Vol. 13(1), pp.

52-65, January 2019

DOI: 10.5897/AJEST2018.2586
Article Number: 179801159491
ISSN: 1996-0786
Copyright ©2019 African Journal of Environmental Science and
Author(s) retain the copyright of this article
http://www.academicjournals.org/AJEST Technology

Full Length Research Paper

Understanding nexuses between precipitation changes

and climate change and variability in semi-arid
lowlands of Mwanga District, Tanzania
Bagambilana Francis Rweyemamu and Rugumamu William Mulokozi*
Department of Geography, College of Humanities, University of Dar es Salaam, Tanzania.
Received 25 September, 2018; Accepted 12 November, 2018

Contextually, precipitation fluctuated due to climate variability and evolved due to emissions of
anthropogenic greenhouse gases and land-use changes since the industrial revolution in the 1880s.
However, some studies problematize that there is little understanding of nexuses between precipitation
changes and climate change and variability in tropical Africa. Therefore, this paper sought to assess
such linkages in semi-arid lowlands of Mwanga District, Tanzania. The findings revealed statistically
significant decrease of annual rainfall, Kendall’s tau rτ (44) = -.230, p = .019 and Kendall’s tau rτ (39) = -
.223, p = .024, at Same Meteorological station (1970 to 2009 and 2012 and 2015) and Nyumba ya Mungu
Meteorological stations (1977 to 2015), respectively; thus confirming occurrence of human-induced
climate change in the study area. Also, the findings revealed statistically significant correlations
between amounts of rainfall (September – February) and Niño 3.4 index and between amounts of rainfall
(October – December) and dipole mode index at both stations, hence confirming that precipitation
changes during short rainy seasons (Vuli) in the lowlands were significantly influenced by cycles of El
Niño-southern oscillation and Indian Ocean dipole. Besides, branched and isoprenoid tetraether (BIT)
indices revealed that wettest conditions, due to climate variability. This occurred from 650 to 950 CE
(common era), 1550 to 1700 CE, 22 to 25 ka BP (before present, present defined as 1950). Whereas,
driest conditions occurred from 1968 to 1974, 1780 to 1820 CE, 1170 to 1300 CE (between ca. 0.8 and 0.6
ka BP), 11.7 to 13.1 (the Younger Dryas), 15 to 18 (Heinrich 1 stadial) and 23.4 (Heinrich 2 stadial) ka BP.
Lastly, the paper recommends enhancement of traditional and modern-day environmental knowledge
systems with regard to weather forecasting and prediction.

Key words: Branched and isoprenoid tetraether index, climate change, climate variability, El Niño-southern
oscillation, Indian Ocean dipole, and short rainy season.

INTRODUCTION

According to the intergovernmental panel on climate variability) and evolves (climate change) across space
change (IPCC) (2013), climate fluctuates (climate and over time. On the one hand, the main drivers of

*Corresponding author. E-mail: wrugu@udsm.ac.tz.

Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution
License 4.0 International License
 Bagambilana and Rugumamu 53

climate variability include the Indian Ocean dipole (IOD), precipitation patterns over land, and intensification of
El Niño-southern oscillation (ENSO), Pacific decadal heavy precipitation over land regions where data were
oscillation (PDO) and Atlantic multidecadal oscillation sufficient (medium confidence). Under a representative
(AMO). Initially driven by land-sea temperature gradients, concentration pathway (RCP)8.5 scenario, mean
atmospheric teleconnections of coupled atmosphere- precipitation will likely decrease in many mid-latitude and
ocean modes of climate variability lead to extreme subtropical dry regions by the end of the 21st century but
weather and climate events, such as droughts and floods will likely increase in many mid-latitude wet regions and
at local, meso, synoptic, and planetary spatial scales in the East African region.
(Pokhrel et al., 2012; IPCC, 2013; Cassou et al., 2018; On the contrary, some studies, including those
Yeh et al., 2018). Cassou et al. (2018) explained, for undertaken by human-induced climate change skeptics
instance, that devastating droughts and famines of the and deniers, suggested that the frequency and intensity
1970s and 1980s in the Sahel were associated with of extreme weather events (droughts, floods, and storms)
changing modes of AMO and PDO. had either decreased or did not reveal any particular
On the other hand, the main drivers of human-induced trend during recent decades (Stewart, 2008; Inhofe,
climate change, as explained by the theory of 2012; Brown, 2013). A study conducted by Laliberté et al.
greenhouse effect, are accumulation of anthropogenic (2015) explained, that warmer periods including the
greenhouse gases in the atmosphere and land-use Medieval Climate Anomaly circa (ca.) 900 - 1200 CE and
changes (UN, 1992). To the contrary, however, some the 20th/21st centuries were associated with a decrease
scientists, including those affiliated with the Heartland of extreme weather events and that cool periods
Institute (Bast, 2010) and a nongovernmental international including the Little Ice Age ca. 1300 to 1900 CE were
panel on climate change (NIPCC) are skeptical of associated with an increase of unstable and intense
human-induced climate change or deny the existence of weather. IPCC (2013) and Zheng et al. (2018) also
a consensus on the science of human-induced climate supported such findings.
change (Stewart, 2008; Inhofe, 2012; Brown, 2013; Idso Besides, Verschuren et al. (2000) explained that there
et al., 2016). is a need to understand long-term precipitation-climate
It should be understood that this study adopted a change and variability nexus in order to enhance, among
definition of climate change as developed by the United other things, water-resource and land-use management
Nations framework convention on climate change in semi-arid regions of tropical Africa. Indeed, livelihoods
(UNFCCC). Specifically, UNFCCC stressed that climate of the majority of people in semi-arid tropical Africa were
change: pegged on rain-fed agricultural systems (encompassing
sub-systems of crop, livestock, fishery, and forestry
‘Means a change of climate which is attributed directly or production) that were likely to be exacerbated by the
indirectly to human activity that alters the composition of interplay between climate change and variability and non-
the global atmosphere and which is in addition to natural climate drivers and stressors. It should be noted,
climate variability observed over comparable time however, that an understanding of precipitation-climate
periods’ (UN, 1992:7). change and variability nexus in the region was limited by
lack of long-term instrumental time series of precipitation.
Indeed, IPCC’s definition reveals that climate change is Also, limited high-resolution and well-dated proxy
either caused by natural internal processes/dynamics or records, such as ice cores and tree rings particularly in
external forces, such as modulations of solar cycles areas experiencing bimodal rainfall regime due to
(sunspots), volcanic eruptions, orbital parameters biannual passage of inter-tropical convergence zone
(eccentricity, precession, and obliquity), and anthro- (ITCZ) (Verschuren et al. (2009).
pogenic changes in the composition of the atmosphere Indeed, majority of residents in semi-arid lowlands of
and land-use change (IPCC, 2013). Mwanga District, Tanzania, which covered 69.4% of the
Besides, modeling studies projected that shifts of mean total area of 264,100 ha frequently, faced climate-related
climate would increase the frequency, intensity, duration, stressors, such as meteorological/agricultural droughts,
spatial extent and timing of some extreme weather and dry spells, and floods (MDC, 2016). These has led to
climate events (IPCC, 2012; Ostberg et al., 2018). Higher water shortages, land degradation, frequent crop failures,
temperatures would lead to, for instance, increasing rates famines (locally known as Nzota), and death of domestic
of evapotranspiration, hence increasing severity of animals (Mvungi, 2008; Mashingo, 2010; IFRC and RCS,
droughts on agricultural systems particularly in semi-arid 2013). Such attributes suggested that precipitation
areas (FAO et al., 2018). changes in semi-arid lowlands of Mwanga District either
According to IPCC (2013), it is very likely that the reflected human-induced climate change or natural
global water cycle was affected by anthropogenic climate variability or both climate change and variability.
influences since the 1960s. Such influences (including Coupled with instrumental time series of rainfall and high
increasing atmospheric temperatures) led to increases in resolution and well-dated branched and isoprenoid
atmospheric moisture content, global-scale changes in tetraether (BIT) indices for paleo-precipitation, this study
54 Afr. J. Environ. Sci. Technol.

