See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/280034906

New climatic classification of Nepal

Article  in  Theoretical and Applied Climatology · July 2015

DOI: 10.1007/s00704-015-1549-0

CITATIONS READS

74 9,703

4 authors, including:

Ramchandra Karki Rocky Talchabhadel

University of Hamburg, Germany and Department of Hydrology and Meteorology, … Texas A&M AgriLife Research
32 PUBLICATIONS   477 CITATIONS    79 PUBLICATIONS   343 CITATIONS   

SEE PROFILE SEE PROFILE

Juha Aalto
University of Helsinki
61 PUBLICATIONS   1,278 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Topoclimate, land surface conditions and atmospheric feedbacks View project

Watershed and Morphological Dynamics of Chure Rivers, Nepal View project

All content following this page was uploaded by Rocky Talchabhadel on 12 June 2020.

The user has requested enhancement of the downloaded file.

Theor Appl Climatol (2016) 125:799–808
DOI 10.1007/s00704-015-1549-0
ORIGINAL PAPER
New climatic classification of Nepal
Ramchandra Karki 1 & Rocky Talchabhadel 1 & Juha Aalto 2 & Saraju Kumar Baidya 1
Received: 5 November 2014 / Accepted: 29 June 2015 / Published online: 11 July 2015
# Springer-Verlag Wien 2015
Abstract Although it is evident that Nepal has an extremely
wide range of climates within a short latitudinal distance, there
is a lack of comprehensive research in this field. The climatic
zoning in a topographically complex country like Nepal has
important implications for the selection of scientific station
network design and climate model verification, as well as
for studies examining the effects of climate change in terms
of shifting climatic boundaries and vegetation in highly sen-
sitive environments. This study presents a new high-
resolution climate map of Nepal on the basis of long-term
(1981–2010) monthly precipitation data for 240 stations and
mean air temperature data for 74 stations, using original and
modified Köppen–Geiger climate classification systems.
Climatic variables used in Köppen–Geiger system were cal-
culated (i) at each station and (ii) interpolated to 1-km spatial
resolution using kriging which accounted for latitude, longi-
tude, and elevation. The original Köppen–Geiger scheme
could not identify all five types of climate (including tropical)
observed in Nepal. Hence, the original scheme was slightly
modified by changing the boundary of coldest month mean air
temperature value from 18 °C to 14.5 °C in order to delineate
the realistic climatic condition of Nepal. With this modifica-
tion, all five types of climate (including tropical) were identi-
fied. The most common dominant type of climate for Nepal is
temperate with dry winter and hot summer (Cwa).
* Ramchandra Karki
rammetro@hotmail.com
1
Department of Hydrology and Meteorology, P.O. BOX 406 Naxal
Kathmandu, Nepal
2
Finnish Meteorological Institute, P.O. BOX 503 ,
00101 Helsinki, Finland
1 Introduction
Climate classification refers to grouping of similar climate con-
ditions for understanding the causes, meteorological condi-
tions, and climatic changes that can be expected for specific
types of climate. The Köppen–Geiger (hereafter KG) climate
classification (Köppen 1936; Kottek et al. 2006) is one of the
most widely used climatic classification systems. The classifi-
cation is based on monthly air temperature and precipitation,
and it has been widely used due to its simplicity in representing
the climates around the world.
Löhmann et al. (1993) have applied this classification to
the output from both atmospheric general circulation model
and coupled atmosphere–ocean circulation models and com-
pared these to maps of the Köppen classification using mod-
ern data sets and to Köppen’s 1923 map. Peel et al. (2004)
used KG system as the basis for grouping rivers by climate
types around the world for the comparison of runoff charac-
teristics. Peel et al. (2007) also prepared the updated global
climate map using Köppen’s method. Despite some criticisms
(Strahler 1971; Sanderson 1999), the scheme remains widely
used, although in some cases, slight modifications are applied.
For example, Stern et al. (2000) used a modified Köppen
classification to produce a new map of the climates of
Australia.
In Nepal, Nayava (1975) classified the climates using
Thornthwaite’s scheme (Thornthwaite 1931) which was fur-
ther generalized with elevation into five major climatic groups
based on temperature province (i.e., Tropical, Mesothermal,
Microthermal, Taiga, and Tundra) only. Detailed climatic
subgrouping based on a combination of temperature and hu-
midity province as in Thornthwaite’s original scheme was not
used for this spatial mapping of climates over Nepal.
Moreover, the climatic data used in Nayava (1975)) is limited
to only 15 stations, below an elevation of 3867 m providing
800
data for a period of only 5 years. These data and classification
limitations suggest that the map may not depict the realistic
current climatic condition of Nepal.
The climatic classification of Nepal is extremely challeng-
ing due to complex topographical conditions which produce
