Research ProposalPDF Available

Long Term Rainfall Trend over Meteorological Sub Divisions and Districts of India

Authors:
MAUSAM, 68, 3 (July 2017), 439-450
551.585.3 (540)
(439)
Long term rainfall trend over meteorological sub divisions and districts of India
SURINDER KAUR, SUMANT KUMAR DIWAKAR and ASHOK KUMAR DAS
India Meteorological Department, New Delhi110 003, India
(Received 13 October 2016, Accepted 10 February 2017)
e mail : surinderkaur.imd@gmail.com
सार  , 

 
 1901 - 2013  


, 

       

        -  
(0.25º × 0.25º)  632 34     
   
, 10% 13%
(  )  
 10%    8%       




ABSTRACT. In this paper the long term trend of annual and seasonal rainfall over different districts,
Meteorological (Met.) sub-divisions and whole India have been studied using the long term rainfall data for the period
from 1901 to 2013. The changes in amount and pattern of rainfall have a significant impact on agriculture, water
resources management and overall economy of the country. Mann-Kendall test is applied to check the significance of the
trend. Linear Regression and Theil-Sen’s non parametric test has been applied to estimate the trend. The study is carried
out for 632 districts and 34 sub divisions of India by utilizing the gridded rainfall data (0.25° × 0.25°) over the main land
except Andaman & Nicobar and Lakshadweep islands. Many authors have studied that extreme events are increasing but
there is no trend in Pan India’s rainfall. It is observed from the annual rainfall analysis 10% of the number of districts are
showing significant increasing trend and 13% significant decreasing (mainly in Uttar Pradesh) trend whereas irrespective
of high and low rainfall regions, 10% area of the country is showing significant increasing trend and 8% of the area of the
country showing significant decreasing trend in annual rainfall. In Meteorological Sub divisions, east & west UP are
showing significant negative trend and some of the coastal sub divisions are showing positive trend. It is also observed
that the country’s rainfall is not showing any trend.
Key words Rainfall trend, CV, Meteorological sub-division, Theil-Sen, Mann-Kendall test.
1. Introduction
The annual rainfall over India is highly variable in
space and time. More than 75% of rainfall received during
southwest (SW) monsoon period (June to Sept) which
results in scarcity of water in many parts of the country
during the non-monsoon periods, although large number,
around 5000 large hydraulic structures have been
constructed to store the water to use for drinking and
irrigation purposes. The water requirement in India is
increasing; it will be almost double from the year 2000 to
2050. The problem may be aggravated due to global
climate changes which may influence long-term rainfall
patterns impacting the availability of water, along with the
increasing number of extreme rainfall events results in
occurrences of droughts and floods. Changes in climate
over the Indian region, particularly during the SW
monsoon, would have a significant impact on agricultural
production, water resources management and overall
economy of the country.
Water resource has become a prime concern for any
development and planning including food production,
flood control and effective management of water resource.
IPCC (2007) reported that global surface warming is
occurring at a rate of 0.74 ± 0.18 °C in period 1906-2005.
Parry et al. (2007) has also reported that the IPCC has
estimated that even if the concentration of all greenhouse
440 MAUSAM, 68, 3 (July 2017)
Figs. 1(a&b). Map showing no. of stations per Grid for year (a) 1901 & (b) 2013
gases and aerosols will be kept constant at the year 2000
levels, a further warming of about 0.1 °C per decade
would be expected. The impact would be particularly
severe in the tropical areas, and mainly in developing
countries, including India. Pant and Kumar (1997) have
reported an increase in mean annual temperature in India
at the rate of 0.57 °C in the last century.
The impact of climate change in the future is quite
severe as given by IPCC reports which signify that there
will be reduction in the freshwater availability because of
climate change. This has also been revealed that by the
middle of the twenty-first century, decrease in annual
average runoff and availability of water will project up to
10-30%.
Many authors have studied trend in precipitation
over different parts of India in different seasons and the
results shows some places it is increasing and some places
decreasing or no trend. Guhathakutha and Rajeevan
(2007) have studied the rainfall trend over India and
concluded that Indian monsoon rainfall as a whole does
not show any significant trend but trends are observed
over some specific areas. During the southwest monsoon
season, three subdivisions viz., Jharkhand, Chhattisgarh,
Kerala show significant decreasing trend and eight
subdivisions viz., Gangetic WB, West UP, Jammu &
Kashmir, Konkan & Goa, Madhya Maharashtra,
Rayalseema, Coastal AP and North Interior Karnataka
show significant increasing trends. Gajbhiye et al. (2016)
have studied trend analysis over Sind basin and found
increasing trend in seasonal and annual series during
1901-2002. Various studies have been carried out over
Himalayan region. Pant et al. (1999) have found that
seasonal and annual rainfall did not show any significant
trend over Western Himalaya during the period 1893-
1990. Archer and Fowler (2004) found a non-significant
trend (10% per 100 years) over Srinagar. Kumar et al.
(2005) have found a slight downward trend in monsoon
rainfall and a slight upward trend in winter rainfall over
Himachal Pradesh during 1964-1992. Preethi et al. (2016)
have studied the recent trends over South and East Asian
summer monsoons. They have identified two regions over
India namely northeast India showing decreasing trend
and increasing over the northern part of the West coast of
India during SW monsoon.
As mentioned above, many authors have
studied trend in precipitation over different parts of
India or Pan India on a state, sub divisional or river
basin scale. In this paper, the long term trend in
precipitation has been studied on a district scale for the
whole India using the data from 1901 to 2013. There
are significant developments in recent years, so the
trend analysis is also carried out for the recent period
1961-2013.
(a) (b)
KAUR et al.: LONG TERM RAINFALL TREND OVER METEOROLOGICAL SUB DIVISIONS 441
Fig. 2. The India’s rainfall for the period 1901-2013
Fig. 3. Decadal % rainfall departure and CV of India’s rainfall.
2. Data Used
The daily gridded rainfall data Pai et al. (2014) at a
high spatial resolution (0.25° × 0.25) of a longer period of
113 years (1901-2013) over the Indian main land has been
used in this study. The gridded data has been prepared by
IMD using the daily rainfall records from 6995 rain gauge
stations in India after making quality control of basic rain-
gauge stations. The gridded rainfall is computed for main
land area excluding Andaman & Nicobar (AN) and
Lakshadweep (LK) islands. For the preparation of the
gridded data for each day of the data period, on an
average, about 3500 stations that varied between 1450 &
3900 were used. For the interpolation, inverse distance
weighted interpolation (IDW) scheme of Shepard (1968)
was used. The districts rainfalls have been computed for
632 districts from the gridded rainfall data by simple
average over the rainfall at the grids lying in the district.
