The 2009/10 Drought in China: Possible Causes and Impacts on Vegetation

TLDR: In this paper, the authors reported that the severity of the 2009 and 2010 China drought was particularly severe in southwestern and northern China, where the accumulated precipitation from May 2009 to April 2010 was about 25% less than normal.

Abstract: Several provinces of China experienced an intense drought episode during 2009 and 2010. The drought was particularly severe in southwestern and northern China, where the accumulated precipitation from May 2009 to April 2010 was about 25% less than normal. The decline of accumulated precipitation over northern China was mostly noticeable during the summer season of 2009 and it was comparable to recent dry episodes. The southwestern China drought resulted from a sequence of dry months from summer 2009 to winter 2010, corresponding to the driest event since at least 1951. The suppression of rainfall in summer over both regions was in agreement with a weakened broad-scale South Asian summer monsoon, possibly influenced by an El Nino developing phase, whereas the extremely negative phases of the Arctic Oscillation during the winter of 2010 may have contributed to the persistence of the drought in southwestern China. The assessment of the associated impacts indicates that water reservoirs were severely ...

Figures

FIG. 3. Seasonal anomalies of geopotential height at 200 hPa (gpm, contours), vertically integrated moisture fluxes from surface to 300 hPa (kg m21 s21, arrows), and total water transport (kg m21 s21, shading) for (a) summer 2009, (b) autumn 2009, and (c) as in (a) but for winter 2010, and with shading indicating surface temperature anomalies (8C). Black thick arrows indicate the main climatological channels of water flux transport over the areas affected by the drought at the given season. Anomalies are relative to the 1951–2008 period.

FIG. 7. (a) Land-cover classification as obtained from the Global Land Cover 2000 database; (b) spatial distribution of the month with maximum annual mean NDVI, as obtained from monthly average SPOT–VEGETATION data for the period 1998–2010; and (c) distribution of vegetation over China, according to the vegetative cycle of each pixel. The boxes labeled as N and SW indicate the major regions affected by the drought. The boxes are similar (but not equal) to the ones defined in Fig. 1. Pixels corresponding to bare soil and sparse vegetation have been masked and represented in light gray.

FIG. 4. Standardized time series of (a) summer WYMI (1951–2009), (b) summer IMI (1951–2009), (c) autumn WNPMI (1951–2009), d) winter AO (1951–2010), and their correlation coefficient with regional seasonal precipitation in N and SW China. Only correlations significant at p , 0.1 level are shown. For simplicity, the seasonal times series are only shown until 2009 in (a), (b), and (c), and 2010 in (d), so the last bar in the series identifies the relevant value of the index during the drought episode.

FIG. 6. (a) Monthly time series of hydroelectric power output (1010 kW h21) in China for the period January 2003 to August 2010 (blue line). The red line shows the corrected time series after removing the multiplicative seasonality and linear trend. The bottom graphic shows the monthly anomalies (in percentage of the mean values for the period 2003–08). Gray shading identifies the drought episode (May 2009–April 2010). (b) As in Fig. 1 but for the period May 2006–April 2007.

FIG. 1. Spatial distribution of the accumulated monthly precipitation (in percentage relative to the 1951–2008 norm) during the hydrological year 2009/10 (i.e., between May 2009 and April 2010). To avoid misleading results over dry areas, only grid points with climatological accumulated precipitation above 240 mm (i.e., 20 mm month21) are shown. Black boxes approximately delimit regions with maximum deficit in accumulated precipitation and they are referred to as N and SW China in the text.

FIG. 9. Climatological (1998–2010) annual cycles of spatially averaged NDVI over farmland pixels affected by more than 4 months of vegetation stress (dotted line, right vertical axis) and the 2009/10 monthly time series of the weighted NDVI anomalies (bars, left vertical axis) averaged over (top) N, (middle) SW, and (bottom) CE China (see Fig. 8 for the definition of these regions).