substantial spatial variation in variables such as local radia-
tion, air temperature, and precipitation pattern. However, such
classification has broad applications in selection of scientific
station network design, agricultural planning, and climate
change studies in terms of analyzing the shifting climatic
boundaries and vegetation in this highly sensitive environ-
ment. Therefore, despite potential limitations of classification
systems over mountainous terrain, in this study, we have
attempted to use the KG classification system in its original
and modified form, in a conjunction with a newly-developed
high-resolution gridded data set to generate an improved,
higher resolution climatic map of Nepal.
2 Study area
Nepal is a small landlocked and mountainous country
encompassing an area of 147,181 km2 stretching from 26°
22′ to 30° 27′ N and from 80° 04′ to 88° 12′ E. Though small
in size, the elevation of the country varies from 60 m in the
south to 8848 m in the north within a horizontal distance of
less than 200 km (Fig. 1). The country can be divided into five
physiographic regions. From south to north, and with increas-
ing altitude, these are the following: (a) Terai, (b) Siwaliks, (c)
Middle Mountains, (d) High Mountains, and (e) High
Himalaya (Kansakar et al. 2004). Terai is a low-lying plain,
and northwards, middle mountains and hills are elongated
east–west with a series of complex valleys breaks, such as
Kathmandu, Pokhara, Dang, and Surkhet. The climate of
Nepal is affected by two major weather systems, summer
monsoon circulation (June to September) and westerly circu-
lation (November to May). The influence of these two circu-
lation systems is different, with summer precipitation greater
in the southeast and westerly-derived winter precipitation
greatest in the northwest (Nayava 1980; Mani 1981). Nepal
as a whole receives approximately 80 % of its annual precip-
itation during the summer monsoon (Shrestha 2000).
3 Data and methods
3.1 Climate observation data
This study is based on normalized (relative to 1981–2010)
values of monthly mean air temperature data of 74 climate
stations and precipitation data of 240 stations from the
Department of Hydrology and Meteorology (DHM). In order
to cover the high altitude regions, data from four climate
R. Karki et al.
stations in the high altitude region (Khumbu region of
Nepal) up to 5050 m above sea level (asl) elevation main-
tained by EVK2CNR (http://www.evk2cnr.org) available for
a shorter period of at least 5 years were also used. Figure 1
presents the spatial distribution of stations used in the study.
The stations with highest and lowest mean annual air
temperature and precipitation used in this analysis are
presented in Table 1.
3.2 Köppen–Geiger’s classification method
Köppen’s climate classification recognizes five principal
groups of world climates that are intended to correspond with
five main vegetation groups. These five climatic groups are
tropical, arid, temperate, cold, and polar. Each of these cli-
mates is further divided into subdivisions based upon differ-
ences in the seasonal distribution of temperature and precipi-
tation. Table 2 presents the detailed subdivisions of the KG
climatic classification.
Many modifications proposed to the Köppen system exists;
here, we use the system from the Köppen–Geiger Handbook
(Köppen 1936), with the exception of the boundary between
the temperate (C) and cold (D) climates, where we follow Peel
et al. (2007) and use the scheme of Russell (1931) which uses
the temperature of the coldest month >0 °C, rather than
>−3 °C as used by Köppen in defining the temperate-cold
climate boundary (see Wilcock 1968; Essenwanger 2001;
Peel et al. 2007 for more details of this modification).
A description of the symbols and the criteria used to
define the Köppen–Geiger climate types is provided in
Table 2. The 30 possible climate types in Table 2 are
divided into 3 tropical (Af, Am, and Aw), 4 arid (BWh,
BWk, BSh, and BSk), 9 temperate (Csa, Csb, Csc, Cfa,
Cfb, Cfc, Cwa, Cwb, and Cwc), 12 cold (Dsa, Dsb,
Dsc, Dsd, Dfa, Dfb, Dfc, Dfd, Dwa, Dwb, Dwc, and
Dwd), and 2 polar (ET and EF) types.
In the original classification scheme, the boundary of
C and A types of climate on the basis of mean air
temperature is Thot ≥10 and 0<Tcold <18, but this criteria
is slightly modified for the objective classification of
the climate of Nepal in order to delineate the realistic
climatic conditions. According to Lillesø et al. (2005),
the low-lying Terai region of Nepal, ranging from 60 to
330 m asl in general lies in the lower tropical ecologi-
cal zones. Therefore, with the aim of covering this re-
gion of Nepal with tropical climate, the boundary of
coldest month mean air temperature value is shifted
from 18 to 14.5 °C in the modified scheme. Both 14
and 15 °C were also tried for the boundary but 14.5 °C
resulted in the best representation of the Terai region as
a tropical climate zone. All other classification schemes
are kept in their original form.
New climatic classification 801