Figure 1. Location of meteorological stations and a hydroclimate site for study area in Mwanga District, Tanzania.

was undertaken in order to understand linkages between METHODOLOGY

precipitation changes in semi-arid lowlands of the district
Study area
and both climate change and modes of climate variability
(El Niño-southern oscillation and Indian Ocean dipole) Mwanga district (Figure 1) is one of the seven districts of
across different temporal scales. Kilimanjaro Region in northeastern Tanzania and it extends
 Bagambilana and Rugumamu 55

between latitudes 3°46’S and 3°47’ S and between longitudes (2016) and Sinninghe Damsté et al. (2012), respectively. It should
37°35’E and 37°50’E (MDC, 2016). Furthermore, the district covers be noted that Lake Challa is at an altitude of 880 m a.s.l on the
2,641 km2 and borders Moshi Rural district and the republic of lower east slope of Mount Kilimanjaro and along the border
Kenya in the North, republic of Kenya and Lake Jipe in the East, between Tanzania and Kenya. Trend analysis (moving averages)
Same district in the South, and Simanjiro district, NyM Reservoir, was conducted on BIT time series in order to establish patterns of
and Moshi rural district in the West. Specifically, the district’s land paleo-precipitation with regard to past cycles of climate variability.
areas cover 2,558.6 km2 while water bodies cover 82.4 km2. With Moreover, datasets of rainfall, Niño 3.4 index, DMI, and BIT were
regard to water bodies, the NyM reservoir covers 56 km 2 and Lake entered into spreadsheets (IBM SPSS 20 and or 2007 Excel
Jipe covers 26.4 km2 (MDC, 2016). software) and processed into desired formats. Specifically, datasets
Furthermore, the regime of rainfall in Mwanga district was of rainfall were cleaned for errors and missing observation values
bimodal and largely controlled by annual shift in the position of before conducting data analysis. In this regard, the main data
ITCZ and the low-level northeasterly and southeasterly trade cleaning methods were spot-checking, reviewing, and logic checks.
winds/monsoons (Moernaut et al., 2010). Specifically, strong and It should be noted, however, that a period of two years (January
dry winds tended to blow from East to the West (MDC, 2016). 2010 to December 2011) with missing rainfall time series at Same
Additionally, both ITCZ and trade winds/monsoons were driven by Meteorological Station, as original time series had not been
latitudinal differences for radiation, which was received at the corrected, was not included in the analysis of meteorological data.
Earth’s surface (Moernaut et al., 2010). Consequently, long rains Additionally, spatial interpolation could not be conducted with
(Mbua ja Mashikaa in Pare language) were experienced from rainfall time series from three meteorological stations (Same, NyM,
March to June and short rains (Mbua ja Vhuri in Pare language) and Moshi Airport) (Figure 1) as there were no other
were experienced from October to January (MDC, 2016). meteorological/weather stations with relatively longer time series in
In addition, the amount of rain varied spatially. Highest amount of Mwanga District. According to Tao et al. (2009), the use of
rain (800-1350 mm/year) fell in the North Pare Mountains and on its deterministic methods (including Inverse Distance Weighted and
eastern windward slopes to the Indian Ocean and lowest amount of Spline) and geostatistical methods (including Kriging) of spatial
rain (400-600 mm/year) fell in the eastern and western lowlands interpolation of climate parameters is worthwhile when there is
(URT, 2004; MDC, 2016). Besides, temperatures ranged from an sufficient density of meteorological/weather stations.
average minimum of 14°C between July and August to a maximum Besides, cleaned datasets of quantitative data were analyzed in
average of 32°C between January and February. IBM SPSS 20 and or 2007 Excel’s Analysis ToolPak by computing
According to IPCC (2013), shifts of climate mean(s) can reliably univariate and bivariate statistics. Specifically, univariate statistics
be detected by analyzing time series data that cover a period of were computed using the following formulas:
about 50 years. In this regard, attempts were made to source
rainfall time series covering longer time periods from 1) Mean (M)
meteorological/weather stations within Mwanga District or nearest
meteorological/weather stations in Same District (Figure 1). It is n
worth pointing out that, Pare District was split in 1978 to form x i
Mwanga District in the northern part (MDC, 2016) and Same District
in the southern part. x 1
(1)
However, rainfall time series for NyM Meteorological Station
n
(station code 9337090) and Same Meteorological Station (station
code 9437003) that were sourced from Tanzania meteorological Where, = sample mean, n = number of values, = individual
agency (TMA) and Pangani basin water board (PBWB) covered 39 value of the sample or measurement (where i goes from 1 to n).
and 44 years, respectively. Indeed, both stations are found in semi-
arid lowlands but attempts to harmonize attendant time series 2) Range
would have decreased reliability of analyses for detecting climate
change. Otherwise, the same time series were useful for
establishing inter-annual climate variability. Besides, rainfall time (2)
series covering a period of 62 years (1938 to 1999) at Moshi
Meteorological Station (station code 9337004) were sourced from Where, n = range, = maximum value, = minimum value.
Røhr (2003) as originally sourced from PBWB. Indeed, such time
series were used occasionally for comparison with time series for 3) Standard deviation (SD)
NyM and Same Meteorological Stations. Generally, the computed
standardized coefficients of skewness and kurtosis for rainfall time
 x  x 
2
series were positively skewed and < 1.96, hence the coefficients i
revealed approximate normal distributions.
s (3)
Also, time series of Niño 3.4 index (sea surface temperature n 1
anomalies averaged for areas extending between 5°N and 5°S and
between 120° and 170°W in the tropical Pacific Ocean), covering a
period of 46 years (1970 to 2015) were accessed through online Where, s = standard deviation, = sample mean, n = number of
visits from the national oceanic and atmospheric administration values, = individual value of the sample or measurement (where
(NOAA). Similarly, time series of dipole mode index (DMI) covering i goes from 1 to n).
a period of 46 years (1970 to 2015) were accessed through online
visits from NOAA. Indeed, DMI time series represented anomalous 4) Anomaly (deviation score)
SST gradient between the western equatorial Indian Ocean
(extending between 50°E and 70°E and between 10°S and 10°N)
di = (4)
and the southeastern equatorial Indian Ocean (extending between
90° E and 110°E and between 10°S and 0°N).
where, di = deviation score for the ith observation in a set of
Moreover, time series for well-resolved BIT index covering
observations, x = raw score for observation in a set of observations,
periods of the last 2.2 and 25 ka for a hydroclimate of Lake Challa
(Figure 1) were accessed online as originally used by Buckles et al. = mean of all values in a set of observations.
56 Afr. J. Environ. Sci. Technol.

