论文信息

Ahmed Abdiaziz Alasow / Abdirahman Liban Mohamed / Kowsar Abdirahman Afyare 等 8 人

Frontiers in Sustainable Food Systems

论文解读

AI 辅助整理
方法
NARDL框架、FAO生产指数与CRU气候数据、累积气候压力指标及内生结构断点检
发现
短期温度冲击呈非对称性,负向冲击危害更大;作物比畜牧更敏感且恢复更慢。
机会点
可拓展至其他脆弱地区,量化非对称气候冲击并识别作物与畜牧系统差异化的适应阈值。

论文摘要

气候变率对脆弱且暴露于气候风险地区的农业系统构成日益加剧的风险,然而关于非对称性和动态生产响应的证据仍然有限。本研究利用1989—2021年的年度数据,考察气候变率对索马里种植业和畜牧业生产的短期与长期影响。研究将联合国粮食及农业组织(FAO)的作物和畜牧生产指数与气候研究单位(CRU)的降雨和温度数据相结合,采用非线性自回归分布滞后(NARDL)框架检验对正向和负向气候冲击的非对称响应,并辅以累积气候胁迫指标和内生结构断点控制。结构不稳定性通过内生断点检验加以处理。结果表明,温度对两类农业系统的影响在短期内存在统计显著的非对称性。负向温度冲击产生的短期不利影响比正向温度变化更大且更持久,其系数在作物和畜牧生产中均在常规水平上显著。相比之下,长期系数总体上统计不显著且估计不精确,仅提供提示性而非结论性证据表明气候影响具有持续性。误差修正估计为畜牧业生产的收敛提供了有力证据,每年约修正45%的非均衡(p = 0.008),但对种植业生产仅提供边际证据,每年约修正30%(p = 0.097)。基线线性模型忽略了短期非线性响应,凸显了采用非对称框架的必要性。气候胁迫指数显示,近几十年来高胁迫状态的频率和持续时间均有所增加。这些发现表明,气候对索马里农业的短期影响具有非对称性和系统特异性:种植业生产比畜牧业更敏感且恢复更慢。这凸显了制定气候响应型政策的必要性,包括投资耐热作物品种、改善水资源管理以及对脆弱环境中的牧区系统提供适应性支持。

