
Rev. 13, 95–120 (1995).
106. Y. Zhou, L. Zhang, J. Xiao, S. Chen, T. Kato, G. Zhou, A Comparison of Satellite-Derived
Vegetation Indices for Approximating Gross Primary Productivity of Grasslands.
Rangeland Ecol. Manag. 67, 9–18 (2014).
107. L. Xu, A. Samanta, M. H. Costa, S. Ganguly, R. R. Nemani, R. B. Myneni, Widespread
decline in greenness of Amazonian vegetation due to the 2010 drought. Geophys. Res.
Lett. 38, L07402 (2011).
108. Q. Wang, Á. Moreno-Martínez, J. Muñoz-Marí, M. Campos-Taberner, G. Camps-Valls,
Estimation of vegetation traits with kernel NDVI. ISPRS J. Photogramm. Remote Sens.
195, 408–417 (2023).
109. M. Li, S. Cao, Z. Zhu, Z. Wang, R. B. Myneni, S. Piao, Spatiotemporally consistent global
dataset of the GIMMS Normalized Difference Vegetation Index (PKU GIMMS NDVI)
from 1982 to 2022. Earth Syst. Sci. Data 15, 4181–4203 (2023).
110. S. Huang, L. Tang, J. P. Hupy, Y. Wang, G. Shao, A commentary review on the use of
normalized difference vegetation index (NDVI) in the era of popular remote sensing. J.
For. Res. 32, 1–6 (2021).
111. M. Friedl, D. Sulla-Menashe, MODIS/Terra+Aqua Land Cover Type Yearly L3 Global
500m SIN Grid V061, (NASA EOSDIS Land Processes Distributed Active Archive
Center, 2022); [Link]
112. C. Liu, J. Liu, Q. Zhang, H. Ci, X. Gu, A. Gulakhmadov, Attribution of NDVI Dynamics
over the Globe from 1982 to 2015. Remote Sens. 14, 2706 (2022).
113.(21
Correction A. January
P. Williams, E. R. Cook, J. E. Smerdon, B. I. Cook, J. T. Abatzoglou, K. Bolles, S. H.
2025):
Baek, A. M. Badger, B. Livneh, Large contribution from anthropogenic warming to an
emerging North American megadrought. Science 368, 314–318 (2020).