Fig. 1 The spatial distribution of meteorological stations used in this study and the topography of Nepal

3.3 Spatial interpolation resampled to 1 km×1 km resolution by spatial averaging

In Nepal, spatial interpolation of air temperature and precipi-
tation can be challenging due to complex topography (Fig. 1).
Elevation plays a key role in the observed spatial pattern of
temperature, but it can be an important predictor for precipi-
tation as well (Daly 2006; Goovaerts 2000). To include eleva-
tion in the interpolations, we used the 30 m×30 m resolution
ASTER Digital Elevation Model (DEM) which was further
using ArcGis 10.1. Zonal statistics function. In addition to
elevation, latitude and longitude were also included in the
kriging to account for potential geographical trends in the
studied parameters.
Kriging is a widely used spatial interpolation method
which in its simplest form predicts a value for a location with
no data based on the spatial dependency (i.e., autocorrelation)
of observed values (Goovaerts 1997; Haylock et al. 2008;

Table 1 Stations with highest and lowest normal mean annual air temperature and precipitation

Station name Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Elevation (asl)

Mean air temperature (°C)

Karmaiya 16.7 20.1 24.6 28.9 30.1 30.3 29.5 29.5 28.9 26.8 22.9 18.8 25.6 131
Pyramid −8.2 −9.0 −4.5 −3.9 1.1 4.4 4.8 3.7 2.2 −1.6 −5.3 −5.8 −1.8 5050
Precipitation (mm)
Ghami (Mustang) 8.2 6.5 9.7 2.7 6.4 6.8 31.8 35.3 14.0 17.8 2.9 9.3 151.3 3465
Lumle 32.4 49.3 67.6 111.0 319.1 885.9 1468.9 1441.3 886.2 205.5 25.2 22.2 5514.7 1740
802 R. Karki et al.

Table 2 Detail subdivision of

Köppen–Geiger’s climatic
classification (The unit for
precipitation is millimeters (mm)
and for temperature is degrees
Celsius (°C))
1st 2nd 3rd Description Criteria
A Tropical Tcold ≥18
f Rainforest Pdry ≥60
m Monsoon Not (Af) and Pdry ≥100–MAP/25
w Savannah Not (Af) and Pdry <100–MAP/25
B Arid MAP<10×Pthreshold
W Desert MAP<5×Pthreshold
S Steppe MAP≥5×Pthreshold
h Hot MAT≥18
k Cold MAT<18
C Temperate Thot ≥10 and 0<Tcold <18
s Dry summer Psdry <40 and Psdry <Pwwet/3
w Dry winter Pwdry <Pswet/10
f Without dry season Not (Cs) or (Cw)
a Hot summer Thot ≥22
b Warm summer Not (a) and Tmon10≥4
c Cold summer Not (a or b) and 1≤Tmon10<4
D Cold Thot ≥10 and Tcold ≤0
s Dry summer Psdry <40 and Psdry <Pwwet/3
w Dry winter Pwdry <Pswet/10
f Without dry season Not (Ds) or (Dw)
a Hot summer Thot ≥22
b Warm summer Not (a) and Tmon10≥4
c Cold summer Not (a, b, or d)
d Very cold winter Not (a or b) and Tcold <−38
E Polar Thot <10
T Tundra Thot >0
F Frost Thot ≤0