The sub divisional rainfall has been computed for 34 sub
divisions using the weighted average rainfall of the
districts and the country’s rainfall by the weighted rainfall
of the sub divisions. The number of stations per grid used
in preparation of gridded rainfall is given in Figs. 1(a&b).
Fig. 4. District wise annual rainfall trend for the period 1901-2013
3. Methodology
The Mann-Kendall and Sen’s slope estimator has
been used for the determination of the trend.
3.1. Significance of trend
The Mann-Kendall test [Burn et al., 2004; Douglas
et al., 2000; Libiseller and Grimval, 2002; Singh et al.,
2008a; Yu et al., 1993; Yue and Hashino, 2003] is used to
detect monotonic (increasing or decreasing) trends and is
widely used for detecting trends in time series, because it
is simple and robust accommodates missing values and
the data need not conform to any statistical distribution.
Since there are chances of outliers to be present due to
extreme rainfall events, the non-parametric Mann-Kendall
test is useful, because its statistic is based on the (+ or )
signs, rather than the values of the random variable and
therefore, the trends determined are less affected by the
outliers [Birsan et al., 2005; Helsel & Hirsch, 1992].
Trend analysis using the MK test to detect trend in
the rainfall time series, we proceeded as follows:
Assuming that the rainfall time series was
independent, and then the Mann-Kendall statistic S was
computed as:
442 MAUSAM, 68, 3 (July 2017)
TABLE 1
State wise number of districts showing increasing and decreasing trend
S. No. State & UT Total n
o. of
Districts
No. of Districts (1901-2013) No. of Districts (1961-2013)
Increasing trend Decreasing trend Increasing trend Decreasing trend
1. Arunachal Pradesh 16 0 8 1 7
2. Andhra Pradesh 13 6 0 4 0
3. Assam 27 1 15 2 9
4. Bihar 38 1 13 4 2
5. Chandigarh 1 0 1 0 0
6. Chhattisgarh 18 1 12 1 0
7. Delhi 8 0 0 0 8
8. Dadra & Nagar Haveli and Daman & Diu 3 2 0 0 1
9. Goa 2 1 0 2 0
10. Gujarat 26 7 0 4 0
11. Haryana 22 2 0 0 5
12. Himachal Pradesh 12 3 3 0 1
13. Jammu & Kashmir 14 3 4 6 2
14. Jharkhand 24 0 7 3 1
15. Karnataka 30 8 2 6 1
16. Kerala 14 0 9 1 3
17. Madhya Pradesh 52 5 12 0 3
18. Maharashtra 35 8 6 1 0
19. Manipur 10 3 0 4 0
20. Meghalaya 7 5 1 1 2
21. Mizoram 8 2 4 5 0
22. Nagaland 11 0 10 0 0
23. Odisha 30 4 4 12 0
24. Puducherry 4 0 1 0 1
25. Punjab 21 5 0 2 0
26. Rajasthan 33 2 0 0 0
27. Sikkim 4 1 3 1 2
28. Tamil Nadu 32 2 5 1 1
29. Telangana 10 2 0 0 0
30. Tripura 4 1 2 0 0
31. Uttarakhand 13 0 6 0 4
32. Uttar Pradesh 72 0 27 0 32
33. West Bengal 18 7 2 3 0
Total 632 82 (13%) 156 (24.7%) 64 (10.1%) 85 (13.4%)
S = signXj-Xi
1≤i≤n-1
i+1<j<n
(1)
where, and are sequential data for the ith and jth
terms; n is the sample size; and
sign (Xj-Xi) = +1; if Xj-Xi>1
0; if Xj-Xi =0
-1; if Xj-Xi<1 (2)
The statistic S is approximately Gaussian when
n = 18 with the mean E(S) and variance Var (S) of the
statistic S is given by:
E(S) = 0 and Var (S) =
18 )52)(1( +nnn
(3)
However, if ties exist in the dataset, then the
expression for Var(S) has to be adjusted and becomes :
Var (S)= n(n-1)(2n+5)-tp (tp-1) (2tp+5)
q
p=1
18 (4)
The variable q and tp are number of tied group’s and
number of data values in the pth group, respectively. The
standardized statistic (Z) for one-tailed test of the statistic
S is given as follows:
KAUR et al.: LONG TERM RAINFALL TREND OVER METEOROLOGICAL SUB DIVISIONS 443
TABLE 2
Districts with increasing trend in annual rainfall during the period 1961-2013
State Name of Districts Area (sq. km) %
Arunachal Pradesh West Kameng 5510.0 0.1683
Andhra Pradesh Guntur, Krishna, Prakasam, Vishakhapatnam 49634.9 1.5161
Assam Baksa, Barpeta 4725.6 0.1443
Bihar
Bhagalpur, Katihar, Munger, Purnia
10255.9
0.3133
Chhattisgarh Dantewara 8598.0 0.2626
Goa North Goa, South Goa 3718.2 0.1136
Gujarat Ahmedabad, Junagarh, Rajkot, Surendra Nagar 38704.6 1.1823
Jammu And Kashmir Baramula, Mirpur, Punch, Riasi, Udhampur 35030.4 1.0700
Jharkhand Purbi Singbhum, Sahebganj, Saraikela 8441.9 0.2579
Karnataka Chikmagalur, Chitradurga, Davangere, Kolar, Mandhya, Uttar Kannada 41002.9 1.2525
Kerala Idukki 4384.9 0.1339
Maharashtra Pune 15728.0 0.4804
Manipur Bishnupur, Churachandpur, Imphal East 5807.1 0.1774
Meghalaya South Garo Hills 1929.7 0.0589
Mizoram Aizawl, Champhai, Kolasib, Lawngtlai, Serchhip 11624.1 0.3551
Orissa
Baragarh, Cuttack, Kalahandi, Kendujhar, Khordha, Koraput, Malkangiri,
Mayurbhanj, Nabarangapur, Nayagarh, Puri, Subarnapur 68846.9 2.1030
Punjab Patiala, Sas Nagar 3731.6 0.1140
Sikkim East Sikkim 957.7 0.0293
Tamil Nadu Teni 2882.5 0.0880
West Bengal Bankura, Darjeeling, Puruliya 16326.4 0.4987
Total
10.3197
TABLE 3
Districts with decreasing trend in annual rainfall during the period 1961-2013
State Name of Districts Area (sq. km) %
Arunachal Pradesh
West Siang 37673.1 1.1508
Assam Dhemaji, Dibrugarh, Golaghat, Kamrup Metro, Karbi Analog, Lakhimpur,
N.C. Hills, Tinsukia 34098.7 1.0416
Bihar Kishanganj, Siwan 4222.7 0.1290