5) Coefficient of variation V = Crenarchaeol isoGDGTs.

br = branched, iso = isoprenoid, and GDGT = glycerol dialkyl

glycerol tetraether
(5)
In this regard, times with high BIT values implied existence of drier
conditions and times with low BIT values implied existence of wetter
Where, cv = coefficient of variation, s = standard deviation, = conditions (Damsté et al., 2012; Buckles et al., 2016).
sample mean.
Furthermore, findings on trends/moving averages were also
presented as linear graphs. Additionally, statistical significance of
Kendall’s tau test was undertaken basing on the following
RESULTS AND DISCUSSION
hypotheses:
This section presents findings and discussions of
H0: There was no trend (data is independent and randomly findings. Specifically, it focuses on instrumental rainfall
ordered). changes in the study area, nexuses between instrumental
H1: There was a trend. rainfall changes in the study area and El niño–southern
In this regard, Kendall’s tau rτ was computed using the following oscillation and Indian Ocean dipole coupled with BIT
formula: indices for paleo-precipitation changes in the study area.

6) Kendall’s tau rτ
Instrumental rainfall changes in the study area
(6)
To begin with, mean annual rainfall at Same
Meteorological Station decreased by 235.2 mm from
631.1 mm with a relatively higher SD of 239.4 (1970 to
Where, = Kendall’s tau, n = sample size, = number of
1979) to 395 mm with a relatively lower SD of 73.2 (2012
concordant pairs, = number of discordant pairs. to 2015) (Table 1). Similarly, mean annual rainfall at NyM
Besides, bivariate analyses encompassed computation of Meteorological Station decreased by 250.3 mm from
Pearson’s correlation coefficient through the following formula:
611.6 mm with a relatively higher SD of 170.7 (1977 –
7) Pearson’s correlation coefficient: 1979) to 361.3 mm with a relatively with lower SD of 77.1
(2010 to 2015) (Table 2). Generally, the variability of
n xy   x  y (7) annual rainfall at NyM Meteorological Station, typified by
r SD of 157.5 for a mean of 342.9 mm and an average

n x 2
  x 
2
 n y 2
  y 
2
 coefficient of variation of 0.5 (2077 to 2015), was
relatively higher than the variability of rainfall at Same
Meteorological Station. This was typified by SD of 202.1
for a mean of 540.5 mm and an average coefficient of
Where, n = number of pairs of score, sum of the variation of 0.4 (1970 to 2009 and 2012 to 2015) (Tables
products of paired scores, = sum of x scores, = 1 and 2).
sum of y scores, = sum of squared x scores, = sum of Also, the findings revealed, during a period of 44 years
squared y scores. (1970 to 2009 and 2012 to 2015), increasing negative
It is worth pointing out that alpha level (α) was set at 0.05 and the anomalies of annual rainfall from a reference value of
central limit theorem was invoked since the sample size was 567.2 mm (an average for a period of 30 years from
sufficiently large (usually N ≥ 30) (Field, 2013). Consequently, the
January 1977 to December 2006) at Same Meteorological
findings based on analysis of quantitative data were largely
presented in tables and graphs. Station (Figure 2). Additionally, the fitted linear trend line
Besides, time series of BIT (branched and isoprenoid tetraether) is statistically significant at alpha level, α, of 0.05
indices, based on GDGTs (glycerol dialkyl glycerol tetraethers) as (Kendall’s tau rτ (44) = -0.230, p = 0.019) (Table 3).
proxies for paleo precipitation of Lake Challa area (Figure 1), were Similarly, the findings revealed, during a period of 39
sourced through online visits. Indeed, the time series covering a years (1977 - 2015), increasing negative anomalies of
period of the last 2.2 ka were originally used by Buckles et al.
(2016) and time series covering a period of the last 25 ka were
annual rainfall from a reference value of 345.5 mm (an
originally used by Sinninghe Damsté et al. (2012). Additionally, average for a period of 30 years from January 1977 to
authors through the following formula computed the time series: December 2006) at NyM Meteorological Station (Figure
3). In addition, the fitted linear trend line is statistically
8) BIT Index significant at alpha level, α, of 0.05 (Kendall’s tau rτ (39) =
-.223, p =0.024) (Table 4).
BIT index = (8)
In this regard, positive and negative anomalies of
annual rainfall typified annual and inter-annual climate
variability. Additionally, increasing negative anomalies of
Where, VI = brGDGT VI, VII = brGDGT VII, VIII = brGDGT VIII, annual rainfall decreased mean annual rainfall, which led
 Bagambilana and Rugumamu 57

Table 1. Decadal changes in mean annual rainfall at Same Meteorological station, 1970 to 2009 and 2012 to 2015.