Full text

The 2009/10 Drought in China: Possible Causes and Impacts on Vegetation
DAVID BARRIOPEDRO
IDL, Universidade de Lisboa, Lisbon, Portugal, and Departamento de Astrofı´sica y Ciencias de la Atmo
´
sfera/Instituto de
Geociencias, Universidad Complutense de Madrid, Madrid, Spain
CE
´
LIA M. GOUVEIA
IDL, Universidade de Lisboa, Lisbon, and Escola Superior de Tecnologia, Instituto Polite
´
cnico de Setu
´
bal, Setubal, Portugal
RICARDO M. TRIGO
IDL, Universidade de Lisboa, and Departamento de Engenharias, Universidade Luso
´
fona, Lisbon, Portugal
LIN WANG
Center for Monsoon System Research, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
(Manuscript received 22 June 2011, in final form 16 February 2012)
ABSTRACT
Several provinces of China experienced an intense drought episode during 2009 and 2010. The drought was
particularly severe in southwestern and northern China, where the accumulated precipitation from May 2009
to April 2010 was about 25% less than normal. The decline of accumulated precipitation over northern
China was mostly noticeable during the summer season of 2009 and it was comparable to recent dry episodes.
The southwestern China drought resulted from a sequence of dry months from summer 2009 to winter 2010,
corresponding to the driest event since at least 1951. The suppression of rainfall in summer over both regions
was in agreement with a weakened broad-scale South Asian summer monsoon, possibly influenced by an
El Nin
˜
o developing phase, whereas the extremely negative phases of the Arctic Oscillation during the winter
of 2010 may have contributed to the persistence of the drought in southwestern China. The assessment of the
associated impacts indicates that water reservoirs were severely affected with a ;20% reduction in the na-
tionwide hydroelectrical production during the drought event. Furthermore, an analysis of the normalized
difference vegetation index data reveals that large cropland sectors of northern and eastern China experi-
enced up to 8 months of persistently stressed vegetation between May 2009 and July 2010, while southwestern
China was relatively less affected. Such different regional vegetative responses are interpreted in terms of
the land-cover type, agriculture management, and their dependence on seasonal precipitation and water
availability for irrigation.
1. Introduction
During 2009 and 2010, an intense and prolonged
drought episode affected several provinces of north-
ern (Hebei, Shanxi, and Liaoning) and southwestern
(Yunnan, Guizhou, Guangxi, and Sichuan) China, in-
cluding the municipalities of Beijing and Chongqing
(Qiu 2010). Northern China is an important area of grain
production, while the southwest region represents one
of the major water resources of the country, supplying
water to the Yangtze River, the Pearl River head, and
their reservoirs. According to the Chinese Office of State
Flood Control and Drought Relief Headquarters and
the Ministry of Civil Affairs, the drought affected ;25
million people and ;18 million livestock by shortage of
drinking water, as well as about ;8 million ha of arable
land nationwide, with total economic losses of at least
$3.5 billion U.S. dollars (http://www.chinadaily.com.cn/
china/2010drought/index.html).
Droughts are recurrent phenomena in China, but
they can be associated to different causes and complex
Corresponding author address: David Barriopedro, Instituto
Dom Luiz, Universidade de Lisboa, Departamento de Engenharia
Geogra
´
fica, Geofı
´
sica e Energia, Faculdade de Cie
ˆ
ncias, Univ. de
Lisboa, Campo Grande, Ed C8, Piso 6, 1749-016 Lisboa, Portugal.
E-mail: dbarriopedro@fc.ul.pt
A
UGUST 2012 B A R R I O P E D R O E T A L . 1251
DOI: 10.1175/JHM-D-11-074.1
Ó 2012 American Meteorological Society