全文

Frontiers in Sustainable Food Systems

Frontiers in Sustainable Food Systems 01 frontiersin.org
TYPE Original Research
PUBLISHED 17 November 2026
DOI 10.3389/fsufs.2026.1828365
OPEN ACCESS
EDITED BY
Oyegunle Emmanuel Oke,
Federal University of Agriculture, Nigeria
REVIEWED BY
A. Amarender Reddy,
National Institute of Agricultural
Extension Management (MANAGE), India
Zubair Tanveer,
National Tariff Commission, Pakistan
CORRESPONDENCE
Ahmed Abdiaziz Alasow
alasow@hu.edu.so
RECEIVED 11 March 2026
ACCEPTED 21 October 2026
PUBLISHED 17 November 2026
CITATION
Alasow AA, Mohamed AL, Afyare KA,
Ali II, Hersi AA, Osman AM, Adan MD and
Abdi SM (2026) Asymmetric and
dynamic effects of climate variability on
crop and livestock production in
Somalia.
Front. Sustain. Food Syst. 10:1828365.
doi: 10.3389/fsufs.2026.1828365
COPYRIGHT
© 2026 Alasow, Mohamed, Afyare, Ali,
Hersi, Osman, Adan and Abdi. This is an
open-access article distributed under
the terms of the Creative Commons
Attribution License (CC BY). The use,
distribution or reproduction in other
forums is permitted, provided the
original author(s) and the copyright
owner(s) are credited and that the
original publication in this journal is
cited, in accordance with accepted
academic practice. No use, distribution
or reproduction is permitted which does
not comply with these terms.
Asymmetric and dynamic effects
of climate variability on crop and
livestock production in Somalia
Ahmed Abdiaziz Alasow 1
, Abdirahman Liban Mohamed 2,
Kowsar Abdirahman Afyare 1, Ibrahim Isse Ali 3,
Abdifatah Ahmed Hersi 4, Abdirahman Mohamed Osman 5,
Mohamed Dahir Adan 1 and Said Mohamed Abdi 1
1Faculty of Postgraduate Studies, Hormuud University, Mogadishu, Somalia, 2Department of
Environmental Management, Faculty of Forestry and Environment, Universiti Putra Malaysia, Serdang,
Selangor, Malaysia, 3Department of Plant Protection, Faculty of Agriculture, Selcuk University, Konya,
Türkiye, 4Faculty of Economics and Management Sciences, Hormuud University, Mogadishu, Somalia,
5Earth Observation Centre, Institute of Climate Change, Universiti Kebangsaan Malaysia, Bangi,
Selangor, Malaysia
Climate variability poses escalating risks to agricultural systems in fragile and
climate-exposed regions, yet evidence on asymmetric and dynamic production
responses remains limited. This study investigates the short-run and long-run
effects of climate variability on crop and livestock production in Somalia using
annual data for 1989–2021. Crop and livestock production indices from the
Food and Agriculture Organization (FAO) are combined with rainfall and tem -
perature data from the Climatic Research Unit (CRU). A nonlinear autoregressive
distributed lag (NARDL) framework is applied to examine asymmetric responses
to positive and negative climate shocks, complemented by cumulative climate
stress indicators and endogenous structural break controls. Structural instability
is addressed using endogenous break tests. The results reveal statistically signifi-
cant short-run asymmetries in the effects of temperature on both agricultural
systems. Negative temperature shocks exert larger and more persistent adverse
short-run effects than positive temperature changes, with coefficients signifi -
cant at conventional levels for both crop and livestock production. In contrast,
the long-run coefficients are generally statistically insignificant and imprecisely
estimated, providing suggestive rather than conclusive evidence of persistent
climate effects. Error-correction estimates provide strong evidence of conver-
gence for livestock production, with approximately 45% of disequilibrium cor -
rected annually ( p = 0.008), but only marginal evidence for crop production,
for which approximately 30% is corrected annually ( p = 0.097). Baseline linear
models overlook short-run nonlinear responses, emphasizing the need for asym-
metric frameworks. The Climate Stress Index shows high-stress regimes have
increased in both frequency and duration in recent decades. These findings show
that short-run climate impacts on Somalia’s agriculture are asymmetric and sys-
tem-specific: crop production is more sensitive and recovers more slowly than
livestock. This underscores the need for climate-responsive policies, including
investment in heat-tolerant crop varieties, improved water management, and
adaptive support for pastoral systems in fragile environments.
KEYWORDS
agricultural production, asymmetric effects, climate variability, food security, heat
stress, NARDL, Somalia