Parameter 1970 - 2015 1970 - 1979 1980 - 1989 1990 - 1999 2000 - 2009 2012 - 2015
M 540.5 631.1 559.3 528.5 500.7 395.9
SD 202.1 239.4 122.9 228.4 219.0 73.2
Coefficient of variation 0.4 0.4 0.2 0.4 0.4 0.2
Range 808.7 754.4 391.6 675.5 753.9 169.4
Minimum 265.3 319.6 376.5 299.7 265.3 293.3
Maximum 1074.0 1074.0 768.1 975.2 1019.2 462.7
Source of data: TMA and PBWB.

Table 2. Decadal changes in mean annual rainfall at NyM Meteorological station, 1977 – 2015.

Parameter 1977 - 2015 1977 - 1979 1980 - 1989 1990 - 1999 2000 - 2009 2010 - 2015
M 342.9 611.6 377.6 332.6 226.6 361.3
SD 157.5 170.7 112.5 145.9 142.1 77.1
Coefficient of variation 0.5 0.3 0.3 0.4 0.6 0.2
Range 731.1 338.2 400.8 547.8 487.5 215.4
Minimum 63.2 456.1 198.1 84.3 63.2 272.5
Maximum 794.3 794.3 598.9 632.1 550.7 487.9
Source of data: TMA and PBWB.

Figure 2. Annual rainfall anomalies for Same Meteorological Station, 1970– 2009 and 2012 – 2015.
Source of data: TMA and PBWB.

Table 3. Mann-Kendall’s test for trend of rainfall at Same Met Station, 1970– 2009 and 2012 – 2015.

a b
Value Asymp. Std. Error Approx. T Approx. Sig.
Kendall's tau-b -0.230 0.098 -2.349 0.019
Ordinal by Ordinal
Kendall's tau-c -0.230 0.098 -2.349 0.019
N of Valid Cases 44
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming the null hypothesis.
Source of data: TMA and PBWB.
58 Afr. J. Environ. Sci. Technol.

Figure 3. Annual rainfall anomalies for NyM Meteorological Station, 1977– 2015.
Source of data: TMA and PBWB.

Table 4. Mann-Kendall’s test for trend of rainfall at NyM Met Station, 1977 – 2015.

a b
Parameter Value Asymp. Std. Error Approx. T Approx. Sig.
Kendall's tau-b -0.223 0.098 -2.265 0.024
Ordinal by Ordinal
Kendall's tau-c -0.223 0.098 -2.265 0.024
N of Valid Cases 39
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming the null hypothesis.
Source of data: TMA and PBWB.

to climate change. Indeed, rainfall decreased faster at forests in the highlands of Mount Kilimanjaro).
-1
NyM Meteorological Station (slope -4.900 mm y and Specifically, cultivated areas increased from 54% in
change is explained by 12.5% of the observed variance) 1973, to 62 and 63% in 1984 and 2000, respectively. The
-1
than at Same Meteorological Station (slope -4.186 mm y main underlying drivers for such land-use/land cover
and change is explained by 7% of the observed variance) changes were demographic, government policies, socio-
(Figures 2 and 3). economic/cultural factors encompassing land-tenure
For comparison purposes, the findings did not reveal, system, institutional factors, technological change and
during a period of 62 years (1938 – 1999), increasing infrastructure development.
negative anomalies of annual rainfall from a reference Besides, a study conducted by Huang et al. (2016)
value of 905 mm (an average for a period of 30 years revealed that precipitation had decreased by 57 mm in
from 1967 to 1996) at Moshi Airport Meteorological the newly formed semi-arid regions of East Asia during a
Station (Figure 4). Additionally, the fitted linear trend line 61-year period from 1948 to 2008. Additionally, potential
is non-statistically significant at alpha level, α, of 0.05 evapotranspiration had increased by 132 mm in the same
(Kendall’s tau rτ (62) =.021, p = .810) (Table 5). regions and same timescale. A decrease of precipitation
It is worth pointing out that climate change is partly and concomitant increase of potential evapotranspiration
caused by land-use/land cover changes (UN, 1992). led to a considerable decrease of aridity index, computed
Furthermore, statistically significant decrease of annual as the ratio of precipitation to potential evaporation,
rainfall at both Same and NyM Meteorological stations implying an increase of drying trends in the newly formed
could partly be explained by land-use/land cover semi-arid regions of East Asia. The consistency of
changes. In this regard, a study conducted by Misana et precipitation patterns with aridity patterns implied
al. (2012) in Kilimanjaro Region (also encompassing precipitation was a key factor influencing aridity index. It
semi-arid areas in Mwanga District) revealed, through is worth pointing out that precipitation had decreased at
analyses of satellite images dated 1973, 1984 and lesser rate and potential evapotranspiration had
1999/2000, expansion of cultivated areas in the southern increased at lesser rate in the old semi-arid regions of
and eastern lowlands of Mount Kilimanjaro. This was at East Asia.
the expense of natural vegetation, particularly natural Likewise, a study conducted by Mabhuye et al. (2015)
 Bagambilana and Rugumamu 59

Figure 4. Annual rainfall anomalies for Moshi Airport Meteorological Station, 1938 – 1999.
Source of data: Røhr (2003).

Table 5. Mann-Kendall’s test for trend of rainfall at Moshi Airport Met Station, 1938 to 1999.

a b
Parameter Value Asymp. Std. Error Approx. T Approx. Sig.
Kendall's tau-b 0.021 0.086 0.241 0.810
Ordinal by Ordinal
Kendall's tau-c 0.021 0.086 0.241 0.810
N of Valid Cases 62
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming the null hypothesis.
Source of data: Røhr (2003).

revealed negative sloping trends for rainfall in most of eastern Africa suggested, for instance, a wetter climate
arid and semi-arid lands of Tanzania during a period from with more intense MAM and OND rainy seasons and less
st
1975 to 2004. The same study revealed, however, non- severe droughts by the end of the 21 century. During the
significant positive linear trends for annual, MAM (March, same time, however, regional climate models indicated
April, May), and OND (October, November, December) that most parts of Kenya, South Sudan, and Uganda
rainfall for Dodoma during a period from 1960 to 2007. would experience drier conditions in August and
Besides, a study conducted by IPCC (2013) revealed a September due to the weakening of Somali jet and Indian
high degree of both temporal and spatial variability of Ocean monsoon. Additionally, it was projected that boreal
precipitation in eastern Africa. Specifically, MAM rainfall spring rains would be truncated over eastern Ethiopia,
had decreased in the region during the previous three Somalia, Tanzania, and southern Kenya while boreal fall
decades probably due to rapid warming of the Indian seasons would be lengthened in southern Kenya and
st
Ocean that led to convection and precipitation over the Tanzania in the mid-21 century.
tropical Indian Ocean and subsidence of drier air over
eastern Africa. Additionally, an increasing frequency of
dry spells tended to be accompanied with an increasing Nexuses between rainfall changes in the study area
trend in daily rainfall intensity. Consequently, higher and El niño–southern oscillation and Indian Ocean
rainfall intensity tended to increase both soil erosion and dipole
sediment loads in waterways.
It should be noted, however, that projection of An evaluation of the linear relationships between ENSO’s
precipitation was much more uncertain than projection of Niño 3.4 (five month running mean of extended
temperature (UNEP, 2012). According to IPCC (2013), an reconstructed sea surface temperature version 5
assessment of 12 Coupled Model Intercomparison (ERSST.v5) anomalies in the Niño 3.4 region from
Project Phase 3 GCMs (general circulation models) over September – February) (Figure 5) and amounts of rainfall
60 Afr. J. Environ. Sci. Technol.