MAP=mean annual precipitation, MAT=mean annual temperature, Thot =temperature of the hottest month,
Tcold =temperature of the coldest month, Tmon10 =number of months where the temperature is above 10, Pdry =
precipitation of the driest month, Psdry =precipitation of the driest month in summer, Pwdry =precipitation of the
driest month in winter, Pswet =precipitation of the wettest month in summer, Pwwet =precipitation of the wettest
month in winter, Pthreshold =varies according to the following rules (if 70 % of MAP occurs in winter then
Pthreshold =2×MAT, if 70 % of MAP occurs in summer then Pthreshold =2×MAT+28, otherwise Pthreshold =2×
MAT+14). Summer (winter) is defined as the warmer (cooler) 6-month period of AMJJAS and ONDJFM] (Peel
et al. 2007)

Aalto et al. 2013). More precisely, this means that closely method, with the same specifications for both air tem-
neighboring observations are more likely to have similar perature and precipitation:
values than observations farther apart (Legendre et al. 2002).
A general model of spatial variability can be expressed as: Zðx; y; eÞ ¼ β 0 þ β1 x þ β 2 y þ β3 e ð2Þ

Z ðsÞ ¼ μðsÞ þ εðsÞ
where Z(s) is the interpolated value at a given location,
μ(s) is the deterministic function for a trend component,
and ε(s) denotes the stochastically varying but spatially
dependent error term (Høst 1999). Many regionalized
variables show a trend structure when analyzed spatial-
ly. Therefore, the mean (μ) need to be estimated from
the data. The trend was modelled using the least-squares
ð1Þ where x and y are the longitude and latitude, respectively, and
e denotes the elevation.
In kriging (here a global approach), the spatial autocorrela-
tion of the residuals is modelled using a semivariogram
(Fig. 2), where the variance between the observations is pre-
sented as a function of distance between the data pairs (see
Goovaerts 1997 for more detailed description of the method).
Kriging interpolations were performed in R (R
Development Core Team 2011) using gstat (Pebesma 2004)
New climatic classification
Fig. 2 Two examples of the
empirical semivariograms for the
residuals of (a) mean annual
temperature and (b) mean annual
precipitation
on the normalized values (1981–2010 average) of all the me-
teorological variables (MAP, MAT, Thot, Tcold, Tmon10, Pdry,
Psdry, Pwdry, Pswet, Pwwet, Pthreshold) required in KG classification
system. For both the temperature- and precipitation-based pa-
rameters, the range parameter in the semivariogram model was
fixed to 1° (roughly equal to 111 km) to avoid singular matrix
errors. Such problems are common, when the variogram model
parameters cannot be estimated due to, for example, duplicated
observations or perfectly correlated predictors (Bivand et al.
2008). For temperature interpolations, the nugget parameter
was optimized separately by the model fitting function (See
Bivand et al. 2008). For precipitation, the nugget corresponded
to the standard deviation of the precipitation sum.
The interpolations were evaluated using a cross-validation
scheme; for every interpolation run, the observations were
randomly split to independent calibration (70 %) and evalua-
tion data (30 %). Subsequently, the interpolations produced
with the calibration data was compared with the evaluation
data. Two statistical measures were calculated: mean absolute
difference (MAD) and Pearson’s correlation coefficient
(Hofstra et al. 2008; Aalto et al. 2013).
The interpolated gridded air temperature- and precipitation-
based variables were used to generate a climate map of Nepal.
Since all locations that satisfy the B climate criteria also satisfy
one of the other (A, C, D, or E) climate criteria, the B climate
was identified first. The set of locations defined as having a B
climate is based on a combination of mean annual precipita-
tion and mean annual air temperature.
4 Results
The high-resolution interpolated maps for mean annual air
temperature and precipitation are presented in Figs. 3 and 4.
Air temperature variation in local features such as the rivers,
valleys, and mountains are well captured with this 1 km×1 km
spatial interpolation. From the interpolated map, it can be seen
that the mean annual air temperature in the Terai plains
803
reaches above 26 °C, while mountain tops have value below