Delhi Central Delhi, East Delhi, North Delhi, Ne Delhi, Nw Delhi, South Delhi, Sw
Delhi, West Delhi 1489.6 0.0455
Daman & Diu Diu 28.2 0.0009
Harayana Gurgaon, Jhajjar, Karnal, Panipat, Sonipat 9048.4 0.2764
Himachal Pradesh Shimla 5138.0 0.1569
Jammu And Kashmir Kathua, Tribal Territory 5202.1 0.1589
Jharkhand Palamu 4408.8 0.1347
Karnataka Dharwad 4281.2 0.1308
Kerala Kozikod, Palakkad, Thiruvanathpuram 9039.4 0.2761
Madhya Pradesh Datiya, East Nimar, Hoshangabad 16873.6 0.5154
Meghalaya Jaintia Hills, Ri Bhoi 6233.1 0.1904
Pondicherry Yanam 38.0 0.0012
Sikkim North Sikkim, West Sikkim 5481.7 0.1674
Tamil Nadu
1692.8
0.0517
Uttarakhand Dehradun, Nanital, Pauri Garhwal, Uttarkashi 20438.1 0.6243
Uttar Pradesh
Etah, Etawah, Farrukhabad, Fatehpur, Firozabad, Gautam Budh Nagar,
Gazipur, Gonda, Gorakhpur, Jalaun, Kannauj, Kanpur, Kanpur Dehat,
Kanshiram Nagar, Kushinagar, Mahamaya Nagar, Maharajganj, Mainpuri,
Mathura, Muzafarnagar, Pratapgarh, Raibeareli, Siddharthnagar, Unnao
97663.9 2.9832
Total
8.0352
444 MAUSAM, 68, 3 (July 2017)
TABLE 4
% area of the country showing increasing and decreasing trend in annual rainfall
Region Period (1961-2013)
Increasing trends Decreasing trends
India* 14% 8%
India* except Leh & Ladakh district 10% 8%
*India except Andaman & Nicobar and Lakshadweep Island
TABLE 5
Mann-Kendall test parameters of some districts showing increasing trend in annual rainfall
Parameters Aizawl Bhagalpur Dantewara Kolar Mayurbhanj North Goa Patiala Prakasam
n 53 53 53 53 53 53 53 53
Z
3.61
3.01
2.01
2.35
1.8
3.96
2.24
2.64
S 471 405 263 307 235 517 293 345
OLS (Regression Line)
Slope 18.55 11.41 5.37 3.53 5.28 36.21 4.93 4.08
Intercept 1858 1148 1317 665 1441 2946 780 759
Theil-Sen Trend Line
Slope 20.25 10.93 5.15 3.43 5.13 34.86 4.07 5.07
Intercept 1865 1206 1261 636 1383 3035 787 717
TABLE 6
Mann-Kendall test parameters of some districts showing decreasing trend in annual rainfall
Parameters East Nimar Etawah Gorakhpur Jalaun Kozikod Nainital Tinsukia Tirap
n
53
53
53
53
53
53
53
53
Z -2.03 -4.22 -1.87 -3.54 -2.79 -3.1 -3.64 -2.22
S -265 -551 -245 -463 -365 -405 -475 -291
OLS Regression Line
Slope -3.99 -9.07 -4.92 -8.5 -20.3 -10.14 -19.92 -4.35
Intercept 1090 895 1261 1066 3426 1667 3149 2449
Theil-Sen Trend Line
Slope -4.69 -9.67 -4.69 -8.46 -17.89 -12.27 -22.82 -6.69
Intercept 1086 917 1254 971 3372 1740 3023 2495
=
S-1
(Var(S) , if S > 0
0, if S = 0
S+1
Va r (S), if S < 0 (5)
If Z is positive, then the trend is increasing, and if Z
is negative, then the trend is decreasing.
3.2. Magnitude of trend
In addition to identifying whether the trend exists,
the magnitude of the trend was also estimated by a slope
estimator β, which was extended by Hirsch et al. (1982)
that was proposed by Sen (1968). β is the robust estimate
of the trend magnitude. In other words, the slope estimator
β is the median of overall possible combinations of pairs
for the whole dataset. A positive value of β indicates an
upward trend(increasing values with time), while a
negative value of β indicates a downward trend
(Karpouzos et al., 2010; Xu et al., 2007). The magnitude
of trend was predicted by the Sens slope estimator with
the slope (Ti) of all data pairs was computed as follows:
Ti = XjXi / j-i (6)
where, Xj and Xi are considered as rainfall values at
time j and i (j > i). The median of these N values of Ti is
KAUR et al.: LONG TERM RAINFALL TREND OVER METEOROLOGICAL SUB DIVISIONS 445
Fig. 5. District wise trend in monsoon rainfall for the period 1901-2013
represented as Sens estimator of slope. Sens estimator
is computed as Qmed=T(N+1)/2 if N is odd and
it is considered as Qmed=[T(N/2) +T(N+1)/2]/2 if N
is even. At the end, Qmed is tested at 100 (1−α)%
confidence interval, and then a true slope can be
obtained by the non-parametric test. A positive value of
Qi indicates an upward or increasing trend, and a negative
value gives a downward or decreasing trend in the
time series.
4. Results and discussion
4.1. Mean, CV and % departure from normal of
annual rainfall
The year wise India’s rainfall from 1901 to 2013 is
given in Fig. 2. The maximum and minimum rainfall
received in the years 1917 and 1972 as 1426 and 912 mm
(excluding Andaman & Nicobar and Lakshadweep Island)
respectively. The years 1905, 1920, 1951, 1965, 1972,
1979, 2002 and 2009 have received less than 1000 mm
rainfall. The decadal % rainfall departure and CV of
India’s rainfall are given in the Fig. 3. The annual decadal
% departure rainfall is showing a pattern of 30 years cycle
with positive and negative departures upto 1980 but it is
showing a positive departure in the decade 1981-1990
different from the pattern. From 1991 onwards it is
showing negative departures till 2013. The years 2014 &
2015 are also having negative departure of -12% and -9%
Fig. 6. District wise trend in annual rainfall for the period 1961-2013
respectively. The value of CV is high in case of negative
decadal annual rainfall % departure and low in case of
positive decadal annual rainfall % departure.