Figure 5. ENSO’s Niño 3.4, 1970/71 to 2015/2016.

Source of data: NOAA.

Table 6. Correlation matrix for ENSO’s Niño 3.4 and rainfall at

Same and NyM meteorological Stations.

1 2 3
1 ENSO’s Niño 3.4 (Sep - Feb)
2 Sep – Feb rainfall at Same Station 0.54**
3 Sep – Feb rainfall at NyM station 0.45** 0.85**
**Correlation is significant at the 0.01 level (2-tailed).

for corresponding periods at Same and NyM Meteorological Station from September to February
Meteorological stations was made through hypothesis during strong El Niño events of 1972/73, 1982/83, and
testing for statistical significance of Pearson’s correlation 1997/98. This was when corresponding Niño 3.4 values
coefficients. were 1.0, 1.3, and 1.4 respectively while low amount of
rain, 108.5 mm, was recorded (during the same months)
H0: ρ = 0 (there was no linear relationship between at the same station during a strong La Niña event of
ENSO’s Niño 3.4 and amounts of rainfall during periods 1974/75 when the corresponding Niño 3.4 value was -
from September to February). 1.1. Also, high amounts of rain, 421 mm and 598, were
Ha: ρ ≠ 0 (there was a linear relationship between recorded (during similar months) at NyM Meteorological
ENSO’s Niño 3.4 and amounts of rainfall during periods Station during strong El Niño events of 1982/83 and
from September to February). 1997/98, when corresponding Niño 3.4 values were 1.3
Pearson’s correlation coefficient indicated that there and 1.4 respectively. However low amounts of rain, 44
was a significant positive linear relationship between and 64 mm, were recorded (during the same months) at
ENSO’s Niño 3.4 and amounts of rainfall (September – the same station during strong La Niña events of 1999/00
February) at Same Meteorological Station, r(41) = .54, p and 2000/01 when corresponding Niño 3.4 values were -
< .001 (Table 6), which led to rejection of the null 1.3 and -1.1, respectively.
hypothesis. Similarly, there was a significant positive Similarly, an evaluation of the linear relationships
linear relationship between ENSO’s Niño 3.4 and between IOD’s DMI (overall anomalies of DMI averaged
amounts of rainfall (September – February) at NyM for periods from October to December) (Figure 6) and the
Meteorological Station, r(37) = .45, p = .004, which led to amounts of OND rainfall at Same and NyM Meteorological
rejection of the null hypothesis. Besides, the amounts of stations was made through hypothesis testing for
rainfall at both stations were highly correlated during a statistical significance of a Pearson’s correlation
period from September to February, r(34) = .85, p < .001. coefficient.
It is worth pointing out that high amounts of rain, 401
mm, 540 mm, and 920 mm, were recorded at Same H0: ρ = 0 (there was no linear relationship between IOD’s
 Bagambilana and Rugumamu 61

Figure 6. IOD’s DMI, 1970 to 2015.

Source of data: NOAA.

Table 7. Correlation matrix for IOD’s DMI and rainfall at

same and NyM meteorological Stations.

1 2 3
1 IOD’s DMI (OND)
2 OND rainfall at Same station 0.57**
3 OND rainfall at NyM station 0.42** 0.71**
**Correlation is significant at the 0.01 level (2-tailed).

DMI and amounts of OND rainfall). and 0.7 respectively. It should be noted that lower
Ha: ρ ≠ 0 (there was a linear relationship between IOD’s amounts of rain, 0, 15.7, 0, 30, and 38 mm, were
DMI and amounts of OND rainfall). recorded (during the same months) at the same station in
1998, 1999, 2000, 2003, and 2005, respectively, when
An analysis using Pearson’s correlation coefficient DMI values ranged between 0 and 0.2.
indicated that there was a significant positive linear Similarly, Ogwang et al. (2015:7) explained the
relationship between IOD’s DMI and amounts of OND influence of IOD on East African OND rainfall, hence
rainfall at Same Meteorological Station, r(42) = .57, p < climate variability in the region.
.001 (Table 7), which led to rejection of the null ‘Positive IOD (Negative IOD) event results into flood
hypothesis. Similarly, there was a significant positive (drought) in the region. The evolution of these
linear relationship between IOD’s DMI and amounts of phenomena can thus be keenly observed and applied in
OND rainfall at NyM Meteorological Station, r(37) = .42, p seasonal forecasting to avert the huge losses associated
= .008, which led to rejection of the null hypothesis with extreme weather events. During wet years, the wind
Besides, the amounts of rainfall at both stations were circulation is observed to converge in the low levels over
highly correlated during a period from October to the western Indian Ocean and diverge in the upper
December, r(35) = .71, p < .001. levels, the opposite is observed in dry years.
It is worth pointing out, for example, that high amounts Indeed, prolonged periods of drought from 1973 to
of rain, 400 and 596 mm, were recorded at Same 1976 for Same Meteorological station (Figure 2) and from
Meteorological Station from October to December during 1998 to 2005 for NyM Meteorological station (Figure 3)
positive IOD phases in 1982 and 1997 when were largely linked to climate variability. With regard to
corresponding IOD’s DMI values were 0.5 and 0.7, the former, Kavishe and Mushi (1993) and Limbu (1995)
respectively. Furthermore, lower amounts of rain, 29 and explained that a severe drought of 1973 – 1975 led to
18 mm, were recorded at the same station during crop failure in several parts of Tanzania. Indeed, the
negative IOD phases in 1975 and 1996 when drought coupled with Government’s implementation of
corresponding IOD’s DMI values were -0.1 and -0.2, villagization program that displaced farmers and the
respectively. Besides, high amounts of rain, 374 mm and world’s oil crisis led to severe famine, which increased
272, were recorded (during similar months) at NyM importation of cereals from 12,000 tons in 1972 to
Meteorological Station during positive IOD phases in 521,000 tons in 1975. Consequently, the drought
1982 and 1997 when corresponding DMI values were 0.5 prompted the Government Tanzania to launch several
62 Afr. J. Environ. Sci. Technol.