−12 °C. The north–south gradient of air temperature following
the elevation pattern is clearly seen in Nepal (Fig. 3).
Based on the precipitation interpolation, three high rainfall
pockets are observed in Nepal (Fig. 4). Firstly, the station at
Lumle located North of Pokhara valley receives mean annual
rainfall over 5000 mm with highest rainfall in Nepal.
Secondly, pronounced annual precipitation exceeding
3500 mm occurs in and around the Gumthang area to the
south of Langtang Himalayan range, just northeast of the
Kathmandu. Thirdly, annual precipitation in excess of
3500 mm is also found east of Mount Everest region in
Num area. The lowest annual precipitation (<200 mm) is ob-
served in and around Mustang, Manang, and Dolpa regions
located at leeward side north of the Annapurna range.
Similarly, low precipitation regions are observed in the high
altitude regions (Fig. 4). The cross-validation statistics of all
the major variables used in the KG implied a good interpola-
tion accuracy for all the parameters (Table 3).
Based on the spatial interpolation of all the parameters, a
map of climates of Nepal was prepared using the original KG
climate classification system. The map showed four main cli-
mate types present in Nepal with the temperate climate C
being the dominant climate type by area (Fig. 5). The other
main climate types are polar climate E, cold climate D, and
arid climate B. The sub-climate types are identified as Cwa
(Temperate climate with dry winter and hot summer), Cwb
(Temperate climate with dry winter and warm summer),
Dwb (Cold climate with dry winter and warm summer),
Dwc (Cold climate with dry winter and cold summer), ET
(Polar Tundra climate), EF (Polar Frost climate), and BSk
(Arid Cold Steppe climate).
Because of the broad classification of the temperate climate
group Cwa using the 0 to 18 °C mean isotherm for the coldest
month and for hottest month≥22 °C, it resulted the most dom-
inant climate with more than 50 % areal coverage in Nepal.
This resulted in the Kathmandu airport at 1337 m elevation
(Hill area) and Simara Airport at 130 m (hot, low-lying Terai)
804
Fig. 3 Spatial variation of mean annual air temperature (°C) in Nepal
being classified into the same climate group despite their dra-
matic differences in ecological conditions (Fig. 5).
Therefore, this climate group was modified to give a more
realistic picture of the climate in Nepal. After modifications,
the mean air temperature range of the coldest month was lim-
ited to 14.5 °C (instead of the original 18 °C) and the coldest
month mean air temperature of 14.5 °C was categorized into
Tropical group.
With this updated classification, Aw (Tropical Savannah
climate) type climate can now be observed at all low altitude
Terai areas of Nepal. Figure 6 presents the climatic map of
Nepal based on modified KG classification. Table 4 shows the
percentage of areal coverage of climate types in Nepal with
this modification.
5 Discussion
The climate zoning from KG method is primarily based on
normal monthly mean air temperature- and precipitation-
R. Karki et al.
based parameters. The spatial variation in air temperature
and precipitation is strongly affected by latitude, longitude,
and elevation. Hence, the high-resolution spatial interpolation
of mean air temperature- and precipitation-based parameters
has been carried out with the consideration of altitude, lati-
tude, and longitude for better representation of variation of
these parameters over Nepal. The interpolation of major me-
teorological variables like mean annual precipitation and
mean annual air temperature parameters used in these climatic
classifications showed very good results. However, as com-
pared to mean annual precipitation which has a very high
spatial-temporal variation, mean annual air temperature inter-
polation shows a very good representation of general variation
of air temperature over Nepal due to its high dependency with
elevation. The air temperature variation over all the major
plains, river valleys, and complex mountain range has been
well represented. Similarly, precipitation interpolation has
well captured the three major high rainfall pocket areas
(around Pokhara, Gumthang, and Num areas) in Nepal where
mean annual precipitation ranges from 3500 mm to more than
New climatic classification 805

