raju & nath.pdf

Give to AgEcon Search
The World’s Largest Open Access Agricultural & Applied Economics Digital Library
This document is discoverable and free to researchers across the
globe due to the work of AgEcon Search.
Help ensure our sustainability.
AgEcon Search
http://ageconsearch.umn.edu
aesearch@umn.edu
Papers downloaded from AgEcon Search may be used for non-commercial purposes and personal study only.
No other use, including posting to another Internet site, is permitted without permission from the copyright
owner (not AgEcon Search), or as allowed under the provisions of Fair Use, U.S. Copyright Act, Title 17 U.S.C.

Heterogeneous Climatic Impacts
on Agricultural Production: Evidence
from Rice Yield in Assam, India
Hiranya K. Nath
Sam Houston State University
Huntsville, Texas, USA
Email: eco_hkn@shsu.edu
Raju Mandal
Assam University
Silchar, Assam, India
Email: rajum06@gmail.com
ABSTRACT
Understanding the nature and extent of climatic impacts on agricultural productivity under a variety
of scenarios is extremely important for developing countries, where a sizable portion of the population
relies on agriculture for life and livelihood. Thus, this paper presents evidence of heterogeneity in
climatic impacts on crop yield in Assam, India. In particular, applying the non-parametric quantile
regression technique to district-level data from 1978 to 2005, this study examined heterogeneity in the
impacts of temperature and rainfall across seasonal rice varieties (autumn, winter, and summer), agro-
climatic (AC) zones, and the distribution of rice yield. The results suggested that, in general, the effects
of temperature on yield were not statistically significant for any of the three seasonal rice varieties.
However, these effects were not uniform in their magnitudes, signs, and statistical significance across
AC zones and yield distribution for each variety of rice. Similarly, there were wide variations in the
effects of total precipitation across seasonal varieties, AC zones, and yield distribution. The results
also suggested that an increase in temperature variability is beneficial and that rainfall variability is
harmful to autumn and winter rice yield. For summer rice, the effects of these two climate variables
were positive but statistically insignificant. Given the importance of rice yield for food security
and poverty alleviation in Assam, these results could inform the design of appropriate adaptation
strategies and public policies to counter the adverse impacts of climate change on agriculture in
Assam. Furthermore, since most people in rural areas are engaged in agriculture, these results are
important for the sustainability of rural economies.
Keywords: climatic impacts, rice yield, Assam, India, autumn rice, winter rice, summer rice,
quantile regression, median regression
JEL Classification: Q11, Q18, Q54

24 Hiranya K. Nath and Raju Mandal
INTRODUCTION
The study of climatic impacts on
agricultural productivity is extremely important
for developing countries where a sizable portion
of the population relies on agriculture for life
and livelihood. Although what causes climatic
changes is debatable, the evidence of changes
in temperature, precipitation, and extreme
weather events is indisputable. Since climatic
conditions directly affect agriculture, it is but
natural to examine the impacts of changes in
these conditions on crop yield.
There is substantial empirical literature on
how climatic changes impact agriculture using
data on various crops from different parts of the
world. In the beginning, the literature primarily
focused on developed countries (Kaiser et al.
1993; Mendelsohn, Nordhaus, and Shaw 1994;
Adams et al. 1998; Adams, Hurd, and Reilly
1999; Lewandrowski and Schimmelpfennig
1999; Bryant et al. 2000). However, some more
recent studies examined climatic impacts on
agriculture in developing countries (Sanghi
and Mendelsohn 2008; Molua 2009; Deressa
and Hassan 2009; Wang et al. 2009; Sarker,
Alam, and Gow 2012; Poudel and Kotani 2013;
Burney and Ramanathan 2014; Singh et al.
2017). In general, these studies found evidence
of significant impacts of climatic changes
on mean yield, growth, and yield variability
of different crops, and used these findings to
derive implications for adaptation strategy in
cropping patterns.
Several recent studies that investigated
climatic impacts on agriculture in India have
mixed results—while some studies found
evidence of negative impacts of climate change
(Auffhammer, Ramanathan, and Vincent
2006; Cline 2007; Aggarwal 2008; Guiteras
2009; Burney and Ramanathan 2014; Rao
et al. 2014; Kumar et al. 2015; Singh et al.
2017), others showed that climatic changes
have positive effects (Mohandass et al. 1995;
Lal et al. 1998; Rathore et al. 2001; Aggarwal
and Mall 2002; Abeysingha et al. 2016).
There were other studies that reported both
positive and negative effects under different
climate change scenarios (Dubey et al. 2014;
Yadav et al. 2015). An interesting finding was
that the climatic impacts on agriculture are
non-linear (Auffhammer, Ramanathan, and
Vincent 2012).1
This is particularly important
for policymakers for adoption of appropriate
countermeasures in response to the negative
impacts of climate change. The distributional
heterogeneity in the climatic impacts on crop
yield was further highlighted by Krishnamurthy
(2012) and Barnwal and Kotani (2013).
These studies applied quantile regression
technique to district-level data for India and the
state of Andhra Pradesh, respectively.
In this paper, the heterogeneity in climatic
impacts on agricultural production was studied
further. The case of rice yield in Assam, a state
in the northeast region of India where it is a
staple crop for its population of over 30 million,
was considered. In particular, heterogeneity
in the impacts of temperature and rainfall on
rice yield was studied along three dimensions:
rice varieties grown in different seasons,
agro-climatic conditions that would primarily
reflect differences in soil quality, and different
levels along the distribution of rice yield.
The knowledge of these heterogeneities would
be useful in developing an understanding of the
nature and magnitude of adaptations to climatic
changes, and in formulating appropriate policy
to ensure food security in a state like Assam.
This paper primarily applied a non-
parametric regression technique to district-level
data from 1978 to 2005 to examine the above-
mentioned heterogeneities in climatic impacts
on rice yield in Assam. By considering rice
1 Schlenker and Roberts (2006, 2009) document
nonlinearity in the effects of weather on corn yield for
the U.S.