campaigns to ensure food security, including kilimo cha 650 and 950 CE, and between 1550 and 1700 CE
kufa na kupona (‘Agriculture as a matter of life and (Figure 7). To the contrary, lowest BIT values (< 0.5),
death). implying driest periods, occurred between 1170 and 1300
With regard to the latter, a study conducted by Kijazi CE, and between 1780 and 1820 CE. In addition, low BIT
and Reason (2008) established out-of-season peak in values occurred between 1870 and 1895 CE and
January (for years of 1998, 2000, 2001, 2003 and 2004, between 1968 and 1974 CE.
hence temporary displacement of short (Vuli) rainy In this regard, a study conducted by Scroxton et al.,
seasons from OND to NDJ (November, December, (2017) reported similar wet phases between the 6th and
January) in the lowlands of northeastern highlands. With 10th centuries and between the 16th and 18th centuries
18
respect to the 2004 MAM rainy season, for instance, the for: i) Anjohibe speleothem, Madagascar based on δ O
18
onset was on the 14th pentad (period of five days) (6th to record, ii) Cave Defore speleothem, Oman based on δ O
10th March) and the cessation was on the 24th pentad record, iii) Lake Edward, Uganda/Democratic Republic of
(26th to 30th April), hence a shorter rainy season. Congo based on Mg/Ca in Calcite, iv) Lake Naivasha,
Additionally, total rainfall (169 mm) for the 2004 MAM Kenya, based on Lake level (m), and v) Liang Luar Cave
rainy season was less than half of the normal average of speleothem, Flores, Indonesia based on PC1.
388 mm. Also, the 2004 MAM rainy season was Specifically, Verschuren et al. (2000) observed an
characterized with unusual dry spells from 16th to 20th agreement between (multi-) decadal trends in Lake
March, 6th to 10th April, and 16th to 20th March. Challa’s varve thickness record and the 1100-year
According to Kijazi and Reason (2008), the multi-year moisture-balance reconstruction from Lake Naivasha,
droughts from 1998 to 2005 were associated with three Central Kenya.
atmospheric circulation patterns. Firstly, wind circulation Moreover, such state of affairs was attributed to both
patterns diverted atmospheric moisture away from zonal Walker Cell dynamics controlled by the Indian
Tanzania. Secondly, there was strong moisture flux Ocean and meridional shifts in the average position of
divergence in the northeastern highlands. Thirdly, there ITCZ that influenced the intensity of northeasterly and
was strong subsidence, which was linked to the eastward southeasterly monsoons/rains. Indeed, monsoon rainfall,
displacement of ascending limb of the Walker-type controlled by insolation forcing and ENSO dynamics,
circulation. varied at half-processional (11,500 years) interval
Moreover, both MAM and OND rainy seasons in the (Verschuren et al., 2009; Wolf et al., 2011; Buckles,
study area were largely driven by the yearly double 2016).
passage of ITCZ between the Equator and Tropic of Furthermore, the driest period that occurred between
Capricorn (Owiti et al., 2008; Williams and Hanan, 2011; 1110 and 1310 CE was partly associated with a period of
Schmidt and Spero, 2011). In turn, the migration of ITCZ high volcanic activity (the Medieval Volcanic Activity)
was largely driven by the intensification and relaxation of between 1250 and 1500 CE when global temperatures
migrating Azores and Siberian anticyclones in the cooled dramatically as explained by the theory of volcanic
northern hemisphere and Mascarene and St. Helena activity (IPCC, 2013). In addition, a study conducted by
anticyclones in the southern hemisphere. Scroxton et al.,(2017) reported similar drier periods
During January when the overhead Sun was between the 11th and 15th centuries and in the 20th
experienced near the southern hemisphere, for instance, century for five hydroclimate stations as mentioned
the Mascarene and St. Helena anticyclones relatively above. Likewise, a study conducted by Verschuren et al.
relaxed in the southern hemisphere while the Azores and (2000) suggested the existence of drier conditions in
Siberian anticyclones intensified in the northern Equatorial East Africa during the Mediaeval Warm Period
hemisphere. Consequently, the rain-bearing ITCZ was (1000 to 1270 CE) and the relatively wetter conditions
largely confined in the southern hemisphere. Conversely, during the Little Ice Age (1270 – 1850) that were
during July when the overhead Sun was experienced interrupted by severe drought episodes (1380 to 1420
near the northern hemisphere, the Mascarene and St. CE, 1560 to 1620 CE, and 1760 to 1840 CE). Generally,
Helena anticyclones relatively intensified in the southern such findings suggested that both zonal and meridional
hemisphere while the Azores and Siberian anticyclones mechanisms controlled tropical precipitation variability
relaxed in the northern hemisphere. Thus, the rain- during the late Holocene. It should be noted, however,
bearing ITCZ was largely confined in the northern that the relatively drier and windier conditions existed in
hemisphere. Lake Malawi, within the African Rift Valley, partly due to
the weakening of the Congo Basin Monsoon during the
Little Ice Age (Scroxton et al., 2017).
Paleo-precipitation changes in the study area Pertaining to the BIT index covering a period of the last
25 ka, lowest BIT values (< 0.6), implying driest periods,
To begin with BIT index covering a period of the last 2.2 occurred during Heinrich 2 stadial, the late Last Glacial
ka, highest BIT values (> 0.8), implying wettest periods Maximum and early Late-glacial period (including
that were accompanied with floods, occurred between Heinrich 1 stadial) (20.4 - 15.9 ka BP) and during the
 Bagambilana and Rugumamu 63

Figure 7. BIT index for Lake Challa area during the last 2.2 ka. Orange-shaded bars approximate periods of
documented drought episodes in East Africa: 1 (1780 – 1820 CE), 2 (1870 – 1895 CE), and 3 (1968 – 1974 CE).
Source of data: NOAA and Buckles et al. (2016).

Figure 8. BIT index for Lake Challa area during the last 25 ka.
Source of data: Sinninghe Damsté et al. (2012).