Fig. 4 Spatial variation of mean annual precipitation (mm) in Nepal

5000 mm. The wettest region in Nepal is the Pokhara and central Nepal with annual precipitation of more than
surrounding region located south of the Annapurna range in 5000 mm. The resulting high precipitation in the region is
due to the unique topographical settings. Furthermore, Dhar
and Nandargi (2005) attributed this to the large orographic
Table 3 The cross-validation statistics related to the major temperature- differences over very short distance because it promotes rapid
and precipitation-based parameter interpolations ascent and condensation of moist air. The lowest precipitation
Parameter name Cross-validation statistics
in Nepal is in Mustang, Manang, and Dolpa regions located
leeward side of massive Annapurna mountain range which is
Correlation Mean absolute mainly due to rain shadow effects (Dhar and Nandargi 2002).
difference Other high altitude areas located in the leeward side of moun-
tain range also have low precipitation.
Mean annual air temperature 0.99 0.01
As the kriging interpolation is known to be very sensitive to
Mean annual precipitation 0.78 0.28
the spatial coverage of the observation network (Legendre
Mean air temperature of the hottest month 0.98 0.02
et al. 2002), the quality of the analysis could be further im-
Mean air temperature of the coldest month 0.98 0.04
proved by incorporation of more station data. This is especial-
Precipitation of direst month in winter 0.61 0.02
ly important at the border regions, where the uncertainty of the
Precipitation of wettest month in summer 0.73 0.49
interpolation increases due to the station network becoming
Number of months where the mean air 0.94 0.02
temperature is above 10 °C
sparse. In our study, the spatial coverage of the meteorological
stations was good (Fig. 1), excluding very high altitude sites
806
Fig. 5 The climatic map of Nepal based on Köppen–Geiger classification
(>5050 m) and some regions in the western mountains of
Nepal.
There are many problematic issues related to the interpola-
tion of precipitation variables. Firstly, there is already uncer-
tainty related to precipitation measurements; the station cov-
erage has to be dense to capture the often very local rain
patterns. Mean annual precipitation however is straightfor-
ward to interpolate, since at long-term period, the rainfall field
is more or less continuous. In contrast, the quality of interpo-
lation of the precipitation of the driest month suffers from a
very local precipitation events and a lack of real trend
structures.
Mean air temperature, on the other hand, is strongly related
to altitude due to the vertical lapse rate, and the inclusion of
elevation as a covariate often improves the interpolation sig-
nificantly. Additionally, temperature as a continuous field of-
ten follows geographical trend (longitude and latitude) caused
by for example continental-maritime gradients, the location of
nearby deserts, and the direction of movement of weather
R. Karki et al.
systems among other things. Such issues make the interpola-
tion of mean air temperature potentially simpler compared to
precipitation.
The classification with original KG climate method reveals
only four of the main climate types in Nepal. With the original
classification scheme, areas with elevation range from 72 m
up to about 1800 m lie in Cwa (temperate climate with dry
winter and hot summer) type of climate covering more than
half the area of Nepal. This was due to the broad range of
temperature of coldest month defined to delineate types A
and C, which resulted in Kathmandu airport at 1337-m eleva-
tion and Simara Airport at 130 m into the same climate group
despite dramatic differences in climates between these two
locations.
The areas above this altitude range and up to about 3500 m
lie in Cwb (temperate climate with dry winter and warm sum-
mer) type. The small very dry area lying north of massive
Annapurna Mountain in the Mustang region of Nepal falls
in the Bsk (arid cold steppe climate). The areal coverage of
New climatic classification 807