Asian Journal of Agriculture and Development, Vol. 15 No. 1 25
grown in three seasons of the year, it uncovers
seasonal disparities in climatic impacts.
High resolution gridded daily temperature
and rainfall data were used to construct the
climate variables.2
The results suggested that, in
general, the effects of temperature on yield were
not statistically significant for any of the three
seasonal rice varieties. However, these effects
were not uniform in their magnitudes, signs,
and statistical significance across agro-climatic
zones (AC zones) and yield distribution for
each variety of rice.
Similarly, there were wide variations
in the effects of total precipitation across
seasonal varieties, agro-climatic zones, and
yield distribution. The results also suggested
that an increase in temperature variability
was beneficial, but rainfall variability was
harmful to autumn and winter rice yields.
For summer rice, although the effects of
these two variables were positive, they were
statistically insignificant. Given the importance
of crop yield for food security and poverty
alleviation, these results seem to suggest that
growing season, location, and current yield
should be taken into account in formulating
appropriate adaptation strategies and public
policies for rice cultivation to counter the
adverse effects of climatic changes. No other
study has examined the climatic impacts on rice
yield focusing on their heterogeneity across
seasonal varieties, agro-climatic zones, and
yield distribution for the state of Assam.
2 Previousstudies(Mearns,Rosenzweig,andGoldberg
1996; Schlenker and Roberts 2009) showed that
day-to-day variations in weather conditions such as
temperature and rainfall during the growing season
have crucial effects on the growth and yield of crops.
Therefore, high resolution daily weather data are
extremely valuable in studying the impact of climate
change on agriculture. However, most studies on
climatic impact on agricultural yield in India used
monthly weather data primarily due to a lack of high-
resolution daily weather data until recently.
METHODS AND DATA
Empirical Model and Methodology
For the empirical analysis, the model
below, which postulates that rice yield depends
on a number of climatic and other variables,
was used:
(1)
y = x'β + z'γ + ϵ
where y is the rice yield, x is a (k × 1) vector of
climate variables, z is a (m × 1) vector of control
variables, β and γ are the corresponding vectors
of coefficients to be estimated, and ϵ is the error
term. In the baseline specification, climatic
variables included mean of daily temperature
during a growing season, temperature variability
(standard deviation of daily temperature
during the season), total rainfall, and rainfall
variability during the season. Control variables
included a set of five dummy variables that
captured AC zone-specific fixed effects,
a time trend that represented technology,
total cropped area under the specific seasonal
rice variety, fertilizer consumption, and a
dummy variable for the districts that were
frequently and heavily affected by floods.3,4
In the extended specification, the interactions of
mean temperature and total rainfall with the AC
zone dummies were also included to examine
differential effects of the climate variable across
these zones.
3 There are six agro-climatic zones in Assam and
five dummies were included. In a similar study for
India, Krishnamurthy (2012) used district fixed
effects. Since the districts in Assam were small and
adjacent districts were very similar, location-specific
fixed effects at the agro-climatic zone level were
controlled. This was similar to the study of Barnwal
and Kotani (2013).
4 The flood dummy took the value of 1 for a district
that experienced frequent and widespread floods
(in terms of total cropped area inundated) and 0
otherwise.

The quantile regression (QR) technique
was used to estimate model (1).5
First, a median
regression (50th quantile regression) was used
to examine the heterogeneity in climatic impacts
on rice yield across three varieties grown in
three different seasons, and then across six
AC zones.6
In both cases, panel ordinary least
square (OLS) regression was also estimated for
comparison with the results from the median
regression. In order to investigate the differences
in the impacts of temperature and rainfall across
distribution, regressions were estimated for the
20th, 40th, 60th, and 80th quantile.
Unlike the conventional parametric linear
regression techniques, this non-parametric
approach allowed the consideration of the
climatic impacts on the entire distribution of rice
yield and not merely on its conditional mean.
Thus, it provided a richer characterization of
the data by uncovering the heterogeneity in the
impacts of climatic changes. This is important
because, as Barnwal and Kotani (2013) argued,
the potential non-stationarity of climatic
variables and crop yield unfolded the possibility
of asymmetric effects of temperature, rainfall,
and other covariates across the conditional
distribution of rice yield. Other advantages of
QR over OLS regression included the fact that
while OLS estimates could be inefficient in case
of non-normal errors, QR estimates were robust
to non-normal errors and outliers. Furthermore,
by allowing for different coefficients at different
quantiles, the technique took care of potential
heteroskedasticity. Only Barnwal and Kotani
(2013) and Krishnamurthy (2012) used the QR
technique to examine the climatic impacts on
agriculture in the Indian context.
5 Quantile regression technique was proposed by
Koenker and Bassett (1978)
6 Sarker, Alam, and Gow (2012) used median
regression to examine the relationship between
climate change and rice yield in Bangladesh.
The estimation of the coefficients involved
minimizing a non-differentiable objective
function through the simplex method that was
guaranteed to yield a solution in a finite number
of iterations. The standard errors were computed
by bootstrapping. Since QR is a non-parametric
method, no specific assumption about the
distribution of the error term was required.
Study Area
Located south of the eastern Himalayas,
Assam comprises the Brahmaputra and the
Barak river valleys along with Karbi Anglong
and the North Cachar Hills with an area of
78,437.79 km2
. Assam has sub-tropical humid
climate conducive to rice cultivation. Average
temperatures in the state range from high
90ºF (about 32ºC) in August to mid-40ºF
(about 4ºC) in January. Although the state
receives some rain between March and May,
the heaviest precipitation comes with the
southwest monsoon, which arrives in June and
stays through September. With annual average
rainfall that varies from about 1,800 mm in the
west to more than 3,000 mm in the east, Assam
ranks among the world’s regions with the
highest rainfall. Assam is primarily an agrarian
economy with about two-thirds of its workforce
engaged in agriculture. Rice is the staple crop
that accounts for more than 60 percent of the
total cropped area in the state.
Based on rainfall pattern, geographic
terrain, and soil characteristics, Assam has been
divided into six AC zones: Lower Brahmaputra
Valley zone, Central Brahmaputra Valley zone,
North Bank Plain zone, Upper Brahmaputra
Valley zone, Hill zone, and Barak Valley zone.
The AC zones are shown in inset in Figure 1.
Table 1 lists the districts in each AC zone. The
table also shows the relative sizes (in total areas)
of these zones and the distribution of cropped
areas under the three seasonal rice varieties.

Data Collection
This study focused on the impacts of
climate variables, namely, temperature and
rainfall, on average yield of three distinct
seasonal varieties of rice inAssam: autumn rice,
winter rice, and summer rice.7
The data used in
this study covered 22 districts of the state from
1978 to 2005.8
The selection of the sample was
determined primarily by the availability of data
on all relevant variables in a consistent manner.
The districts have undergone boundary changes
7 They are so named according to the respective
seasons of their harvesting.
8 Districts are administrative units with clearly
demarcated geographical boundaries. They are
much like the counties in the United States. A district
is often divided into two or more sub-divisions.
with the creation of new districts from time to
time. Until 1981, there were only 10 districts.
Between 1981 and 1991, eight of them were
split to create 13 new districts, bringing the total
number to 23.
Since these new districts were sub-
divisions of the parent districts and data for
most relevant variables were collected at the
level of those sub-divisions, the data series
for the entire sample period for all but the
Bongaigaon district could easily be constructed.
Therefore, this district was dropped from the
sample. Four more districts were carved out
in 2003–04. However, since these new
districts were created combining parts of more
than one erstwhile districts, it was difficult to
construct data for these new districts for the
entire sample period. Furthermore, data for
these new districts were published separately
Figure 1. Map of India (with the AC zones of Assam in inset)
Sources: Maps of India (2014) and NESAC (n.d.)