Younger Dryas (13.1 - 11.7 ka BP) (Figure 8). In addition, area (based on BIT), Greenland composite (NGRIP-
drier conditions occurred during the mid Holocene (6 ka GISP2-GRIP) based on atmospheric CH4 record,
18
BP) and late-Holocene between 0.8 and 0.6 ka BP Greenland (NGRIP) based on ice core δ O with respect
(ca.1100 - 1300 CE) and between 0.2 and 0.15 ka BP to GICC05 timescale, and Hulu/Dongge cave based on
18
(ca.1750-1810 CE). Excluding Heinrich 2 stadial, wetter stalagmite δ O (Verschuren et al., 2009). Additionally,
conditions occurred from 25 to 22 ka BP. Additionally, there was agreement of a period with wetter conditions
wetter conditions occurred from 14.5 to 8.5 ka BP between the Younger Dryas and Heinrich 1 stadial
(excluding the Younger Dryas) and since 4.5 ka BP to between the four hydroclimate stations. Such findings
present. suggest that precipitation patterns for Lake Challa area
Indeed, paleo-precipitation reconstruction (Figure 8) during a period of the last 25 ka was largely influenced by
was in agreement with the reconstruction of Lake teleconnections of natural atmosphere-ocean modes of
Challa’s surface levels, encompassing high stands and climate variability including IOD and ENSO and
low stands, based on a high-resolution seismic-reflection meridional shifts in the average position of ITCZ.
stratigraphy survey (Moernaut et al., 2010; Buckles et al., Moreover, a study conducted by Wolff et al. (2011)
2016). revealed that thin varves from Lake Challa highly
Besides, there was agreement of periods with drier correlated with El Niño events (positive Niño3.4 values)
conditions during the Younger Dryas (13.1- 11.7 ka BP) during a period of the previous 155 years while thick
and Heinrich 1 stadial (15 – 18 ka BP) for Lake Challa varves highly correlated with La Niña events (negative
64 Afr. J. Environ. Sci. Technol.

Niño3.4 values). Indeed, La Niña events were associated (Adansonia digitata). Pertaining to MEKS, studies could
with existence of strong winds that enhanced upwelling of focus on ways of enhancing reliability of modern weather
nutrients and intensification of seasonal blooms of algae forecasting and timely dissemination of weather
and, consequently, the formation of thicker varves. To the advisories to stakeholders.
contrary, El Niño events were associated with existence
of weak winds that led to the formation of thinner varves.
Basing on analysis of varve structure, El Niño events with CONFLICT OF INTERESTS
high rainfall over Lake Challa area occurred in years of
1997/98, 1982/83, 1941/42, 1914/15, 1905/06 and La The authors have not declared any conflict of interests.
Niña events with low rainfall occurred in 1988/89,
1971/72, 1955/56, 1950/51, 1924/25, 1917/18, and
1897/98. REFERENCES

Bast JL (2010). Seven theories of climate change. Chicago, IL:

Heartland Institute.
CONCLUSIONS AND RECOMMENDATIONS
Brown DA (2013). Climate change: Navigating the perfect moral storm.
London: Routledge.
This paper sought to assess linkages between Buckles LK, Verschuren D, Weijers JWH, Cocquyt C, Blaauw M,
precipitation changes in semi-arid lowlands of Mwanga Sinninghe Damsté JS (2016). Interannual and (multi-) decadal
variability in the sedimentary BIT index of Lake Challa, East Africa,
District and both human-induced climate change and over the past 2200 years: Assessment of the precipitation proxy.
natural modes of climate variability across different Climate of the Past 12:1243-1262.
temporal scales. Since the findings revealed statistically Cassou C, Kushnir Y, Hawkins E, Pirani A, Kucharski F, Kang I–S,
significant decrease of annual rainfall at both Same and Caltabiano N (2018). Decadal climate variability and predictability:
Challenges and opportunities. Bulletin of the American Meteorological
NyM Meteorological stations, it was concluded that the
Society 99(3):479-490.
human-induced climate change had occurred in the study FAO, IFAD, UNICEF, WFP, WHO (2018). The state of food security and
area between the 1970s and 2010s. Besides, the findings nutrition in the world 2018. Building climate resilience for food
revealed, during the same time and stations, statistically security and nutrition. Rome, FAO.
http://www.fao.org/3/I9553EN/i9553en.pdf
significant correlations between amounts of rainfall Field A (2013). Discovering statistics using IBM SPSS statistics. 4th Ed.
(September – February) and ENSO’s Niño 3.4 index and London: SAGE.
between amounts of OND rainfall and IOD’s DMI. In this Huang J, Ji M, Xie Y, Wang S, He Y, Ran J (2016). Global semi-arid
regard, it was concluded that precipitation changes climate change over last 60 years. Climate Dynamics 46(4):1131-
1150.
during short rainy seasons (Vulis) in the lowlands were
Idso CD, Carter, RM, Singer SF (2016). Why scientists disagree about
significantly influenced by cycles of climate variability global warming: The NIPCC report on scientific consensus. 2nd Ed.
(namely, ENSO and IOD). Chicago, IL: Heartland Institute.
Indeed, rainfall fluctuations during the last 25 ka were International Federation of Red Cross and Red Crescent Societies
(IFRCRCS) (2013). Tanzania drought and food insecurity:
also influenced and shaped by meridional shifts in the
Emergency appeal n° MDRTZ012 - final report. International
average position of ITCZ and zonal Hadley/Walker Cell Federation of Red Cross and Red Crescent Societies.
dynamics in the Indian Ocean due to changes in Inhofe J (2012). The greatest hoax: How the global warming conspiracy
insolation forces. Furthermore, the findings revealed threatens your future. Washington, D. C: WND Books.
Intergovernmental Panel on Climate Change (IPCC) (2012). Managing
higher average coefficient of variation for rainfall at NyM
the risks of extreme events and disasters to advance climate change
Meteorological Station and lower average coefficient of adaptation. Cambridge: Cambridge University Press.
variation for rainfall at Same Meteorological Station. In Intergovernmental Panel on Climate Change (IPCC) (2013). Climate
this respect, it was concluded that rainfall variability at change 2013: The physical science basis. Contribution of Working
Group I to the Fifth Assessment Report of the IPCC. Cambridge:
Same Meteorological Station was lower as influenced by, Cambridge University Press.
among other things, topographical factors and Kavishe FP, Mushi SS (1993). Nutrition-relevant actions in Tanzania.
continentality. TFNC Monograph Series no. 1. UN ACC/SCN Country Study
Since increasing rainfall perturbations pose significant Supported by UNICEF. A Case Study for the XV Congress of the
International Union of Nutritional Sciences, September 26 to October
risks on economic activities, including rain-fed agricultural
1, 1993, Adelaide.
production systems, that provide livelihoods to the Kijazi AL, Reason CJC (2008). Analysis of the 1998 to 2005 drought
majority of residents in the study area, there is a need to over the northeastern highlands of Tanzania. Climate Research
undertake studies aimed at enhancing synergies between 38:209-223.
Laliberté F, Zika, J, Mudryk L, Kushner PJ, Kjellsson J, Döös K (2015).
traditional environmental knowledge systems (TEKS) and
Constrained work output of the moist atmospheric heat engine in a
modern-day environmental knowledge systems (MEKS) warming climate. Science 347(6221):540-543.
as early warning systems. With regard to TEKS, studies Limbu F (1995). Agriculture and rural development in Tanzania: A
could focus on traditional weather prediction through survey of the 1980-1985 literature. Dar es Salaam: Economic and
Social Research Foundation.
observation of changing appearance and positions of Mabhuye E, Yanda P, Maganga F, Liwenga E, Kateka A, Henku A,
astronomical objects, such as a moon/stars (Ngalakeri in Malik N, Bavo C (2015). Tanzania: country situation assessment.
Pare language) and fruiting/ripening of fruits of wild PRISE working paper.
plants, such as Mikayo (Salvadora persica) and Baobab Mashingo M (2010). Vulnerability assessment on pastoral livestock and
 Bagambilana and Rugumamu 65