Fig. 6 The climatic map of Nepal based on modified Köppen–Geiger classification

ET (Polar Tundra) climate is relatively larger, and only the
very high altitude mountain peaks above 6000 m of the coun-
try show EF (Polar Frost type of climate).
The areal coverage of D (cold) type of climate is relatively
small. Although the very high-resolution (1 km×1 km) inter-
polation of air temperature and precipitation data has resulted
in good representation of the Köppen’s climatic classification
scheme in the complex mountain ranges as well, it has not
identified the tropical climate of Terai plain due to its broad
classification of the temperate climate group Cwa using the 0
to 18 °C isotherm for the coldest month.
With this realization, the original KG classification method
was slightly modified with shifting the boundary of coldest
month mean air temperature between C and A climate to
14.5 °C from its original 18 °C in order to cover the Terai
region of Nepal with tropical climate type. This modification

Table 4 The areal coverage of

different climate types in Nepal Symbol Climate Area in km2 % area
based on modified Köppen–
Geiger climatic classification Aw Tropical Savannah 31078 21.1
BSk Arid Steppe cold climate 864 0.6
Cwa Temperate climate with dry winter and hot summer 46011 31.3
Cwb Temperate climate with dry winter and warm summer 33133 23.5
Dwb Cold climate with dry winter and warm summer 6232 4.2
Dwc Cold climate with dry winter and cold summer 669 0.5
ET Polar Tundra climate 26797 18.2
EF Polar Frost climate 2399 1.6
808
only affects the Cwa climate area keeping other climates areal
coverage same. By doing this, the Aw (Tropical Savannah)
type climate is also identified for low-lying areas below about
400-m altitude of Nepal. From the original scheme, the areal
coverage of Cwa climate is about 50 % of total area of Nepal
but with the modification around 20 % is covered by Aw
climate and only about 30 % by Cwa. Importantly, the mod-
ified scheme seems to provide a better general representation
of the climates of Nepal.
6 Conclusions
The climatic classification of Nepal was carried out using both
the original and a slightly modified KG system and high-
resolution interpolated data. Based on the original scheme,
only four of the major climate types were observed in Nepal
but with modification and all five types of climate (including
tropical) have been classified. Therefore, the modified scheme
seems to provide a more realistic representation of the cli-
mates in Nepal. This improved classification offers a powerful
tool for the optimizing the network of scientific station loca-
tions and climate model verification. This climatological map
developed here can serve as a valuable basis for the forthcom-
ing studies examining the effects of climate change in terms of
shifting climatic boundaries and vegetation in the highly sen-
sitive environment of Nepal.
Acknowledgments The authors would like to thank the Department of
Hydrology and Meteorology, Nepal, for the permission to use meteoro-
logical data. They also express their sincere thanks to Kamal Prakash
Budhathoki, Chiranjibi Bhetuwal, and Srijana Bista for their comments
and suggestions during the preparation of this manuscript. We are also
grateful to Dr. Lindsey Nicholson for reviewing the manuscript and im-
proving the English. Finally, the authors wish to thank an anonymous
reviewer for the critical and constructive comments.
References
Aalto J, Pirinen P, Heikkinen J, Venäläinen A (2013) Spatial interpolation
of monthly climate data for Finland: comparing the performance of
kriging and generalized additive models. Theor Appl Climatol 112:
99–111
Bivand R, Pebesma E, Còmez-Rubio V (2008) Applied spatial data anal-
ysis with R. Springer, New York
Daly C (2006) Guidelines for assessing the suitability of spatial climate