after 2004–05; but it was almost impossible to
map those data back to the pre-2003–04 districts
that were used in the sample. Therefore, the
sample period was extended only up to 2005.
The data used in this study can be divided into
two broad categories: agricultural data and
climate data.
Agricultural data
Data on agricultural variables were
obtained from three government agencies:
Directorate of Agriculture, Government of
Assam (DoA-GOA); Directorate of Economics
and Statistics, Government of India (DES-
GOI); and Directorate of Economics and
Statistics, Government of Assam (DES-
GOA). The three main agricultural variables
used in the empirical model were average
rice yield (output of a seasonal rice variety
in kilogram per hectare), total rice area (area
under a seasonal rice variety in hectares), and
fertilizer consumption (use of all kinds of
chemical fertilizer in ‘000 kg for all crops).
The data on average rice yield and total rice
area were compiled from various issues of
Basic Agricultural Statistics published by DoA
(1978–1979 to 1996–1997) and Estimates of
Area, Production andAverageYield of Principal
Crops in Assam published by DES (1978–1979
to 2005–2006). Fertilizer consumption data
were obtained from DoA; however, fertilizer
data were available for all crops and not
separately for rice.
Zone No.
(as used in
the paper)
AC Zones Districts
Total
Geographical
Area (ha)
Percentage of Total Cropped Area
(Average over 1978–1979
to 2004–2005)
Autumn
Rice
Winter
Rice
Summer
Rice
1 Lower
Brahmaputra
Valley zone
Dhubri,
Bongaigaon,
Goalpara,
Kokrajhar, Barpeta,
Nalbari, Kamrup
2,014,800
(25.69)
21.58 37.53 5.85
2 Central
Brahmaputra
Valley zone
Nagaon,
Morigaon
553,500
(7.06)
14.03 39.71 13.18
3 North Bank
Plain zone
Darrang, Sonitpur,
Lakhimpur,
Dhemaji
1,431,900
(18.26)
16.41 43.88 3.12
4 Upper
Brahmaputra
Valley zone
Dibrugarh,
Tinsukia, Sibsagar,
Jorhat, Golaghat
1,619,200
(20.64)
4.91 51.24 0.43
5 Hill zone Karbi Anglong,
North Cachar
1,532,200
(19.53)
8.11 53.87 0.42
6 Barak Valley
zone
Cachar, Hailakandi,
Karimganj
692,200
(8.82)
9.82 59.66 4.32
Assam 7,843,800
(100.00)
13.05 44.29 5.91
Table 1. Agro-climatic zones (AC zones) in Assam
Sources: Compiled from DoA (1978–1979 to 1996–1997) and DES (1978–1979 to 2005–2006)
Note: Percentage shares of the agro-climatic zones in total area are shown in parentheses.

Climate data
The study used high-resolution (1º×1º
latitude-longitude) daily gridded temperature
and rainfall data compiled by the National
Climate Centre of India Meteorological
Department (IMD), Government of India. IMD
uses Shepard's angular distance weighting
method on historical data from 395 quality-
controlled stations for the period 1969 to 2005
to create the gridded temperature data
(Srivastava, Rajeevan, and Kshirsagar 2009).
The dataset has been recently updated until
2014. Similarly, IMD uses daily rainfall data
from 1,803 stations across India to create a
high-resolution gridded precipitation dataset for
the period 1951–2003 (Rajeevan et al. 2006),
which was later updated to 2015.
A modified Shepard's inverse weighting
interpolation method was used to obtain
district-level daily temperature and rainfall
data from the high-resolution gridded data. The
geographical center of each district was taken
and the grid points that fell within 100 km from
this center were identified. Then the district-
level temperature and rainfall as a weighted
average of their respective recorded values at
the identified grid points were calculated. The
inverse square roots of the distances between
these grid points and the district center were
used as corresponding weights. The daily
weather data, thus obtained, were then used to
construct various climatic variables for different
growing seasons.
According to the Directorate of Rice
Development (DRD), India (2014), autumn rice
was sown between mid-February and April and
harvested between June and July every year
in Assam. There were substantial variations in
sowing and harvesting dates across districts.
For the construction of climate data, the period
between the mid-points of the sowing and
harvesting seasons (i.e., March 24–June 30) as
the growing season for autumn rice was taken.
Similarly, the sowing and harvesting seasons for
winter rice were June–August and November–
December, respectively. For summer rice,
the corresponding seasons were December–
February and May–June. Accordingly, July 16–
November 30 was considered as the growing
season for winter rice and 16 January–31 May
for summer rice.9
Using the district-level daily temperature
and rainfall data as obtained above, data on four
weather variables for each of the above growing
seasons were constructed. From the temperature
data, the mean and standard deviation of daily
average temperature during a particular season
were calculated. From the daily rainfall data,
total rainfall and standard deviation of daily
rainfall for the respective seasons were also
calculated.
The agricultural data reported for the crop
year begins in July and ends in June of the
subsequent calendar year. For consistency, the
constructed seasonal data on climatic variables
to the corresponding crop year was mapped.10
The summary statistics of the data are presented
in Table 2.
RESULTS AND DISCUSSION
Since the study's objective was to examine
heterogeneity in climatic impacts on rice
yield across seasonal varieties, AC zones, and
yield distribution, the results are reported and
discussed in such a way that the differences
along these three dimensions are highlighted.
Although the focus was primarily on the
median/quantile regression results for reasons
discussed above, panel OLS results were also
reported for comparison.
9 The climate data were constructed using alternative
specifications of the growing seasons (instead of
using mid-points). The results were robust to this
alternative data construction.
10 For example, climatic data for the year 1978–1979
refer to a period from July 1978 to June 1979.

Variables Mean Median Maximum Minimum Std. Dev. Obs.
Autumn Rice
Yield (kg/ha) 992.4 926.0 2,224.0 18.0 390.4 594
Daily temperature (ºC) 25.3 25.3 27.1 23.4 0.7 594
Temperature variability (ºC) 2.1 2.1 3.7 1.3 0.4 594
Total rainfall (cm) 89.4 86.3 193.6 19.2 28.0 594
Rainfall variability (cm) 1.3 1.2 3.7 0.3 0.6 594
Fertilizer ('000 kg) 1,274.6 481.7 17,251.0 0.0 2,151.2 594
Area sown (ha) 25,008.9 16,302.5 90,000.0 1,443.0 21,386.1 594
Winter Rice
Yield (kg/ha) 1,380.0 1,345.5 2,441.0 407.0 329.7 594
Fertilizer ('000 kg) 1,223.1 515.0 14,872.0 0.0 1,900.9 593
Area sown (ha) 75,132.0 75,361.0 155,015.0 5,400.0 31,034.6 594
Summer Rice
Yield (kg/ha) 1,441.5 1,389.0 3,092.0 139.0 463.7 587
Fertilizer ('000 kg) 1,274.6 481.7 17,251.0 0.0 2,151.2 594
Area sown (ha) 6,323.1 1,500.0 59,833.0 1.0 10,988.5 590
Table 2. Descriptive statistics
Heterogeneity Across Seasonal Rice
Varieties
Table 3 presents the regression results
of the baseline specification for all three
seasonal rice varieties: autumn, winter, and
summer rice. The coefficient estimates for the
AC zone dummy variables were not reported.
The results showed that an increase in mean
daily temperature had a positive effect on
autumn rice yield and a negative impact on
winter and summer rice yield. However, none
of these effects was statistically significant. In
general, increased temperatures may decrease
crop yield due to spikelet sterility. As Rahman
et al. (2017) stated, “rice is hypersensitive
to high-temperature stress during panicle
development and meiosis causing anomalous
pollen maturity and absolute sterility.”11
The fact that no evidence of significant impact
11 A situation in which there is no grain within the
glumes of the rice plant.