4.2. Trend in districts rainfall (1901-2013)
The Mann-Kendall (MK) trend test is applied to
district wise annual rainfall series (1901-2013) to check
the trend. The state wise number of districts showing
increasing and decreasing trend are given in Table 1. It is
found that 82 districts are showing significant increasing
trend and 156 districts are showing significant negative
trend at 95%. Out of the 632 districts, 13% of the districts
are showing increasing trend, 25% of the districts are
showing decreasing trend and 62 % of the districts are not
showing any trend or the trend is not significant at 5%
level of significance. It can be seen from the Table that
Andhra Pradesh (6), Gujarat (7), Karnataka (8),
Maharashtra (8), W. Bengal (7) districts are showing an
increasing trend whereas Assam (15), Bihar (12),
Chhattisgarh (12), Kerala (9), Madhya Pradesh (12),
Nagaland (10), Uttar Pradesh (27) are showing decreasing
trend. The district wise information of annual rainfall
showing significant increasing trend at 95% & 90%,
decreasing trend at 95% & 90% and increasing &
decreasing trend but not significant are shown in Fig. 4.
The district wise trend in monsoon rainfall is also studied
and results are shown in Fig. 5. It is found that the pattern
of increasing and decreasing trend is similar to annual
446 MAUSAM, 68, 3 (July 2017)
Fig. 7. Mann-Kendall trend test of some districts showing increasing trends
(Note: Red line denotes Theil-Sen trend line and blue line denotes OLS regression line)
KAUR et al.: LONG TERM RAINFALL TREND OVER METEOROLOGICAL SUB DIVISIONS 447
Fig. 8. Mann-Kendall trend test of some districts showing decreasing trends
(Note: Red line denotes Theil-Sen trend line and blue line denotes OLS regression line)
448 MAUSAM, 68, 3 (July 2017)
Figs. 9(a&b). Meteorological (Met.) sub-division wise (a) Annual and (b) Monsoon rainfall, trend for the period (1961-2013)
rainfall. However, the number of districts for significant
increasing and decreasing trend is less in the monsoon
rainfall as compared to annual rainfall. This may be due to
less variability in monsoon rainfall because more than
75% of rainfall occurs during the monsoon period (June-
September).
4.3. Trend in districts rainfall (1961-2013)
The Mann-Kendall (MK) trend test is applied to
district wise annual rainfall series for the recent period
(1961-2013) to check the trend. The state wise number of
districts showing increasing and decreasing trend are
given in Table 1. It is found that 64 districts are showing
significant increasing trend and 85 districts are showing
significant negative trend at 95%. Out of the 632 districts,
10% of the districts are showing increasing trend 13% of
the districts are showing decreasing trend and 77 % of the
districts are not showing any trend or the trend is not
significant at 5% level of significance. It can be seen from
the table that Andhra Pradesh (4), Bihar (4), Gujarat (4),
Manipur (4), Jammu & Kashmir (6), Karnataka (6),
Mizoram (5) and Odisha (12) districts are showing an
increasing trend whereas Arunachal Pradesh (7), Assam
(9), Delhi (8), Haryana (5) and Uttar Pradesh (32) are
showing decreasing trend. The districts with significant
increasing trend at 95% & 90%, decreasing trend at 95%
& 90% and increasing & decreasing trend but not
significant are shown in Fig. 6. It is observed that there is
decreasing trend over NE India and Indo-Gangetic plains.
It is also observed that in the coastal areas there is
increasing trend over northern part of western coast and
eastern coast except Tamil Nadu. Preethi et al. (2016)
mentioned that the decrease in rainfall may be due to
intensification and expansion of the South Asian High
(SAH) in recent times.
The state-wise name of the districts and their
corresponding areas showing increasing and decreasing
significant trend in annual rainfall during the period 1961-
2013 are given in Table 2 and Table 3 respectively. It can
be seen from Table 4, annual rainfall for the period 1961-
2013 shows that 10% area of the country is showing
significant increasing trend whereas 8% of the area of the
country showing significant decreasing trend after
excluding the area of the Leh & Ladakh (J & K) district
due to its large area and sparse network. It is showing the
% area of the country for increasing and decreasing trend
irrespective of the magnitude of the rainfall.
After comparing the periods 1901-2013 and 1961-
2013 (Table 1), it is found that total numbers of districts
are descendants both in increasing and decreasing trend.
In Eastern coast there is an increasing trend in more
number of districts as we move from coast to inland in the
state of Odisha and Andhra Pradesh. It may be due to
more intense storm activities and penetrating more in the
land areas. It can be also seen that, the states of Bihar,
(a)
(b)
KAUR et al.: LONG TERM RAINFALL TREND OVER METEOROLOGICAL SUB DIVISIONS 449
Chhattisgarh, East UP and West MP showing significant
decreasing trend to insignificant decreasing trend. There
are some changes observed in the North Eastern India and
other parts of India. The changes in north eastern India
may be due to the increase in rain-gauge network in the
recent period which computes more precise rainfall
information.
In recent period, 149 districts are showing significant
increasing or decreasing trend. Out of these some districts
showing significant increasing and decreasing trends are
shown in the Fig. 7 and 8 respectively.
The parameters n, S, Z and Slope & Intercepts by
Ordinary Least Square (OLS) Regression and Theil-Sen
Trend line for some districts showing increasing and
decreasing trend in annual rainfall during the period 1961-
2013 are given in Table 5 and Table 6 respectively.
4.4. Trend in meteorological sub-division
4.4.1 Annual rainfall
The trend analysis has been carried out on the annual
rainfall series from 1961-2013 on the 34 sub-divisions
except AN & LK for which gridded data is not available
and the results are shown in the Fig. 9(a), it can be seen
that the sub-divisions of West UP, East UP and Arunachal
Pradesh are showing significantly negative trend in
rainfall whereas the sub-divisions Odisha, Coastal AP and
Konkan & Goa are showing significantly positive trend at
95%.