processes. For example, the recently severe drought epi-
sode during winter 2008/09 in eastern China was shown
to be partially driven by the 2008/09 La Nin
˜
a episode
and also influenced by the anomalous high tempera-
tures over the Tibetan Plateau (Gao and Yang 2009).
Interannual rainfall variations in China are also asso-
ciated with the Asian monsoon—a complex and multi-
faceted system encompassing the East Asian monsoon
(EAM), the Indian monsoon (IM), and the western
North Pacific monsoon (WNPM) (e.g., Wang et al. 2001;
Wang and LinHo 2002). The EAM (e.g., Ding and Chan
2005) is characterized by a cold, dry East Asian winter
monsoon (EAWM) affecting Japan and southeastern
China (e.g., Zhang et al. 1997) and a warm, rainy East
Asian summer monsoon (EASM), whose main signature
is a meridional rainfall tripole and the quasi-stationary
mei-yu subtropical front (e.g., Chen et al. 2004). Fluctu-
ations in the EASM are often associated with floods
and droughts in the Yangtze River basin (e.g., Huang
et al. 2007). Relatively, other regions of the country are
more affected by interactions between the different sub-
systems of the Asian monsoon, midlatitude processes,
and tropical cyclones (e.g., Wang et al. 2008), which
in turn are subject to remote influences. In particular,
El Nin
˜
o–Southern Oscillation (ENSO) explains a sig-
nificant part of the interannual Asian monsoon vari-
ability by displacing the heat sources in the tropics
(Webster and Yang 1992; Wang et al. 2001), although
such relationship is nonstationary (Kumar et al. 1999;
R. Wu and B. Wang 2002). Other reported factors af-
fecting seasonal precipitation in China are the Arctic
Oscillation/North Atlantic Oscillation (AO/NAO; Gong
and Wang 2003; Sung et al. 2006), the stationary plan-
etary waves (Chen et al. 2005), the Antarctic Oscillation
(AAO; Nan and Li 2003), dynamic and thermal effects of
the Tibetan Plateau (Wu and Zhang 1998; Hsu and Liu
2003), and the Eurasian snow cover (Zhang et al. 2004).
In addition to the assessment of physical mechanisms
associated with extreme events, it is of equal relevance
to quantify their impacts. This is a particularly daunting
task in China, where the area influenced by droughts
reaches nearly 20 million km
2
on average every year,
accounting for about
1
/
6 of the national arable land
(Huang et al. 2000). Given the key role of agriculture
(;13% of China’s GDP) sustaining the huge number of
Chinese people (;1.3 billion), droughts can cause large
economic losses in China (circa 50% of the total losses
induced by climatic hazards on average; Huang et al.
2000). With the advent of satellite-derived products, the
monitoring of droughts and the related assessment of
vegetation health and net primary production have
experienced significant improvements (e.g., Kogan
1995; Fang et al. 2003). Studies based on the normalized
difference vegetation index (NVDI) over China have
shown a nationwide enhancement of vegetation cover-
age during recent decades, which is not only due to the
advance of agricultural technology but also to the re-
gional precipitation and temperature changes that have
caused an extended growing season and more rapid plant
growth (Piao et al. 2010). This general trend in NDVI is,
however, punctuated by severe droughts with spatially
coherent NVDI anomaly patterns (Habib et al. 2009),
which have become relatively frequent and widespread
in northern China since the 1970s (e.g., Zou et al. 2005).
The application of NDVI as a tool to assess droughts’
impacts in agriculture production is also of great interest
in China because the large variety of climate regimes
and crops, together with an uneven distribution of water
resources and agriculture management, make that a given
combination of climate anomalies can produce either
beneficial, irrelevant, or damaging effects on vegetation
(e.g., Tao et al. 2008).
The main goal of this paper is to provide a descrip-
tion of the 2009/10 drought in China from multiple
perspectives by addressing some of the aforementioned
issues. In more detail, the objectives are (i) to charac-
terize the temporal and spatial extent of this extreme
drought event, (ii) to assess the regional precipitation
deficits in terms of the associated atmospheric circulation
anomalies, and (iii) to evaluate some socioeconomic
impacts of this extreme drought episode, particularly on
vegetation greenness and hydroelectrical production.
The data and methods employed in this study are
described in section 2. Section 3 analyzes the spatial and
temporal characteristics of this drought. The associated
anomalous atmospheric circulation is presented in sec-
tion 4. Section 5 then describes the social–economic im-
pacts of this drought, including the impacts on national
hydroelectrical production and on vegetation. Finally, the
main conclusions are presented in section 6.
2. Data and methods
Monthly precipitation data on land surface at 1.083
1.08 resolution grid were provided by the Global Precipi-
tation Climatology Centre (GPCC; Rudolf and Schneider
2005). This dataset has already been used to analyze ex-
treme dry precipitation episodes over different regions,
such as Iberia (Garcı
´
a-Herrera et al. 2007) and the Mid-
dle East (Trigo et al. 2010). The GPCC products are
basedoninsituraingaugedataandcovertheperiod
from 1901 to present through two gridded datasets: the
so-called full data product (1901–2009), which is contin-
uously updated with all available stations, and the mon-
itoring product (2007–present) that only uses a network
of near-real-time stations. Both datasets are subject to
1252 JOURNAL OF HYDROMETEOROLOGY VOLUME 13