Table 3. Median and panel OLS regression results for autumn, winter, and summer rice
Independent
Variable
Median Regression Panel OLS Regression
Autumn Rice Winter Rice Summer Rice Autumn Rice Winter Rice Summer Rice
Est. coeff t-stat Est. coeff t-stat Est. coeff t-stat Est. coeff t-stat Est. coeff t-stat Est. coeff t-stat
Temperature 27.83 1.28 –15.97 –0.69 –5.00 –0.24 29.78 1.32 –9.88 –0.43 –26.71 –1.01
Temperature
variability
72.88** 2.14 72.47** 2.15 6.01 0.35 63.49* 1.82 81.93*** 2.43 23.66 1.48
Rainfall –0.15 –0.19 1.19* 1.89 –1.18 –1.17 –0.76 –0.91 1.16* 1.82 –1.93 –1.53
Rainfall
variability
–38.32 –0.96 –106.13*** –2.53 82.85 1.42 40.83 0.99 –113.85*** –2.67 74.07 1.04
Trend 15.25*** 9.50 20.68*** 13.83 32.44*** 19.21 13.51*** 8.11 21.73*** 14.37 27.13*** 12.26
Fertilizer 0.001 0.16 0.01 0.87 0.03*** 3.42 –0.01 –0.88 0.001 0.23 0.04*** 3.96
Area –0.002** –2.19 0.00 1.23 0.004* 1.75 –0.002*** –2.68 0.001 1.53 0.003 1.39
Flood dummy –62.43** –2.22 36.36 1.45 100.61*** 3.80 –124.96*** –4.30 24.94 1.00 84.58*** 2.42
Pseudo
R-squared
0.39 0.36 0.34 0.57 0.57 0.49
No. of
observations
594 593 587 594 593 587
Notes: *** Significant at 1 percent level of significance
** Significant at 5 percent level of sign¬ificance
* Significant at the 10 percent level of significance

was found might indicate that the increase in
temperature does not yet represent the high-
temperature stress that adversely affects the
reproductive structure of rice in Assam.
Temperature variability had a positive
impact on yield for all three seasonal varieties
but statistically significant (at the 5% level)
only for autumn and winter rice. One unit
increase in temperature variability during the
growing season raised the yield by more than
72 kg per hectare for both autumn and winter
rice. As noted previously, day-to-day variations
in weather conditions such as temperature
and rainfall during the growing season have
crucial effects on the growth and yield of crops
(Mearns, Rosenzweig, and Goldberg 1996;
Schlenker and Roberts 2009).
While the exact science behind these effects
is under the purview of agrometeorology,
the result indicated that increased day-to-day
variations in temperature were beneficial for
autumn and winter rice in Assam. Rainfall had
a significant positive effect only on winter rice
yield.A1 cm increase in total rainfall during the
growing season led to an increase of more than
1 kg in median yield of winter rice. In general,
rice requires hot and humid conditions. Rainfall
increases relative humidity, which in turn
may contribute to higher rice yield. Rainfall
variability is harmful to autumn and winter rice
yield but beneficial to summer rice. However,
the negative effect was statistically significant
only for winter rice.
Among the control variables, technology
trend had significant positive effects on
yield for all three seasonal varieties.
The yield increase was largest for summer rice.
The effects of fertilizer consumption were
positive but statistically significant only for
summer rice. Note that these results might have
been influenced by the fact that fertilizer data
were not available separately for rice.12
Area had a statistically significant
negative effect on the yield of autumn rice
and a statistically significant positive effect
on summer rice. Although no information
about farm sizes was available, the negative
(or positive) effect may be indicative of
diseconomies (or economies) of scale. The
significant negative coefficient for the flood
dummy in case of autumn rice suggests that
yield was significantly lower in the districts that
were frequently and heavily affected by floods.
However, flood seemed to be beneficial to other
two seasonal varieties although the effect was
statistically significant only for summer rice.
The higher yield might be due to the fact that
the flood water leaves behind soil nutrients
beneficial to summer rice cultivation in the
flood affected districts. It might also be due to
the fact that farmers in flood affected districts
put more intensive efforts to its cultivation in
order to compensate for lower yields for the
other two seasonal varieties.13
The panel OLS estimates were qualitatively
similar but quantitatively different. Overall,
results indicated significant heterogeneity in the
effects of changes in temperature and rainfall on
yield across seasonal varieties. Since these rice
varieties were grown in three different seasons,
the results also reflected seasonal heterogeneity
of climatic impacts.
12 Barnwal and Kotani (2013) also noted this weakness
of the fertilizer data for Andhra Pradesh. They also
did not find any significant effects of fertilizer on rice
yield.
13 The findings from a field survey reported in Mandal
(2014) showed that prolonged water logging from
floods rendered winter rice cultivation impossible in
the frequently flood-prone areas of Dhubri district.
Hence, the farmers in these areas allocated a
considerable proportion of total cropped area to
summer rice and they put more intensive efforts to its
cultivation.

Heterogeneity Across AC Zones
The estimation results of the model
with interactions between the climate
variables (temperature and rainfall) and
AC zone dummies are presented in Table 4.
For convenience of interpretation, the net
effects of temperature and rainfall for each
AC zone were reported. That is, for AC
zone 1 (the base against which the impacts in
other zones were evaluated), the estimated
coefficient value of temperature (and rainfall)
was reported, and for zones 2, 3, 4, 5, and 6,
the sum of the estimated coefficients for zone
1 and those for the interaction terms between
temperature (and rainfall) and the respective
zone dummies were taken. The corresponding
standard errors for the net effects that were not
reported in the table were also calculated.14
Instead, the t-statistics are indicated.
The results showed that temperature had a
significant positive effect on autumn rice yield
in zone 2 (Central Brahmaputra Valley zone).
A 1°C increase in mean daily temperature led
to about 177 kg/ha increase in median yield for
this seasonal variety in that zone. Furthermore,
it had a positive effect on autumn rice yield
in zones 1, 3, and 4, and negative effects in
zones 5 and 6; although, these effects were not
statistically significant.
For winter rice, increase in temperature was
harmful in all but zones 2 and 6. However, the
negative effect was statistically significant only
14 The following formula was used to calculate the SEs
of the net effects:
where is the estimated SE of the estimated
coefficient of temperature for the qth quantile
(q = 25th, 33rd, 50th, 67th, and 75th) in zone 1;
is the SE of the estimated coefficient of the
interaction term between temperature and zone j
dummy for the same quantile in agro-climatic zone
j (j = 2, 3, 4, 5, and 6), and is the
estimated covariance between and .
in zone 5 (Hill zone). In contrast, the positive
effects in zones 2 and 6 were relatively large and
statistically significant. Finally, an increase in
temperature was beneficial to summer rice yield
in zones 1, 2, and 4. The positive effect was not
statistically significant in zone 4. In contrast,
the negative effect was large and statistically
significant only in zone 3.
Thus, an increase in temperature was found
to be beneficial to all three seasonal varieties of
rice in the Central Brahmaputra Valley zone,
while it was harmful for all three varieties
(although significantly so only for winter rice)
in the Hill zone. Additionally, it was favorable
to winter rice yield in the Barak Valley zone and
detrimental to summer rice yield in the North
Bank Plain zone.
The effect of an increase in total rainfall
during the growing season on autumn rice yield
was negative in all but zone 3. None of these
effects was statistically significant. In contrast,
rainfall had significant positive impacts on
winter rice yield in zones 1, 3, and 6 where a 1 cm
increase in total rainfall raised median yield by
1.43, 2.92, and 3.96 kg/ha, respectively. Finally,
rainfall was beneficial for summer rice in zones
1 through 3 and harmful in zones 4 through 6.
However, the effect was statistically significant
only in zone 3. Thus, higher rainfall was
beneficial to winter rice yield in the Lower
Brahmaputra Valley and North Bank Plain zone
(a contiguous geographic region to the north of
the Brahmaputra River) and the Barak Valley
zone in the southern part of Assam. The effects
of temperature and rainfall variability, and other
control variables were similar to those reported
in the previous sub-section.
As before, the panel OLS estimates
were qualitatively similar but there were
quantitative differences. Overall, the results
demonstrated substantial heterogeneity in
terms of magnitude, direction, and statistical
significance in the effects of temperature across
various AC zones of Assam. The differences