data sets. Int J Climatol 26:707–721
Dhar ON, Nandargi S (2002) Precipitation distribution around the
Annapurna range of Nepal Himalayas—a brief appraisal. J
Meteorol 27:377–382
Dhar ON, Nandargi S (2005) Areas of heavy precipitation in the Nepalese
Himalayas. Weather, Vol. 60, No. 12, RMETS, UK
Essenwanger OM (2001) Classification of climates. In World Survey of
Climatology 1C, General Climatology, Elsevier, Amsterdam
View publication stats
R. Karki et al.
Goovaerts P (1997) Geostatistics for natural resources evaluation.
Applied Geostatistics Series, Oxford University Press
Goovaerts P (2000) Geostatistical approaches for incorporating elevation
into the spatial interpolation of rainfall. J hydrol 228:113–129
Haylock M, Hofstra N, Klein Tank A, Klok E, Jones P, New M (2008) A
European daily high‐resolution gridded data set of surface tempera-
ture and precipitation for 1950–2006. J Geophys Res Atmos 113:
D20
Hofstra N, Haylock M, New M, Jones P, Frei C (2008) Comparison of six
methods for the interpolation of daily, European climate data. J
Geophys Res 113:D21,110
Høst G (1999) Kriging by local polynomials. Comput Stat Data Anal 29:
295–312
Kansakar SR, Hannah DM, Gerrard J, Rees G (2004) Spatial pattern in
the precipitation regime of Nepal. Int J Climatol 24:1645–1659
Köppen W (1923) Dle Klirnate der Erde. Walter de Gruyter, Berlin
Köppen W (1936) Das geographisca System der Klimate. In: Köppen W,
Geiger G (ed) Handbuch der Klimatologie. 1. C. Gebr, Borntraeger,
pp 1–44
Kottek M, Grieser J, Beck C, Rudolf B, Rubel F (2006) World map of the
Köppen-Geiger climate classification updated. Meteorol Z 15:259–
263
Legendre P, Dale MRT, Fortin MJ, Gurevitch J, Hohn M, Myers D (2002)
The consequences of spatial structure for the design and analysis of
ecological field surveys. Ecography 25:601–615
Lillesø JP, Shrestha TB, Dhakal LP, Nayaju RP, Shrestha R (2005) The
map of potential vegetation of Nepal: a forestry/agro-ecological/bio-
diversity classification system. Københavns Univ Dev Environ No.
2/2005
Löhmann U, Sausen R, Bengtsson L, Cubasch U, Perlwitz J, Roeckner E
(1993) The Köppen climate classification as a diagnostic tool for
general circulation models. Max-Planck-Institutfür Meteorol,
Report No. 92
Mani A (1981) The climate of Himalaya. In: Lall JS, Moddie AD (eds)
The Himalaya: aspects of changes. Oxford University Press, New
Delhi, pp 3–15
Nayava JL (1975) Climates of Nepal. Himal Re, l. VII
Nayava JL (1980) Rainfall in Nepal. Himal Rev 12:1–18
Pebesma E (2004) Multivariable geostatistics in S: the gstat package.
Comput Geosci 30:683–691
Peel MC, McMahon TA, Finlayson BL (2004) Continental differences in
the variability of annual runoff—update and reassessment. J Hydrol
295:185–197
Peel MC, Finlayson BL, McMahon TA (2007) Updated world map of the
Köppen-Geiger climate classification. Hydrol Earth Syst Sci 5:
1633–1644
R Development Core Team (2011) R: a language and environment for
statistical computing. R Foundation for Statistical Computing,
Vienna, Austria. http://www.R-project.org, ISBN 3-900051-07-0
Russell RJ (1931) Dry climates of the United States: I climatic map. Univ
Calif Publ Geogr 5(1–41):1931
Sanderson M (1999) The classification of climates from Pythagoras to
Koeppen. Bull Am Meteorol Soc 80:669–673
Shrestha ML (2000) Interannual variation of summer monsoon rainfall
over Nepal and its relation to Southern Oscillation Index. Meteorol
Atmos Phys 75:21–28
Stern H, De Hoedt G, Ernst J (2000) Objective classification of Australian
climates. Aust Meteorol Mag 49:87–96
Strahler AN (1971) The Earth Sciences. Harper and Row, New York
Thornthwaite CW (1931) The climates of North America according to
new classification. Geogr Rev XXXI:633–655
Wilcock AA (1968) Köppen after fifty years. Ann Assoc Am Geogr 58:
12–28