similar processing and quality-control steps but they dif-
fer in the number of underlying stations. A preliminary
analysis revealed high consistence between these two
GPCC products during their overlapping period (2007–
09). Therefore, the monitoring product for 2010 was
added to the full data product (available until the end of
2009) to provide full temporal coverage of the drought
episode (2009/10). Since the number of stations over
China experiences a significant decrease before the 1950s,
the period of analysis has been limited to 1951–2010. To
better characterize regional features of the drought ep-
isode, regional precipitation averages have been com-
puted over northern (N) and southwestern (SW) China
(see Fig. 1) after excluding grid points with very dry cli-
mates (e.g., inner Mongolia) and/or at high elevation sites
(e.g., southeasternmost sector of the Tibetan Plateau).
The spatial domain of these sectors is further supported
by a clustering analysis based on observational precipi-
tation stations in China (Song et al. 2007).
The atmospheric data include geopotential height,
temperature, humidity, and zonal and meridional wind
data at different pressure levels obtained from the Na-
tional Centers for Environmental Prediction–National
Center for Atmospheric Research (NCEP–NCAR) re-
analysis in a 2.5832.58 regular grids and for the period
1951–2010 (Kistler et al. 2001). The vertically inte-
grated moisture flux (from the surface to 300 hPa) was
computed following the approach adopted in Trenberth
and Guillemot (1995). This field displayed almost identical
features to the water vapor flux at low levels (e.g., 850 hPa)
for the analyzed period of the drought and it also showed
strong resemblance with the low-level wind, except over
relatively dry land areas of midlatitudes. With the aim
of assessing remote influences in the regional precip-
itation variability, we have used atmospheric circula-
tion indices of ENSO [measured by the El Nin
˜
o 3.4 index
(EN3.4)] and the AO for the 1951–2010 period, as pro-
vided by the Climate Prediction Center of the National
Oceanic and Atmospheric Administration (NOAA; www.
cpc.ncep.noaa.gov/).
Additionally, taking into account the regional mani-
festations of the Asian monsoon, a suite of distinct win-
ter and summer monsoon indices have been computed
from the NCEP–NCAR reanalysis for the period 1951–
2010. The summer indices include (i) the Webster–Yang
Asian Monsoon Index (WYMI; Webster and Yang 1992),
(ii) the Western North Pacific Summer Monsoon Index
(WNPMI; Wang and Fan 1999), and (iii) the Indian
Monsoon Index (IMI; Wang and Fan 1999). In what
concerns the winter season, the strength of the EAWM
was characterized by the East Asian Winter Monsoon
Index (EAWMI) proposed by Jhun and Lee (2004).
Seasonal values are computed from the seasonal mean
fields and then standardized with reference to the 1951–
2008 period. The WYMI reflects the broad-scale South
Asian summer monsoon variability, which is primarily
driven by two relatively independent convective heat
sources over the Bay of Bengal and the Philippine Sea
(Wang and Fan 1999), herein represented by the IMI
and WNPMI, respectively. Although the WNPM is not
exactly the same system as EASM, it is closely related to
the variations of mei-yu and therefore is able to capture
the main features of the EASM (Wang et al. 2008). The
sign of WNPMI used here is reversed compared with the
original index proposed by Wang and Fan (1999) so that
positive value indicates an enhanced mei-yu rainfall.
Other summer and winter monsoon indices (see Wang
et al. 2008 and Wang and Chen 2010a) may capture better
regional precipitation anomalies, but for the sake of
simplicity, the discussion will focus on some of the most
traditional indices that are of standard use in operational
centers.
The response of vegetation was assessed with fields
of NDVI since a close relationship between NDVI and
annual rainfall variations has been reported for different
land-cover types in China (Fang et al. 2001). Meanwhile,
it should be noted that water intensive crops (e.g., rice
and cotton) and areas with multicropping rotations are
more sensitive to seasonal rains, springtime snow, and
glacier melt and water resources (Piao et al. 2010). The
NDVI can also reflect temperature anomalies since low
minimum temperatures during the growing season cause
FIG. 1. Spatial distribution of the accumulated monthly precipi-
tation (in percentage relative to the 1951–2008 norm) during the
hydrological year 2009/10 (i.e., between May 2009 and April 2010).
To avoid misleading results over dry areas, only grid points with
climatological accumulated precipitation above 240 mm (i.e.,
20 mm month
21
) are shown. Black boxes approximately delimit
regions with maximum deficit in accumulated precipitation and
they are referred to as N and SW China in the text.
A
UGUST 2012 B A R R I O P E D R O E T A L . 1253