Table 4. Median and panel OLS regression results for autumn, winter, and summer rice across AC zones
Independent
Variable
Median Regression Panel OLS Regression
Autumn Rice Winter Rice Summer Rice Autumn Rice Winter Rice Summer Rice
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Temperature
AC zone 1 36.99 1.05 −40.12 −0.95 68.57* 1.76 48.00 1.29 −16.27 −0.38 1.31 0.03
AC zone 2 177.09*** 2.56 200.84** 2.08 174.87** 1.96 96.79 1.34 253.04*** 2.74 110.45 1.21
AC zone 3 56.61 1.36 −56.72 −1.47 −144.51*** −3.11 29.83 0.68 −89.16** −2.31 −160.91*** −2.97
AC zone 4 27.81 0.62 −42.12 −0.84 17.58 0.37 −0.68 −0.01 −30.83 −0.63 29.01 0.52
AC zone 5 −67.13 −1.08 −200.22*** −2.51 −59.95 −0.72 −23.52 −0.36 −93.92 −1.21 −116.55 −1.34
AC zone 6 −41.95 −0.71 251.56*** 3.24 −25.94 −0.35 −44.18 −0.70 264.55*** 3.44 56.41 0.70
Temperature variability 76.51** 2.26 120.65*** 3.42 4.28 0.25 50.79 1.40 100.56*** 2.87 20.85 1.24
Rainfall AC zone 1 −1.10 −0.84 1.43* 1.88 1.47 0.76 −1.81 −1.30 1.29* 1.71 0.86 0.36
AC zone 2 −1.01 −0.40 −0.09 −0.04 5.61 1.17 −1.59 −0.61 −1.23 −0.64 6.07 1.20
AC zone 3 0.65 0.57 2.92*** 3.35 3.63** 2.19 −0.06 −0.05 2.58*** 2.94 −1.00 −0.51
AC zone 4 −0.91 −0.95 0.44 0.53 −0.62 −0.47 −0.93 −0.91 0.64 0.78 −1.87 −1.17
AC zone 5 −4.23 −1.62 −0.03 −0.01 −2.38 −0.53 −2.32 −0.84 −0.42 −0.21 −2.58 −0.53
AC zone 6 −0.82 −0.69 3.96*** 2.92 −1.78 −1.12 −1.41 −1.12 3.79*** 2.82 −2.80 −1.53
Rainfall variability −8.41 −0.20 −117.38*** −2.73 34.86 0.59 56.85 1.30 −126.58*** −2.95 55.22 0.76
Trend 14.76*** 9.43 20.81*** 13.61 32.28*** 18.43 13.81*** 8.10 21.89*** 14.12 27.24*** 12.17
Fertilizer 0.001 0.10 −0.002 −0.30 0.02*** 2.39 −0.006 −0.87 −0.005 −0.71 0.036*** 3.27
Area −0.001 −1.59 0.001** 2.08 0.004* 1.70 −0.002*** −2.71 0.001 1.43 0.004 1.62
Flood dummy −74.04*** −2.67 65.74*** 2.62 93.76*** 3.27 −127.28*** −4.29 33.19 1.29 90.08*** 2.50
Pseudo R-squared 0.40 0.38 0.35 0.56 0.58 0.49
No. of observations 594 593 587 594 593 587
Notes: *** Significant at 1 percent level of significance; ** Significant at 5 percent level of significance; * Significant at 10 percent level of significance

in the effects of rainfall were relatively less
pronounced. Since AC zones represented not
only the spatial variations in climatic conditions
but also the differences in soil quality, changes
in temperature and rainfall and their variability
interacted with the soil quality to produce the
specific impacts as described above.
Heterogeneity Across Yield Distribution
The quantile regression results for the
20th, 40th, 60th, and 80th quantiles of the
yield distribution are presented here. Only the
estimated net effects of climate variables are
reported separately for autumn, winter, and
summer rice, respectively in Panels A, B, and
C of Table 5.
The effect of temperature on autumn rice
was positive for all quantiles in zones 1, 2, and
3 but statistically significant at conventional
levels only for the 80th quantile in zone 1, the
40th and 60th quantile in zone 2, and the 40th
quantile in zone 3. The significant positive
effects were also quantitatively larger than
those that were not significant. In contrast, the
impact of increase in temperature was negative
for the 20th and 80th quantiles in zone 4, all but
the 80th quantile in zone 5, and all quantiles in
zone 6. These negative effects were statistically
significant at the conventional levels for all but
the 40th quantile only in zone 6.
In contrast, significant negative effects
of increase in temperature on winter rice
yield was found for all but the 20th quantile
in zone 3, and only the 60th quantile in zone
5. In these two zones along with zones 1 and
4, the effect of temperature was negative for
almost all quantiles. However, temperature had
significant positive effects on winter rice yield
for all quantiles in zones 2 and 6. The positive
effects were also quantitatively larger. Finally,
temperature had significant negative impacts
on summer rice yield for all four quantiles
in zone 3 and the 20th quantile in zone 5.
These negative effects were also quantitatively
large. Furthermore, temperature had
significantly large positive effects on summer
rice yield for the 60th and 80th quantiles in zone
2, and for the 20th quantile in zone 6.
Temperature variability had statistically
significant positive impact on autumn and
winter rice yields for all four quantiles of their
distributions considered here. However, there
were quantitative differences across quantiles
with relatively larger impact on the lower end of
thedistribution.Finally,theeffectoftemperature
variability was significantly positive only for
the 80th quantile of summer rice distribution.
These results are broadly consistent with the
earlier results reported in Tables 3 and 4 but
highlighted the distributional heterogeneity in
climatic impact.
The effects of total precipitation on autumn
rice yield were positive for all quantiles in zone
3 and negative for most quantiles in other zones.
However, the negative effects were statistically
significant only for the 40th quantile in
zone 5 and 60th quantile in zone 6. In the case
of winter rice, rainfall had positive effects on
yield for all four quantiles in zones 1, 3, and 6.
However, these effects were statistically
significant only for the 60th quantile in zone
1, for all but the 20th quantile in zone 3, and
for the 20th and 60th quantiles in zone 6.
The positive impact of rainfall was also
statistically significant for the 60th quantile in
zone 4. The negative effects of rainfall for all
but the 60th quantile in zone 2, the 80th quantile
in zone 4, and the 20th and 80th quantiles
in zone 5 were not statistically significant.
Finally, rainfall had positive effects across the
yield distribution for summer rice in zone 2
and negative effects in zone 4, but these effects
were statistically significant only for the 40th
and 80th quantiles in zone 2. Moreover, the
positive effects were statistically significant
for the 80th quantile in zone 1 and for the 40th
and 80th quantiles in zone 3, while the negative