4.4.2 Monsoon rainfall
From the Fig. 9(b), it can be seen that the sub-
divisions of West UP and East UP are showing
significantly negative trend in monsoon rainfall whereas
the sub-divisions Jammu & Kashmir, Odisha, S.I.
Karnataka and Kankan & Goa are showing significantly
positive trend in monsoon rainfall at 95%.
4.4.3 Comparison of trend in Met.sub-division wise
monsoon and annual rainfall (1961-2013)
When we compare the trend in the sub-divisions of
Annual and Monsoon Rainfall, it is found that there is
difference in the significance of trend for some Met Sub-
divisions. In Jammu & Kashmir trend in monsoon rainfall
is significant at 95% whereas in annual rainfall it is
significant at 90%, In coastal AP trend in monsoon
rainfall is significant at 90% whereas in annual rainfall it
is significant at 95%, In SI Karnataka trend in monsoon
rainfall is significant at 95% whereas in annual rainfall it
is significant at 90%, In Arunachal Pradesh trend in
monsoon rainfall is significant at 90% whereas in annual
rainfall it is significant at 95%. In few subdivisions the
trend is changing from negative to positive side or vice-
versa. In Tamil Nadu & Puducherry sub-division,
significant negative trend in monsoon rainfall at 90%
while it is showing positive trend in annual rainfall which
indicates that there is a positive trend in the NE monsoon
and Bihar subdivision showing positive trend in monsoon
rainfall and negative trend in annual Rainfall. In Assam &
Meghalaya it is showing negative trend in monsoon
rainfall but positive significant trend in annual rainfall.
Further study will be carried out in future publication.
5. Conclusions
This paper gives the information on the districts of
India giving significant increasing, decreasing and no
trend in rainfall which is useful for the state government
officials working in the field of management of water
resources, agriculture, flood management etc. The salient
features are given below:
(i) The annual rainfall for the period 1901-2013, it
varies from 912 mm to 1426 mm (excluding Andaman &
Nicobar and Lakshadweep Island). The years 1905, 1920,
1951, 1965, 1972, 1979, 2002 and 2009 have received less
than 1000 mm rainfall.
(ii) The decadal average rainfall showing a pattern of 30
years cycle with positive and negative departures from
1901-1980 there after it is showing negative decadal
departures except a positive departure in the decade 1981-
1990. It shows that if there is high variability in rainfall in
the recent decades.
(iii) No district is showing increasing trend in the
states/UT of Uttar Pradesh, Uttarakhand, Delhi and
Puducherry in both the periods.
(iv) In the recent period, the maximum number of
districts showing decreasing rainfall trend are 32 in Uttar
Pradesh.
(v) In the recent period, the districts in Odisha, Jammu
& Kashmir and Arunachal Pradesh showing increasing
trend while Assam, Bihar, Mizoram, Nagaland, Madhya
Pradesh, Kerala and Chhattisgarh showing decreasing
trend. It is also seen that all the districts of Delhi showing
negative trend.
(vi) Annual rainfall for the period 1961-2013 shows that
the met sub divisions of Arunachal Pradesh, west & east
UP are showing negative trend and Konkan & Goa,
coastal AP and Odisha are showing significant positive
trends.
450 MAUSAM, 68, 3 (July 2017)
(vii) Monsoon rainfall for the period 1961-2013
shows that the met sub divisions of west & east UP are
showing negative trend and Konkan & Goa, SI Karnataka,
J&K and Odisha are showing significant positive trends.
(viii) When we compare the trends in annual and monsoon
rainfall in the met sub-divisions, they are not following
the same pattern. In few met sub-divisions, there is
negative trend in monsoon rainfall whereas positive in
annual rainfall or vice-versa (Tamil Nadu & Puducherry,
Bihar, Assam & Meghalaya) and in some met sub-
divisions there is same pattern of trend but change in the
level of significance. In Tamil Nadu & Puducherry, a
positive trend is indicated in NE monsoon.
(ix) Annual rainfall for the period 1961-2013 shows that
10% area of the country is showing significant increasing
trend whereas 8% of the area of the country showing
significant decreasing trend irrespective of magnitude of
rainfall.
References
Archer, D. R. and Fowler, H. J., 2004, “Spatial and temporal variations
in precipitation in the Upper Indus Basin, global teleconnections
and hydrological implications”, Hydrol. Earth Syst. Sci., 8,
47-61.
Birsan, M., Molnar, P., Burlando, P. and Pfaundler, M., 2005,
Streamflow trends in Switzerland”, Journal of Hydrology, 314,
312-329.
Burn, D. H., Cunderlik, J. M. and Pietroniro, A., 2004, “Hydrological
trends and variability in the Liard river basin”, Hydrological
Sciences Journal, 49, 53-67.
Douglas, E. M., Vogel, R. M. and Knoll, C. N., 2000,“Trends in flood
and low flows in the United States: Impact of spatial
correlation”, Journal of Hydrology, 240, 90-105.
Gajbhiye Sarita, Meshram Chandrashekhar, Singh S. K., Srivastava
Prashant, K. and Islam Tanvir, 2016, “Precipitation trend
analysis of Sindh River basin, India, from 102-year record
(1901-2002)”, Atmospheric Science Letters, 17, 1, 71-77.
Guhathakurta, P. and Rajeevan, M., 2007, “Trends in the rainfall pattern
over India”, Int. J. Climatol., 28, 1453-1469.
Helsel, D. R. and Hirsch, R. M., 1992, Statistical Methods in Water
Resources”, Elsevier, Amsterdams, the Netherlands, Elsevier
Publishers, 529.
Hirsch, R. M., Slack, J. R. and Smith, R. A., 1982,“Techniques of trend
analysis for monthly water quality data”, Water Resources
Research, 18, 1, 107-121.
Intergovernmental Panel on Climate Change, 2007, The physical
science basis; In: Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on
Climate Change (eds), Cambridge University Press,
Cambridge.
Karpouzos, D. K., Kavalieratou, S. and Babajimopoulos, C., 2010,
Trend analysis of precipitation data in Pieira Region (Greece)”,
European Water, E.W. Publications, 30, 31-40.
Kumar, V., Singh, P. and Jain, S. K., 2005, “Rainfall trends over
Himachal Pradesh, Western Himalaya, India. In: Development
of Hydro Power Projects - A Prospective Challenge”, Shimla.
Libiseller, C. and Grimval, A., 2002, “Performance of partial Mann-
Kendall test for trend detection in the presence of covariates”,
Environmetrics, 13, 71-84.