damage to crops (e.g., Tao et al. 2008). NDVI values were
derived from the so-called 10-day global synthesis
(S10) products of the SPOT-VEGETATION partner
(VITO) database (http://free.vgt.vito.be), which provide
atmospherically corrected and geometrically calibrated
data acquired by the VEGETATION instrument on board
both Satellite Pour l’Observation de la Terre (SPOT)
SPOT-4 and SPOT-5 satellites between September 1998
and August 2010 (Maisongrande et al. 2004; Hagolle et al.
2005). NDVI fields are supplied at the resolution of
0.0089288 (i.e., about 1-km
2
resolution at the equator) in
geographic coordinates and on a 10-day basis, following
the maximum value composite (MVC) method, which
selects the date of maximum NDVI among 10 consec-
utive daily images (Holben 1986). All months are divided
in three 10-day values, the first one corresponding always
to the first 10-day period of that month. For months
with 31 days and February (28 or 29 days), the last 10-
day value actually represents the maximum NDVI within
slightlylonger(11day)orshorter(8or9day)periods.
Monthly fields of the NDVI were subsequently derived
by averaging at each pixel the three 10-day values of the
given month.
It should be noticed that NDVI values tend to un-
derestimate the green biomass of stands over areas with
strong foliage density and high production (Hobbs 1995;
Tucker et al. 1986; Gilabert et al. 1996), causing satu-
ration of NDVI. Despite this caveat, which is expected
to cause relatively lower impacts in the assessment of
droughts, the time series of MVC–NDVI composites
have proven to be a source of valuable information for
monitoring surface vegetation dynamics at the global
and the regional scales (Zhou et al. 2001; Nemani et al.
2003; Gouveia et al. 2008). Therefore, despite the rela-
tively short period analyzed herein, the use of this da-
taset is preferred because the high spatial resolution is
expected to bring a more accurate assessment of the
drought impacts in land-cover types. Information about
the land-cover type associated to each pixel was obtained
from the Global Land Cover 2000 database (http://bioval.
jrc.ec.europa.eu/products/glc2000/products.php).
3. Spatial and temporal characteristics
of the drought
The precipitation regime in China is characterized by
a strong seasonal behavior, with a rainy season mostly
concentrated between May to September and a large
spatial gradient in the annual totals between the wet
monsoonal areas of the southeast and the semiarid re-
gime of northwest (e.g., Song et al. 2007). Therefore, the
hydrological year used in this study is defined as the
period spanning between May of year (yr) and April of
the next year (yr11). Meanwhile, the amount and tim-
ing of the annual maximum of monthly precipitation
varies across China following the penetration and north-
western advance of the Asian monsoonal flows. Thus,
SW China experiences a long-lasting and intense sum-
mer monsoon, with a broad precipitation peak of
;200 mm month
21
centered in June–July that contrasts
with the relatively narrower, delayed (July–August), and
less intense (150 mm month
21
) maximum of N China
(see Figs. 2b,c). However, when annual totals are con-
sidered N China shows a stronger dependence on sum-
mer rainfall, with July–August precipitation accounting
for more than half of the annual totals. This implies that
dry summer conditions in N China are expected to be
reflected on the accumulated totals, while a dry summer
monsoon in SW can be compensated by near-normal
precipitation in other seasons.
To better assess the drought episode, the spatial pat-
tern of accumulated precipitation percentages for the
hydrological year (May 2009–April 2010) has been com-
puted with respect to the corresponding climatological
(1951–2008) norms (Fig. 1). The drought episode was
particularly severe over N and SW China, with both re-
gions recording nearly 25% below the climatological
normal. In addition, relatively less severe drought condi-
tions were observed in other territories, including north-
eastern India and Burma; northern parts of Thailand,
Laos, and Vietnam (not shown); as well as other prov-
inces of southern China such as Hunan and Guangdong
(Fig. 1).
Figure 2a shows the accumulated monthly precipi-
tation averaged over N and SW areas between May 2009
and April 2010, along with the corresponding evolution
of the climatological accumulated monthly precipi-
tation distribution (whiskers plot). At the end of the
considered period, the regional average accumulated
precipitation in SW (;850 mm) was ;25% less than
the long-term mean value (;1100 mm), corresponding
to the lowest accumulated value in any hydrological year
since at least 1951. In relative terms, the N region ex-
perienced a more severe decrease (;30%) in accumu-
lated totals (;350 mm relative to a ;470 mm mean).
However, this region is also characterized by large in-
terannual variability, with the standard deviation (SD)
of annual accumulated totals (69.7 mm) being of similar
magnitude to that in SW (76.2 mm). Consequently, from
a normalized point of view, the 2009/10 drought was
more severe in SW (23.1 SDs) than in N (21.7 SDs). It
is worth noticing that, despite its extreme magnitude,
the 2009/10 drought episode over N China was not
particularly anomalous in the context of recent years
(Fig. 2a), which have been characterized by a relative
recurrence of extreme droughts. This is in agreement
1254 JOURNAL OF HYDROMETEOROLOGY VOLUME 13