Table 5. Quantile regression results for autumn, winter, and summer rice
Independent Variable
20th Quantile 40th Quantile 60th Quantile 80th Quantile
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Panel A: Autumn Rice
Temperature
AC zone 1 10.30 0.27 26.59 0.84 47.19 1.26 72.89** 1.96
AC zone 2 69.11 0.86 140.61** 2.13 118.95* 1.64 114.72 1.45
AC zone 3 12.21 0.27 63.46* 1.69 48.87 1.11 41.71 0.93
AC zone 4 –17.54 −0.36 1.15 0.03 25.18 0.53 −51.16 −1.06
AC zone 5 −72.31 −0.99 −96.67 −1.63 −47.53 −0.72 38.06 0.54
AC zone 6 −131.39* −1.95 −80.13 −1.47 −151.09*** −2.41 −158.94*** −2.41
Temperature variability 103.91*** 2.88 84.33*** 2.84 91.19*** 2.54 90.72*** 2.53
Rainfall AC zone 1 −1.53 −1.06 −1.03 −0.88 −0.24 −0.17 0.52 0.36
AC zone 2 −1.15 −0.40 −0.33 −0.14 −2.28 −0.87 −2.01 −0.70
AC zone 3 0.27 0.22 0.56 0.54 0.62 0.52 1.40 1.13
AC zone 4 −0.61 −0.59 −0.45 −0.53 −0.40 −0.39 0.57 0.54
AC zone 5 −2.72 −0.88 −4.85* −1.97 −2.25 −0.82 −0.68 −0.23
AC zone 6 −0.99 −0.74 −0.90 −0.82 −2.14* −1.68 −1.14 −0.87
Rainfall variability 21.12 0.47 −0.79 −0.02 −7.43 −0.17 −25.30 −0.56
Panel B: Winter Rice
Temperature
AC zone 1 9.90 0.23 −37.33 −0.81 −41.68 −1.13 3.77 0.08
AC zone 2 234.05** 2.25 178.36* 1.73 207.85** 2.30 302.57*** 2.71
AC zone 3 −22.66 −0.56 −72.25* −1.72 −65.95* −1.90 −107.03*** −2.36
AC zone 4 57.64 1.09 −6.21 −0.11 −48.33 −1.07 −88.76 −1.54
AC zone 5 −11.70 −0.14 −112.49 −1.32 −189.73*** −2.50 −66.42 −0.72
AC zone 6 292.76*** 3.50 258.53*** 3.09 292.72*** 4.08 382.20*** 4.21
Temperature variability 109.31*** 3.10 103.51*** 2.71 106.26*** 3.49 70.42* 1.76
Rainfall AC zone 1 0.83 1.06 0.84 1.02 1.57*** 2.33 1.13 1.29
AC zone 2 −2.75 −1.26 −1.21 −0.57 0.31 0.16 −1.42 −0.61
AC zone 3 0.08 0.09 1.95** 2.05 2.83*** 3.65 3.33*** 3.25
AC zone 4 0.36 0.42 0.27 0.30 1.18* 1.65 −0.32 −0.33
AC zone 5 −2.11 −0.93 0.62 0.28 2.50 1.27 −1.76 −0.73
AC zone 6 7.12*** 5.00 2.40 1.63 2.61** 2.13 1.00 0.64
Rainfall variability −83.08* −1.89 −87.97* −1.87 −148.87*** −3.96 −128.06*** −2.58
** Significant at 5 percent level of significance
* Significant at 10 percent level of significance

Independent Variable
20th Quantile 40th Quantile 60th Quantile 80th Quantile
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Est.
coeff
t-stat
Panel C: Summer Rice
Temperature
AC zone 1 25.65 0.55 55.38 1.34 1.27 0.04 –4.47 –0.12
AC zone 2 42.75 0.39 140.52 1.48 161.98** 1.98 189.92** 2.21
AC zone 3 –209.06*** –3.68 –172.06*** –3.39 –173.48*** –4.08 –107.97*** –2.45
AC zone 4 15.42 0.27 –13.84 –0.27 46.60 1.07 23.15 0.53
AC zone 5 –220.33** –2.15 –59.49 –0.67 –38.42 –0.51 –69.75 –0.88
AC zone 6 155.16* 1.73 –12.92 –0.16 –23.22 –0.35 –90.35 –1.30
Temperature variability 10.47 0.54 11.57 0.66 –3.13 –0.18 112.32*** 6.38
Rainfall
AC zone 1 –0.20 –0.09 0.40 0.19 0.75 0.41 4.96*** 2.75
AC zone 2 3.37 0.57 10.04** 1.96 5.30 1.23 13.16*** 2.90
AC zone 3 –3.29 –1.62 3.34* 1.86 1.19 0.78 3.28** 2.11
AC zone 4 –1.66 –1.05 –1.85 –1.31 –1.04 –0.85 –0.45 –0.36
AC zone 5 –6.17 –1.12 –2.94 –0.61 –0.97 –0.24 3.09 0.72
AC zone 6 –4.00** –2.08 –3.90** –2.29 –0.63 –0.43 1.45 0.98
Rainfall variability 83.70 1.19 78.56 1.24 28.18 0.51 –140.38*** –2.51
Table 5. Continuation
** Significant at 5 percent level of significance
* Significant at 10 percent level of significance
effects were significant only for the 20th and
40th quantiles in zone 6.
The effects (positive or negative) of rainfall
variability on autumn rice yield were not
statistically significant for the four quantiles.
For winter rice, significant negative impact of
rainfall variability was robust across different
quantiles. In contrast, the positive effects of
rainfall variability on summer rice yield for
first three quantiles of its distribution were
not statistically significant. However, the
negative effect on the 80th quantile was large
in magnitude and highly significant. While
these results clearly demonstrate heterogeneous
impacts of climate change on rice yields along
different dimensions, the explanation for these
results would require in-depth investigation in
the fields of agronomy and agrometeorology,
which is beyond the scope of this paper.
To summarize, of the 24 cases (6 AC
zones × 4 quantiles) for each seasonal variety,
a change in temperature had statistically
significant impacts on yield in seven cases
(4 positive and 3 negative) for autumn rice,
in 12 instances (8 positive and 4 negative)
for winter rice, and in 8 cases (3 positive and
5 negative) for summer rice. Similarly, a change
in total rainfall had significant impacts on yield
in only 2 cases (both negative) for autumn rice,
in 7 cases (all positive) for winter rice, and in
7 cases (5 positive and 2 negative) for summer
rice. These results demonstrate substantial
heterogeneity in the impacts of temperature and
rainfall on yield across seasonal rice varieties,