pasture development to the extreme climatic events in Tanzania. Sinninghe Damsté JS, Ossebaar J, Schouten S, Verschuren D (2012).
Paper presented under the study on economics of climate change in Distribution of tetraether lipids in the 25-ka sedimentary record of
Tanzania Phase 3, at the Stakeholder Inception Round Table Lake Challa: Extracting reliable TEX86 and MBT/CBT
Meeting held at the Peacock Hotel, Dar es Salaam, Tanzania, 26 palaeotemperatures from an equatorial African lake. Quaternary
March 2010. Science Reviews 50:43-54.
Mwanga District Council (MDC) (2016). Mwanga district socio-economic Stewart I (2008). Earth: The climate wars. London: BBC.
profile. Mwanga: MDC. https://pefa.org/assessments/tanzania- Tao T, Chaocat B, Liu S, Xin K (2009). Uncertainty analysis of
mwanga-district-council-2016 interpolation methods in rainfall spatial distribution: A case of small
Misana B, Sokoni C, Mbonile MJ (2012). Land-use/cover changes and catchment in Lyon. Journal of Environmental Protection 1:50-58.
their drivers on the slopes of Mount Kilimanjaro, Tanzania. Journal of United Nations (UN) (1992). United Nations framework convention on
Geography and Regional Planning 5(6):151-164. climate change. New York, NY: United Nations.
Moernaut J, Verschuren D, Charlet F, Kristen I, Fagot M, De Batist M United Nations Environment Programme (2012). Global environment
(2010). The seismic-stratigraphic record of lake-level fluctuations in outlook GEO-5: Environment for development. Nairobi: UNEP.
Lake Challa: Hydrological stability and change in equatorial East United Republic of Tanzania (URT) (2004). Lake Jipe awareness raising
Africa over the last 140 kyr. Earth and Planetary Science Letters strategy (2005–2007). Dar es Salaam: Ministry of Natural Resources
290:214-223. and Tourism, Wildlife Division.
Mvungi A (2008). The economy of affection/moral economy and water Verschuren D, Laird KR, Cumming BF (2000). Rainfall and drought in
management for irrigation and environmental conservation in semi- equatorial East Africa during the past 1100 years. Nature 403:410-
arid areas: The case study of Mwanga district. In: Kimambo IN, 414.
Hyden G, Maghimbi S, Sugimura K (eds.). Contemporary Verschuren D, Sinninghe Damsté JS, Moernaut J, Kristen I, Blaauw M,
perspectives on African moral economy. Dar es Salaam: Dar es Fagot… Nowaczyk NR (2009). Half-precessional dynamics of
Salaam University Press. monsoon rainfall near the East African Equator. Nature
Ogwang BA, Ongoma V, Xing L, Ogou FK (2015). Influence of 462(7273):637-641.
Mascarene high and Indian Ocean Dipole on East African extreme Williams CA, Hanan NP (2011). ENSO and IOD teleconnections for
weather events. Geographica Pannonica 19 (2):64-72. African ecosystems: Evidence of destructive interference between
Ostberg S, Schewe J, Childers K, Frieler K (2018). Changes in crop climate oscillations. Biogeosciences 8:27-40.
yields and their variability at different levels of global warming. Earth Wolff C, Haug GH, Timmermann A, Sinninghe Damsté JS, Brauer A,
System Dynamics 9:479-496. Sigman DM, Cane MA, Verschuren D (2011). Reduced inter-annual
Owiti Z, Ogallo LA, Mutemi J (2008). Linkages between the Indian rainfall variability in East Africa during the last ice age. Science
Ocean Dipole and East African rainfall anomalies. Journal of Kenya 333:743-747.
Meteorological Society 2(1): 3-17. Yeh S-W, Cai W, Min S-K, McPhaden MJ, Dommenget D, Dewitte B,
Pokhrel S, Chaudhari HS, Saha SK., Dhakate A, Yadav RK, Salunke K, Kug J-S (2018). ENSO atmospheric teleconnections and their
Mahapatra S (2012). ENSO, IOD and Indian Summer Monsoon in response to greenhouse gas forcing. Reviews of Geophysics
NCEP climate forecast system. Climate dynamics 39:2143-2165. 56(1):185-206.
Røhr PC (2003). A hydrological study concerning the southern slopes of Zheng J, Yu Y, Zhang X, Hao Z (2018). Variation of extreme drought
Mt Kilimanjaro, Tanzania. Dissertation (PhD), Norwegian University and flood in North China revealed by document-based seasonal
of Science and Technology. precipitation reconstruction for the past 300 years. Climate of the
Scroxton N, Burns SJ, McGee D, Hardt B, Godfrey LR, Past 14:1135-1145.
Ranivoharimanana L, Faina P (2017). Hemispherically in-phase
precipitation variability over the last 1700 years in a Madagascar
speleothem record. Quaternary Science Reviews 164:25-36.
Schmidt MW, Spero HJ (2011). Meridional shifts in the marine ITCZ and
the tropical hydrologic cycle over the last three glacial cycles.
Paleoceanography 26(1):PA1206.