with a drying climate trend in northern China (e.g., Hu
et al. 2003) and the increasing tendency for N China to
suffer from severe and long-lasting droughts, which were
particularly acute from 1997/98 to 2002/03 (Zou et al.
2005; Wang et al. 2011).
To better assess the months that specifically contrib-
uted to the marked decline in accumulated precipitation
within the hydrological year, Figs. 2b and 2c show the
temporal evolution of the regional averaged precipi-
tation monthly series during 2009 and 2010 in N and SW
China, respectively. The period with deficit of precipi-
tation started in May 2009 and lasted until October 2009
in N China, further persisting until March 2010 in SW
China. The winter and spring of 2010 were also dry in
the middle and the lower reaches of the Yellow and
Yangtze Rivers (not shown), but slightly wet in N China
(Fig. 2b). A near-normal 2010 monsoon season in SW (N)
China contributed to alleviate (mitigate) the persistent
deficits accumulated throughout the previous hydrologi-
cal year.
Compared with the historical situation, every month
of the May–September monsoon period of 2009 (except
June) exhibited precipitation deficits below the 30th
percentile of the long-term distribution in both N and
SW China (Figs. 2b,c). However, the drought in these
two areas exhibits different features. Thus, in relative
terms, the precipitation deficit over SW China was more
severe (below the 10th percentile of its climatologi-
cal distribution) from September to November 2009
(i.e., after the peak time of local monsoon rainfall). In
contrast, N China received the lowest precipitation
(also below the monthly 10th percentile) during July–
September 2009, which is the peak time of local mon-
soon rainfall. As a consequence, the drought period over
N China was shorter and mainly caused by decreased
summer precipitation, whereas that over the relatively
wetter SW China resulted from the persistence of dry
conditions during several consecutive months.
FIG. 2. (a) Accumulated monthly precipitation averaged over N
and SW China during the hydrological year 2009/10 (black dotted
lines). Gray shaded areas show the overlapping from the corre-
sponding evolution from all grid points embedded in the given area
(defined by the black boxes of Fig. 1). Gray lines indicate the cli-
matological mean evolution in each region, with boxes (whiskers)
representing the 60.5sigma level (10th–90th percentiles) obtained
from all hydrological years between 1951 and 2008. The other two
most severe drought events in each region within the period 1951–
2008 are also indicated with color lines and ranked in the upper-left
corner. (b),(c) Climatological (1951–2008) annual cycle of monthly
precipitation averaged over (b) N and (c) SW China. Two full
annual cycles are shown. The light (dark) shaded areas comprise
the 10th–90th (30th–70th) monthly percentiles obtained from the
1951–2008 precipitation time series, with the median in between.
Dashed line shows the time series for 2009 and 2010, with the
corresponding monthly departure from the climatological mean
being represented in the bottom graphic.
A
UGUST 2012 B A R R I O P E D R O E T A L . 1255