AC zones, and the yield distribution of each
variety in all six zones.15
To gain a clear overall perspective on
these heterogeneities, the estimated coefficients
(net effects) for different quantiles for all
zones and all seasonal rice varieties were
plotted (Figure 2). For easy comparison
across seasonal rice varieties, the scale on the
vertical axis was kept constant across varieties.
This represents changes in rice yield per
hectare. Each chart in the figure clearly shows
the differences in the impact of temperature
and rainfall across yield distribution and AC
zones for each seasonal variety of rice. Looking
across charts, substantial variations can be seen
in the effects of temperature and rainfall across
seasonal rice varieties without losing sight of
within-variety heterogeneities.
That evidence of positive effects of
temperature and rainfall in several instances was
found is interesting, as people would generally
expect, and some other studies (e.g., Barnwal
and Kotani 2013) have shown these to have
negative impacts. However, the results seem
to be quite consistent with the relatively recent
findings that the effects of climate variables are
non-linear. For example, studies showed that
temperature has a positive impact on crop yield
until it reaches a threshold, after which, there is
a decline in crop yield. In the instances where
positive impact was found, the temperature
might not have reached the critical threshold
and, as such, positive effect was still observed.
A similar explanation may apply to the effects
of rainfall as well.
15 The alternative specifications for each of the three
seasonal rice varieties with additional variables for
irrigation, the use of high yielding variety (HYV)
seeds, and drought were estimated. In general, the
results with respect to the effects of temperature and
rainfall on rice yield were qualitatively not different.
To save space, these results were not included in this
paper.
CONCLUSIONS AND IMPLICATIONS
This paper presented evidence of
heterogeneity in climatic impacts on agricultural
yield. Applying the non-parametric quantile
regression technique to district-level data from
1978 to 2005, it examined heterogeneity in
the impacts of temperature and rainfall across
rice varieties grown in different seasons of the
year, AC zones, and distribution of rice yield
in Assam, India. The results suggested that,
in general, the effects of temperature on yield
were not statistically significant for any of the
three seasonal rice varieties. However, these
effects were not uniform in their magnitudes,
signs, and statistical significance across AC
zones and yield distribution for each variety of
rice. Similarly, there were wide variations in
the effects of total precipitation across seasonal
varieties, AC zones, and yield distribution.
The results also suggested that an increase in
temperature variability is beneficial and that
rainfall variability is harmful to autumn and
winter rice yield. For summer rice, although
the effects of these two variables were positive,
these were statistically insignificant.
Given the importance of rice yield for
food security and poverty alleviation in
Assam, these results could be informative in
designing appropriate adaptation strategies
and public policies to counter the adverse
impacts of climate change on agriculture in the
state. Furthermore, since most people in rural
areas are engaged in agriculture, these results
are important for the sustainability of rural
economies as well. For example, while positive
impacts of rising temperature should encourage
farmers in most parts of the Brahmaputra Valley
to grow more autumn rice, the adverse impacts
should discourage them to do so in the Hill zone
and in the Barak Valley zone.
In contrast, since temperature has
significant beneficial effects on winter rice
yield in the Barak Valley, the farmers should

Figure 2. Relative variations in the effects of temperature and rainfall on the distribution of rice yield
Effects of temperature: change in yield (kg/ha) due to an increase in mean daily temperature by 1º C during the growing season
Effects of rainfall: change in yield (kg/ha) due to an increase in total rainfall by 1 cm during the growing season
-200
-100
0
100
200
AC Zone 1 AC Zone 2 AC Zone 3 AC Zone 4 AC Zone 5 AC Zone 6
Autumn Rice
20th quantile 40th quantile 50th quantile
60th quantile 80th quantile
-300
-200
-100
0
100
200
300
400
500
Winter Rice
-300
-200
-100
0
100
200
300
Summer Rice
-6
-5
-4
-3
-2
-1
0
1
2
Autumn Rice
-4
-2
0
2
4
6
8
Winter Rice
-10
-5
0
5
10
15
Summer Rice

focus more on this seasonal variety of rice.
Similarly, temperature had adverse effects
on the yield of winter and summer rice in the
Central Brahmaputra zone but favorable effects
on autumn rice yield. Therefore, the farmers in
that zone may focus more on growing autumn
rice. Further, while increasing precipitation
is beneficial only for winter rice in the Barak
Valley zone, it is favorable only for summer
rice in the North Bank Plain zone. Thus, these
findings not only provide guidance as to which
cropping pattern to choose in order to ensure
food security but also inform the policymakers
to design relevant public policies that provide
incentives and necessary help to the farmers to
adopt the appropriate cropping strategies.
Finally, there are caveats that need to be
noted. First, and as earlier noted, it has been
beyond the scope of this paper to determine and
explain the possible agronomic reasons for the
directions and magnitude of temperature and
rainfall effects across ACs and seasonal rice
varieties. Second, the effects of climate change
may vary at different stages of rice production:
sowing/planting, growing, and harvesting.
By considering the period between sowing/
planting and harvesting, the growing period,
which is an important stage in rice production,
was primarily covered. Thus, there is scope
for examining climatic impacts on rice yield
during sowing/planting and harvesting period
as well. However, as discussed above, there are
substantial variations in sowing and harvesting
time across districts and locations. It would be
an onerous task to examine the climatic impacts
at these stages with so much heterogeneity.
This is a limitation of the present study and
would be a worthwhile subject for future
research.
REFERENCES
Abeysingha N.S., M. Singh, A. Islam, and V.K. Sehgal.
2016. “Climate Change Impacts on Irrigated Rice
and Wheat Production in Gomti River Basin of
India: A Case Study.” SpringerPlus 5 (1): 1250.
doi: 10.1186/s40064-016-2905-y
Adams R.M., B.H. Hurd, S. Lenhart, and N. Leary.
1998. “Effects of Global Change on Agriculture:
An Interpretative Review.” Climate Research 11
(1): 19–30.
Adams R.M., B.H. Hurd, and J. Reilly. 1999.
“Agriculture and Global Climate Change:
A Review of Impacts to U.S. Agricultural
Resources.” In the Report of Pew Center on
Global Climate Change, Arlington, VA, United
States of America.
Aggarwal, P.K. 2008. “Global Climate Change and
Indian Agriculture: Impacts, Adaptation and
Mitigation.” Indian Journal of Agricultural
Sciences 78 (10): 911–919.
Aggarwal, P.K., and R. Mall. 2002. “Climate Change
and Rice Yields in Diverse Agro-Environments
of India.” Climate Change 52 (3): 331–343.
Auffhammer, M., V. Ramanathan, and J.R. Vincent.
2006. “Integrated Model Shows That
Atmospheric Brown Clouds and Greenhouse
Gases Have Reduced Rice Harvests in India.”
In Proceedings of the National Academy of
Sciences of the United States of America 103
(52): 19668–19672.
———. 2012. “Climate Change, the Monsoon, and
Rice Yield in India.” Climate Change 111 (2):
411–424.
Barnwal, P., and K. Kotani. 2013. “Climatic Impacts
Across Agricultural Crop Yield Distributions:
An Application of Quantile Regression on Rice
Crops in Andhra Pradesh, India.” Ecological
Economics 87: 95–109.
Bryant, C.R., B. Smit, M. Brklacich, T. Johnson,
J. Smithers, Q. Chiotti, and B. Singh. 2000.
“Adaptation in Canadian Agriculture to Climatic
Variability and Change.” Climate Change 45 (1):
181–201.
Burney, J., and V. Ramanathan. 2014. “Recent Climate
andAir Pollution Impacts on IndianAgriculture.”
In Procedings of the National Academy of
Sciences of the United States of America 111
(46): 16319–16324.