Pai, D. S., Sridhar L., Rajeevan M., Sreejith O. P., Satbhai N. S. and
Mukhopadhyay, B., 2014, “Development of a new high spatial
resolution (0.25° × 0.25°) Long Period (1901-2010) daily
gridded rainfall data set over India and its comparison with
existing data sets over the region”, Mausam, 65, 1-18.
Pant, G. B., Borgaonkar, H. P. and Kumar, Rupa, 1999, “Climate
variability over the Western Himalaya since little ice age:
dendroclimatic implications”, In: Proceedings of national snow
science workshop“Technology in support of snow and
avalanche research”, Manali (India).
Pant, G. B. and Kumar K. R., 1997, Climates of South Asia”, John
Wiley & Sons Ltd., Chichester, UK
Parry, M. L., Canziani, O. F., Palutikof J. P., van der Linden P. J. and
Hanson, C. E., 2007, “Contribution of Working Group II to the
Fourth Assessment Report of the Intergovernmental Panel on
Climate Change, Cambridge University Press, NY, USA.
Preethi, B., Mujumdar, M., Kripalani, R. H., Prabhu, A. and Krishnan,
R., 2016, “Recent trends and tele-connections among South and
East Asian summer monsoons in a warming environment”,
Climate Dynamics, online DOI 10.1007/s00382-016-3218-0
Sen, P. K., 1968, “Robustness of some nonparametric procedures in
Linear Models”, The Annals of Mathematical Statistics, 39, 6,
1913-1922.
Shepard, D., 1968, “A two-dimensional interpolation function
for irregularly spaced data”, Proc. 1968 ACM Nat. Conf,
517-524.
Singh, P., Kumar, V., Thomas, T. and Arora, M., 2008a, “Changes in
rainfall and relative humidity in different river basins in the
northwest and central India”, Hydrological Processes, 22,
2982-2992.
Xu, Z. X., Li, J. Y. and Liu, C. M., 2007, “Long-term trend analysis for
major climate variables in the Yellow River basin”,
Hydrological Processes, 21, 1935-1948.
Yu, Y. S., Zou, S. and Whittemore, D., 1993, “Non-parametric trend
analysis of water quality data of rivers in Kansas”, Journal of
Hydrology, 150, 61-80.
Yue, S. and Hashino, M., 2003, Long term trends of annual and
monthly precipitation in Japan”, Journal of American Water
Resources Association, 39, 3, 587-596.
... The watersheds of the Ganga and Brahmaputra basin which provide major biodiversity habitats plays a significant role in providing water, food, and climate security to the basin's 630 million inhabitants [28] also showed precipitation deficit in the present scenario (2006-2015) (Fig. 2) while comparing it 20 years back precipitation scenario (1976)(1977)(1978)(1979)(1980)(1981)(1982)(1983)(1984)(1985). Diwakar et al. [29] also found a similar observation of the decreasing trend of rainfall over the river basin of Ganga. ...
Article
Full-text available
We have analyzed the geospatial datasets such as precipitation, runoff, soil moisture, aridity, soil degradation, and future (2050) climate of India and investigated the spatial distribution pattern at the watershed level. Furthermore, we have investigated the long-term TerraClimate present decadal (2006–2015) trend with 20 years back decadal (1976–1985) data for evaluating temporal change in precipitation, runoff, and soil moisture at the watershed level of India. The long term decadal precipitations, as well as soil moisture deficit trend, are found very significant in the watersheds of the Ganga and Brahmaputra basin. The decadal runoff increase (%), when compared with 20 years back decadal runoff showed a high percent (> 50%) increase in the majority of Sabarmati river basin in Gujarat state of India. The three villages Milkipur, Bikapur, and Bantikalan (Faizabad district of Uttar Pradesh) have shown a maximum reduction of soil moisture. The analysis of predicted (2050) temperature and precipitation anomaly showed the precipitation deficit in the majority of watersheds of Indus river basin and their sub-basin. Similarly, the temperature increase in the year 2050 is found very significant in almost all watersheds of India with a range of 0.8 to 1.9 °C but it is more crucial for some of the northern parts of Indus river basin and Brahmaputra basin. Such analysis highlights the need for an adequate management plan with robust soil and water conservation at a watershed level for achieving sustainable development goals (SDGs). Please find the link for view the PDF file https://rdcu.be/b2hel
... The decadal precipitation deficit during kharif and rabi seasons were found more pronounced in Indo-Gangetic plain is a matter of concern. The similar findings of rainfall variation were reported in Indo-Gangetic plain [36] and at the country level of India [37][38][39]. ...
Article
Full-text available
The paper deals with various thematic parameters such as agriculture crop scenario (2000, 2010), water stress, precipitation trend and deficit, climate-induced risk towards crops, drought-prone area, suicide attributes of farmers, agro-ecological regions, prediction of future (2050) precipitation and temperature variation during kharif and rabi seasons of India and their spatial pattern were analyzed in GIS for better understanding of climate change. The analysis revealed about the need of synergic approach/strategies to address the impact of climate change. Few of the Climate-smart villages (CSVs) projects of India were discussed here based on their approach, achievement, and limitation. The CSV conceptual strategies are fully based on climate smart agriculture potentiality to achieve sustainability in food security, enhancing the livelihood, eradication of poverty and magnifying the farm household resilience. The climate-induced high and very high risk to the crops areas were found dominated in the arid and semi-arid regions which will be challenged in future due to water stress, inadequate irrigation facility, increasing trend of temperature and variation in precipitation pattern. The hotspot districts of farmer’s suicide were very significant in climate-induced very high risk zone and majority of them falls in the drought-prone areas/extremely high to high water-stressed areas which leads to crop failure. There is a need to formulate a concrete policy, legal, and institutional actions addressing the farmers problem significantly at country, state, district and village levels which will support investment/technology/guideline in and adoption of Climate-smart village (CSV) practices after seeing the socio-economic background (poverty/tribes/backward class) of them.