Frequently asked questions

Q1. What was the climatological normal for the two regions?

The drought episode was particularly severe over N and SW China, with both regions recording nearly 25% below the climatological normal. 

Q2. What is the effect of the lack of precipitation on the growing season of 2010?

successive months with lack of precipitation resulted in pronounced anomalies in theaccumulated precipitation totals (Fig. 2a), thereby affecting the growing season of 2010 crops (Fig. 9b) by presumably reducing soil moisture and impeding the germination of seeds. 

Q3. What was the period with deficit of precipitation in 2009/10?

The period with deficit of precipitation started in May 2009 and lasted until October 2009 in N China, further persisting until March 2010 in SW China. 

Q4. How many months of drought were affected by the cultivated–managed areas?

cultivated–managed areas lying between the estuaries of the Yellow and Yangtze Rivers (box labeled as CE in Fig. 8) also experienced several months of persistently stressed vegetation, as discussed below. 

Q5. What is the effect of low temperatures on the NDVI?

The NDVI can also reflect temperature anomalies since low minimum temperatures during the growing season causedamage to crops (e.g., Tao et al. 2008). 

Q6. What is the climatological normal for the accumulated rainfall in China?

when annual totals are considered N China shows a stronger dependence on summer rainfall, with July–August precipitation accounting for more than half of the annual totals. 

Q7. What is the effect of the reduction of storage water in damns?

The reduction of storage water in damns was also potentially responsible for negative responses in downstream croplands of the Yangtze and Pearl Rivers. 

Q8. What is the spatial domain of these sectors?

The spatial domain of these sectors is further supported by a clustering analysis based on observational precipitation stations in China (Song et al. 2007). 

Q9. What are the objectives of this paper?

In more detail, the objectives are (i) to characterize the temporal and spatial extent of this extreme drought event, (ii) to assess the regional precipitation deficits in terms of the associated atmospheric circulation anomalies, and (iii) to evaluate some socioeconomic impacts of this extreme drought episode, particularly on vegetation greenness and hydroelectrical production. 

Q10. What is the effect of the accumulated precipitation on the valleys of the major rivers?

the negative impacts on the valleys of the major rivers, where the accumulated precipitation did not show remarkable deficiencies, is well explained by the dominance of multicropping rotations within the year, water intensive crops, and their stronger sensitivity to precipitation in other seasons than summer. 

Q11. What is the climatological norm for the accumulated rainfall in N China?

This implies that dry summer conditions in N China are expected to be reflected on the accumulated totals, while a dry summer monsoon in SW can be compensated by near-normal precipitation in other seasons. 

Q12. What was the effect of the drought on the cultivated land in N China?

As discussed in section 3, the drought episode in N (SW) China was characterized by stronger deficits of precipitation during the peak (post) monsoon season of 2009. 

Q13. What is the effect of the lack of precipitation on the vegetation in the summer?

Taking into account the dependence of crops to growing season precipitation, the earlier deficit of summer precipitation in N China (as compared to its climatological maximum in vegetative activity) would explain the negative impacts in summer farmlands vegetation greenness of this area.