Cline, W.R. 2007. Global Warming and Agriculture:
Impact Estimates by Country. Massachusetts
Avenue, Washington D.C.: Peterson Institute for
International Economics.
Deressa, T.T., and R.M. Hassan. 2009. “Economic
Impact of Climate Change on Crop Production
in Ethiopia: Evidence from Cross-Section
Measures.” Journal of African Economies 18
(4): 529–554.
Directorate of Agriculture (DoA), Government of
Assam. 1978–79 to 1996–97. Basic Agricultural
Statistics. Unpublished Documents. Directorate
Office of Khanapara, Guwahati, Assam.
Directorate of Economics and Statistics (DES),
Government of Assam. 1978–79 to 2005–
06. Statistical HandBook Assam. Annual
publications.
Directorate of Economics and Statistics (DES),
Department of Agriculture, Coperation and
Farmers Welfare, Government of India. n.d. Crop
Production Statistics. Accessed on 15 July 2014.
http://aps.dac.gov.in/APY/Public_Report1.aspx
Directorate of Rice Development. 2014. “Rice in
India: A Status Paper.” Accessed 23 July 2014.
http://drd.dacnet.nic.in/Status%20Paper%20
-%2002.htm
Dubey, S.K., S.K. Tripathy, G. Pranuthi, and R. Yadav.
2014. “Impact of Projected Climate Change
on Wheat Varieties in Uttarakhand, India.”
Journal of Agrometeorology 16 (1): 26–37.
Guiteras, R. “The Impact of Climate Change on
Indian Agriculture.” Unpublished manuscript,
Department of Economics, University of
Maryland, College Park, Maryland, 2009.
Kaiser, H.M., S.J. Riha, D.S. Wilks, D.G. Rossiter, and
R. Sampath. 1993. “A Farm-Level Analysis of
Economic and Agronomic Impacts of Gradual
Climate Warming.” American Journal of
Agricultural Economics 75 (2): 387–398.
Koenker, R., and G.W. Bassett. 1978. “Regression
Quantiles.” Econometrica 46 (1): 33–50.
Krishnamurthy, C.K.B. 2012. “The Distributional
ImpactsofClimateChangeonIndianAgriculture:
A Quantile Regression Approach.” Working
Paper 69/2012, Madras School of Economics,
Chennai, India.
Kumar, S.N., P.M. Govindakrishnan, D.N. Swaroopa-
rani, C. Nitin, J. Surabhi, and P.K. Aggarwal.
2015. “Assessment of Impact of Climate Change
on Potato and Potential Adaptation Gains in the
Indo Gangetic Plains of India.” International
Journal of Plant Production 9 (1): 151–170.
Lal, M., K.K. Singh, L.S. Rathore, G. Srinivasan, and
S.A. Saseendran. 1998. “Vulnerability of Rice
and Wheat Yields in NW India to Future
Changes in Climate.” Agricultural and Forest
Meteorology 89 (2): 101–114.
Lewandrowski, J., and D. Schimmelpfennig. 1999.
“Economic Implications of Climate Change for
U.S. Agriculture: Assessing Recent Evidence.”
Land Economics 75 (1): 39–57.
Mandal, R. 2014. “Flood, Cropping Pattern Choice and
Returns in Agriculture: A Study of Assam Plains,
India.” Economic Analysis and Policy 44 (3):
333–344.
Maps of India. n.d. https://www.mapsofindia.com/
Accessed on 15 June 2014.
Mearns, L.O., C. Rosenzweig, and R. Goldberg. 1996.
“The Effect of Changes in Daily and Interannual
Climatic Variability on CERES-Wheat: A
Sensitivity Study.” Climatic Change 32 (3):
257–292.
Mendelsohn, R., W.D. Nordhaus, and D. Shaw. 1994.
“The Impact of Global Warming on Agriculture:
A Ricardian Analysis.” American Economic
Review 84 (4): 753–771.
Mohandass, S., A. A. Kareem, T. B. Ranganathan,
and S. Jeyaraman. 1995. “Rice Production
in India under Current and Future Climates.”
In Modelling the Impact of Climate Change
on Rice Production in Asia, edited by R. B.
Mathews, M. J. Kropff, D. Bachelet, and H. H. V.
Laar, 165–181. Wallingford, Oxfordshire: CAB
International.
Molua, E.L. 2009. “An Empirical Assessment of the
Impact of Climate Change on Smallholder
Agriculture in Cameroon.” Global Planetary
Change 67 (3–4): 205–208.
North Eastern Space Application Centre (NESAC). n.d.
http://nesac.gov.in/spsc_asp.htm Accessed on 15
June 2014.
Poudel, S., and K. Kotani. 2013. “Climatic Impacts on
Crop Yield and its Variability in Nepal: Do They
Vary Across Seasons and Altitudes?” Climatic
Change 116 (2): 327–355.

Rahman, M.A., S.C. Kang, N. Nagabhatla, and
R. Macnee. 2017. “Impacts of Temperature and
Rainfall Variation on Rice Productivity in Major
Ecosystems of Bangladesh.” Agriculture and
Food Security 6: 10. doi: 10.1186/s40066-017-
0089-5
Rajeevan, M., J. Bhate, J.D. Kale, and B. Lal. 2006.
“High Resolution Daily Gridded Rainfall Data
for the Indian Region: Analysis of Break and
Active Monsoon Spells.” Current Science 91
(3): 296–306.
Rao, B.B., P.S. Chowdary, V.M. Sandeep,
V.U.M. Rao, and B. Venkateswarlu. 2014.
“Rising Minimum Temperature Trends over
India in Recent Decades: Implications for
Agricultural Production.” Global and Planetary
Change 117: 1–8.
Rathore, L.S., K.K. Singh, S. A. Saseendran, and A.K.
Baxla. 2001. “Modelling the Impact of Climate
Change on Rice Production in India.” Mausam
52 (1): 263–274.
Sanghi, A., and R. Mendelsohn. 2008. “The Impact of
Global Warming on Farmers in Brazil and India.”
Global Environmental Change 18 (4): 655–665.
Sarker, M.A.R., K. Alam, and J. Gow. 2012. “Exploring
the Relationship Between Climate Change and
Rice Yield in Bangladesh: An Analysis of Time
Series Data.” Agricultural Systems 112: 11–16.
Schlenker, W., and M. J. Roberts. 2006. “Nonlinear
Effects of Weather on Corn Yields.” Review of
Agricultural Economics 28(3): 391–398.
———. 2009. “Nonlinear Temperature Effects Indicate
Severe Damages to U.S. Crop Yields under
Climate Change.” In Proceedings of the National
Academy of Sciences of the United States of
America 106 (37): 15594–15598.
Singh, P.K., K.K. Singh, S.C. Bhan, A.K. Baxla,
S. Singh, L.S. Rathore, and A. Gupta. 2017.
“Impact of Projected Climate Change on Rice
(Oryza sativa L.) Yield Using CERES-Rice
Modeling Different Agroclimatic Zones of
India.” Current Science 112 (1): 108–115.
Srivastava, A.K., M. Rajeevan, and S.R. Kshirsagar.
2009. “Development of a High Resolution Daily
Gridded Temperature Data Set (1969–2005) for
the Indian Region.” Atmospheric Science Letters
10 (4): 249–254.
Wang, J., R. Mendelsohn,A. Dinar, J. Huang, S. Rozelle,
and L. Zhang. 2009. “The Impact of Climate
Change on China’s Agriculture.” Agricultural
Economics 40 (3): 323–337.
Yadav, M.K., R.S. Singh, K.K. Singh, R.K. Mall,
C.B. Patel, S.K. Yadav, and M.K. Singh. 2015.
“Assessment of Climate Change Impact on
Productivity of Different Cereal Crops in
Varanasi, India.” Journal of Agrometeorology 17
(2): 179–184.

raju & nath.pdf

More Related Content

Similar to raju & nath.pdf (20)

Recently uploaded (20)

raju & nath.pdf