Article
Full-text available
Indian monsoon rainfall has a very strong connection with the Indian economy. Any variation in trend or pattern of Indian summer monsoon rainfall will have serious implications on agronomy, water resources and various associated sectors of the economy in India. In this study, an in-depth investigation of the monsoon rainfall trend is analyzed for 146 years period (1871–2016). Three different spatial scales using a multimethod approach consisting of the Linear Regression Model (LRM), Mann Kendall Test (MKT) and Innovative Trend Analysis (ITA) are analyzed in particular and synchronized way. Monotonic trend with one or the other tests are found in Meteorological Sub-division (MetSD) 3, 4, 14, 11, 10, 19, 20, 27, 8, 29, 28, 23 and 32 (Assam, Meghalaya, Manipur, Mizoram, Nagaland, Tripura, Punjab, Uttar Pradesh, Madhya Pradesh, Chhattisgarh, Jharkhand, Andhra Pradesh, Telangana , Konkan & Goa and Coastal Karnataka). Whereas, no significant monotonic trend was found for India as a unit. Two Homogenous Monsoon Regions (HMR) i.e. Central Northeast and Northeast have the monotonic rainfall trend. Moreover, the synchronized methodology made it possible to identify the most refined significant monotonic trend. It revealed a decreasing monotonic trend in MetSD 4, 20 and 27 (Manipur, Mizoram, Nagaland, Tripura, East Madhya Pradesh and Chhattisgarh) only. ITA based results revealed that MetSD 14, 8, 19 and 29 (Punjab, Jharkhand, West Madhya Pradesh and Telangana) are a new addition to the list of MetSDs with the significant monotonic trend. Changepoint in the trend is obtained for Northeast HMR in the year 1956 and MetSD 4, 20, 23 and 27 in the year 1969, 1961, 1930 and 1961 respectively. This study provides insight into the most refined trend on monsoon rainfall at different spatial scales in India using the updated methods of analysis.
Article
Global warming is a biggest issue around the world. In this research paper, the temporal and spatial trend analysis of seasonal and annual rainfall on Bhogavo River watersheds in Sabarmati lower basin of Gujarat state of India has been analysed using the data of 11 rain gauge stations installed in Bhogavo watershed. Linear regression, Mann–Kendall Test, Sen’s slope test and innovative trend analysis methods are used to carry out monthly and annual rainfall trend analysis. In addition to the rainfall analysis, a number of rainy days change in magnitude as a percentage of mean rainfall have also been analysed using linear regression and Sen’s slope method, respectively. The IDW method is used to develop a spatial distribution of annual and seasonal rainfall trend over the study area. From the results, it is concluded that annual rainfall shown increasing (positive) trend at nine stations out of 11 stations. The highest value of change in magnitude of trend as a percentage of mean monthly rainfall has been obtained in the month of July, attributing increasing trend at Sayla station and lowest value magnitude of trend as a percentage of mean rainfall in the monthly rainfall has been obtained in the month of August, attributing decreasing trend at Bavla station.
Article
Full-text available
Climate change is now widely recognized as the major environmental threat occurring mainly due to increasing greenhouse gases in the atmosphere and causing the extinction of biodiversity and enhances disruptions to ecosystems. Climate change risks are found to be very significant and would have a profound impact on the livelihoods of millions of poor people in India. In the present study, we have analyzed datasets of agriculture Greenhouse gases (GHGs) emission (1990-2016), poverty, anthropogenic biomes, agriculture crop production scenario (2008-2017), seasonal soil moisture status present (2006-2015) and deficit (compared with 20 years' time periods), present (1970-2000) annual mean, future (2050) precipitation and temperature scenario of India and investigated the spatial pattern and relationship incorporating in remote sensing and GIS for the better comprehension of the impact of climate changes on the socioeconomic dimension of the people. The total GHGs emissions (CO 2 equivalent) from agriculture in India were showing the increasing trend (approximately 1% annually) whereas the increasing trend has decreased notably in the last five years. The states of Bihar, Uttar Pradesh, West Bengal, Assam, Punjab, and Tamil Nadu and Andhra Pradesh of India were found producing higher GHGs emissions from agriculture. The present soil moisture and the deficit during the Kharif, Rabi, and Zaid seasons vary geographically whereas soil moisture deficit during the Kharif season was found very significant over most of the districts of Ganga and Brahmaputra basin. The present mean annual temperature and precipitation patterns were found very significant in arid and semi-arid regions which are known as the farmer's suicide hotspot and are predicted to threaten more in the future (2050). The evaluation highlights the need for synergic approaches such as climate-smart agriculture (CSA) to address the impact of climate change in food production and farmers' livelihood. Such an investigation gives a solid understanding of a paradigm shift in the adaptation of CSA an approach in prevailing new climate change reality at country or regional levels for achieving socioeconomic and climate resilience agriculture in India.
Article
Full-text available
Recent trends, variations and tele-connections between the two large regional sub-systems over the Asian domain, the South Asian and the East Asian monsoons are explored using data for the 1901–2014 period. Based on trend analysis a dipole-type configuration with north-drought and south-flood over South as well as East Asia is observed. Two regions over South Asia, one exhibiting a significant decreasing trend in summer monsoon rainfall over northeast India and the other significant increasing trend over the northern parts of the west coast of India are identified. Similarly two regions over East Asia, one over South Korea-southern parts of Japan and the other over South China are also identified both indicating a significant increasing trend in the summer monsoon rainfall. These trends are examined post 1970s. Possible factors associated with the recent trends are explored. Analysis of sea surface temperature (SST), mean sea level pressure and winds at lower troposphere indicates that the entire monsoon flow system appears to have shifted westwards, with the monsoon trough over South Asia indicating a westward shift by about 2–3° longitudes and the North Pacific Subtropical High over East Asia seems to have shifted by about 5–7° longitudes. These shifts are consistent with the recent rainfall trends. Furthermore, while the West Indian Ocean SSTs appear to be related with the summer monsoon rainfall over northern parts of India and over North China, the West Pacific SSTs appear to be related with the rainfall over southern parts of India and over South Korea- southern Japan sector.
Article
Full-text available
The study discusses development of a new daily gridded rainfall data set (IMD4) at a high spatial resolution (0.25° × 0.25°, latitude × longitude) covering a longer period of 110 years (1901-2010) over the Indian main land. A comparison of IMD4 with 4 other existing daily gridded rainfall data sets of different spatial resolutions and time periods has also been discussed. For preparing the new gridded data, daily rainfall records from 6955 rain gauge stations in India were used, highest number of stations used by any studies so far for such a purpose. The gridded data set was developed after making quality control of basic rain-gauge stations. The comparison of IMD4 with other data sets suggested that the climatological and variability features of rainfall over India derived from IMD4 were comparable with the existing gridded daily rainfall data sets. In addition, the spatial rainfall distribution like heavy rainfall areas in the orographic regions of the west coast and over northeast, low rainfall in the lee ward side of the Western Ghats etc. were more realistic and better presented in IMD4 due to its higher spatial resolution and to the higher density of rainfall stations used for its development.
Article
Full-text available
The study of long-term precipitation record is critically important for a country, whose food security and economy rely on the timely availability of water. In this study, the historical 102-year (1901-2002) rainfall data of the Sindh River basin (SRB), India, were analyzed for seasonal and annual trends. The Mann-Kendall test and Sen's slope model were used to identify the trend and the magnitude of the change, respectively. Spatial interpolation technique such as Kriging was used for interpolating the spatial pattern over SRB in GIS environment. The analysis revealed the significantly increasing precipitation trend in both seasonal and annual rainfall in the span of 102years.
Article
Full-text available
Precipitation has been widely considered as a starting point towards the apprehension of climate changes courses, constituting one of the most substantial components of the hydrologic cycle. The purpose of this work is to investigate the temporal variability of precipitation in order to contribute to a better interpretation of the hydrological status of the study area located in Northern Greece within the prefecture of Pieria. The Aison River basin drains the water of its central part, forming the greatest receiver of surface water. The precipitation data refer to the period 1974-2007. The methodology adapts various statistical approaches in order to detect possible precipitation changes in annual, monthly and seasonal basis, including both tests for monotonic trend (Mann-Kendall test, Sequential version of the Mann-Kendall test, Sen's estimator of slope) and a test for step change (the distribution-free CUSUM test). The application of this methodological framework resulted in the identification of a few trends, occurred mainly at smaller than annual time scales. An overall downward trend, even though no statistically significant, has derived for the stations located at lower altitudes, within the agricultural zone of Pieria. Regional (Aison River basin) total precipitation series has also displayed trends similar to the ones observed in the majority of stations. Among the four seasons, a noticeable decrease of precipitation is observed in Spring by both ways of analysis (station based and regional).
Article
This is a study of the South Asian climate, with a description and interpretation of the complex atmospheric dynamics and explanation of the intricacies of monsoon meteorology. The climatic end products of the monsoonal system, especially the rainfall, affect hundreds of millions of the earth's population in India, Pakistan, Bangladesh, Sri Lanka and Nepal. The book is organised into two aprts: The first provides meterological background to understanding the Asian climate. Placing the regional circulation in perspective of the tropical general circulation and describing the specific features dominating the climate. The second part focuses upon the climatological characteristics of South Asia; the mean climate is described, followed by specific features of individual countries.
Article
Seasonal and annual trends of changes in rainfall, rainy days, heaviest rain and relative humidity have been studied over the last century for nine different river basins in northwest and central India. The majority of river basins have shown increasing trends both in annual rainfall and relative humidity. The magnitude of increased rainfall for considered river basins varied from 2-19% of mean per 100 years. The maximum increase in rainfall is observed in the Indus (lower) followed by the Tapi river basin. Seasonal analysis shows maximum increase in rainfall in the post-monsoon season followed by the pre-monsoon season. There were least variations in the monsoon rainfall during the last century and winter rainfall has shown a decreasing trend. Most of the river basins have experienced decreasing trends in annual rainy days with a maximum decrease in the Mahanadi basin. The heaviest rain of the year has increased from 9-27 mm per 100 years over different river basins with a maximum of 27 mm for the Brahamani and Subaranrekha river basins. A combination of increase in heaviest rainfall and reduction in the number of rainy days suggest the possibility of increasing severity of floods. Such information is useful in the planning, development and management of water resources in the study area. Further, the majority of river basins have also experienced an increasing trend in relative humidity both on seasonal and annual scales. An increase in annual mean relative humidity for six river basins has been found in the range of 1-18% of mean per 100 years, while a decrease for three river basins from -1 to -13% of mean per 100 years was observed, providing a net increase in the study area by 2.4% of mean per 100 years. It is understood that an increase in areal extent of vegetation cover as well as rainfall over the last century has increased the moisture in the atmosphere through enhanced evapotranspiration, which in turn has increased the relative humidity.
Article
In many fields using empirical areal data there arises a need for interpolating from irregularly-spaced data to produce a continuous surface. These irregularly-spaced locations, hence referred to as “data points,” may have diverse meanings: in meterology, weather observation stations; in geography, surveyed locations; in city and regional planning, centers of data-collection zones; in biology, observation locations. It is assumed that a unique number (such as rainfall in meteorology, or altitude in geography) is associated with each data point. In order to display these data in some type of contour map or perspective view, to compare them with data for the same region based on other data points, or to analyze them for extremes, gradients, or other purposes, it is extremely useful, if not essential, to define a continuous function fitting the given values exactly. Interpolated values over a fine grid may then be evaluated. In using such a function it is assumed that the original data are without error, or that compensation for error will be made after interpolation.
Book
This is the original first edition published as a physical book by Elsevier. It is woefully out of date. An updated electronic version was published in 2002 by the U.S. Geological Survey, and a completely revised 2020 version with updated methods and supporting materials is listed in my publication list, and is available for download at https://doi.org/10.3133/tm4a3 .
Article
Some previous studies have shown that drying-up of the lower Yellow River resulted from decreasing precipitation and excessive industrial and agricultural consumption of water from the middle and downstream regions of the Yellow River. On the basis of average air temperature, precipitation, and pan evaporation data from nearly 80 gauging stations in the Yellow River basin, the monotonic trends of major climate variables over the past several decades are analysed. The analysis was mainly made for 12 months and the annual means. The isograms for annual and typical months are given in the paper. The result shows that the average temperature in the study area exhibits an increasing trend, mainly because of the increase of temperature in December, January and February. The largest trend is shown in December and the smallest is in August. There are 65 of 77 stations exhibiting a downward trend for annual precipitation. In all seasons except summer, there is a similar trend in the upstream region of the Yellow River, south of latitude 35°N. It is interesting to note that the pan evaporation has decreased in most areas of the Yellow River basin during the past several decades. April and July showed the greatest magnitude of slope, and the area from Sanmenxia to Huayuankou as well as the Yiluo River basin exhibited the strongest declining trend. The conclusion is that the decreasing pan evaporation results from complex changes of air temperature, relative humidity, solar radiation, and wind speed, and both climate change and human activities have affected the flow regime of the Yellow River during the past several decades. Copyright © 2007 John Wiley & Sons, Ltd.