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 02 frontiersin.org
1 Introduction
Climate variability and climate change are increasingly recognized
as major sources of risk for agricultural production systems, particu-
larly in arid and semi-arid regions where livelihoods depend heavily
on rainfall and temperature conditions ( Alasow et al., 2024 , 2025;
Omar et al., 2024). A large empirical literature has documented that
variability in precipitation and rising temperatures can significantly
affect agricultural output through impacts on crop growth, soil mois-
ture, and livestock health (Lobell et al., 2011). These impacts tend to
be most severe in low-income and fragile contexts, where limited
infrastructure, constrained adaptive capacity, and repeated exposure
to climatic shocks reduce the ability of agricultural systems to recover
following adverse events (FAO, 2021; FAO, 2018).
Somalia represents a particularly climate-exposed and structurally
vulnerable case. The country is characterized by high interannual rain-
fall variability, recurrent droughts, and rising temperatures, with agri-
culture and pastoralism accounting for a substantial share of
livelihoods and food supply (Food and Agriculture Organization of
the United Nations (FAO) and World Bank, 2018). Recent assessments
consistently show that drought frequency and severity have increased
over the past decades, contributing to crop failures, livestock losses,
and widespread food insecurity (Ahmed and Ali, 2024; Warsame et
al., 2021; Yusuf et al., 2024). Observational evidence indicates that
agricultural production in Somalia is strongly influenced by rainfall
shortfalls and prolonged heat stress, particularly in rain-fed cropping
systems and pastoral areas (Alasow et al., 2024; Omar et al., 2025).
Despite the clear exposure of Somalia’s agricultural sector to cli-
mate variability, the empirical literature remains limited in several
important respects. Existing studies focusing on Somalia and the
wider Horn of Africa have primarily examined climate trends, drought
monitoring, or food security outcomes, often using descriptive or
correlation-based approaches (Funk et al., 2018).
While these studies provide valuable insights into climate hazards
and humanitarian risk, they offer less evidence on the dynamic pro-
duction responses of agricultural systems to climate variability. Few
studies explicitly quantify how agricultural output adjusts over time
following climate shocks, or whether adverse shocks exert dispropor-
tionately larger effects than favorable conditions.
More broadly, much of the climate–agriculture literature relies on
linear or average relationships, implicitly assuming that positive and
negative climate deviations affect production symmetrically. However,
recent climate-economy research increasingly emphasizes that climate
impacts are often nonlinear and asymmetric, with negative shocks
such as droughts or heatwaves generating larger and more persistent
losses than the gains associated with positive anomalies (Burke et al.,
2015; Dell et al., 2014). In agricultural contexts, repeated droughts and
sustained heat stress can produce cumulative damage to crops, range-
lands, and livestock health, effects that are not adequately captured by
models based on annual averages alone (Lobell et al., 2013). Empirical
applications of asymmetric and dynamic modeling frameworks to
Somalia’s agricultural sector, however, remain scarce.
Another important limitation of existing work concerns the
treatment of heterogeneity across agricultural systems. Cropping
and livestock production differ fundamentally in their exposure,
buffering capacity, and adjustment mechanisms in response to
climate variability ( Mohamednur and Ali, 2025 ; Rojas-Downing
et al., 2017 ; Thornton et al., 2011 ). Y et many studies treat
agriculture as a single aggregate sector, masking potentially
important differences in climate sensitivity between crops and
livestock. This issue is especially relevant in Somalia, where pas -
toral and agro-pastoral livelihoods play a central role in food
security and household resilience. Moreover, prolonged conflict
and institutional disruption have created structural changes that
may influence agricultural production independently of climate
conditions ( Ali and Ahmed, 2024 ). Without accounting for these
shifts, empirical models may incorrectly attribute broader struc -
tural effects to climate variability. Despite their importance in
fragile-state contexts, studies of climate and agricultural produc -
tion in Somalia have rarely formally incorporated struc -
tural breaks.
To cover these gaps, this study examines the effects of climate vari-
ability on crop and livestock production in Somalia from 1989 to 2021.
The nonlinear autoregressive distributed lag (NARDL) approach is
employed to estimate the asymmetric short- and long-run effects of
positive and negative changes in rainfall and temperature while
accommodating variables with different orders of integration (Pesaran
et al., 2001; Shin et al., 2014). Dynamic multipliers are used to trace
the adjustment of agricultural production following climatic shocks.
The analysis also accounts for relevant non-climatic determinants and
addresses structural instability through endogenous break tests and
regime controls.
The cumulative nature of climatic pressure is captured through
accumulated rainfall deficits, persistent heat stress, and a context-
specific Climate Stress Index that identifies periods of compound hot–
dry stress in Somalia. Rather than replacing the individual climate
variables, the index complements the NARDL analysis by supporting
climate-regime classification and robustness assessment. Modeling
crop and livestock production separately further enables the study to
assess differences in climate sensitivity, adjustment, recovery, and the
reversibility of climate-related production losses.
Using a national dataset for Somalia, this study advances the cli-
mate–agriculture literature by jointly examining asymmetric climate
responses, cumulative climatic stress, structural instability, and differ-
ences between crop and livestock production. This integrated frame-
work provides new, system-specific evidence on agricultural
vulnerability and supports targeted adaptation strategies in fragile and
climate-vulnerable settings (World Bank, 2023).
2 Methodology
2.1 Data and study scope
This study examines the impacts of climate variability and stress
on agricultural production in Somalia using annual data for the period
1989–2021. Agricultural outcomes are represented by the Crop
Production Index and Livestock Production Index, obtained from the
Food and Agriculture Organization (FAO), which provide standard-
ized measures of production dynamics over time. Climate variables
include annual rainfall (mm) and annual mean temperature ( °C)
derived from the Climatic Research Unit Time Series (CRU TS) data-
set, a widely used and quality-controlled global climate product. The
analysis period is selected to ensure consistency across all variables
and to capture post-conflict dynamics following Somalia’s early-1990s

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 03 frontiersin.org
institutional collapse. All variables are converted to annual frequency
and aligned to ensure a balanced sample.
2.2 Construction of climate stress variables
To capture nonlinear and asymmetric climate effects, rainfall and
temperature series are decomposed following the partial sum decom-
position approach proposed by Shin et al. (2014). First differences of
rainfall and temperature are separated into positive and negative
changes, which are then cumulatively summed to obtain long-run
partial sums representing favorable and adverse climate movements,
respectively. This approach allows positive and negative climate shocks
to exert distinct dynamic effects on agricultural production. In addi-
tion, climate stress indicators are constructed to reflect cumulative
exposure rather than isolated annual shocks. Rainfall stress is defined
by the accumulation of negative rainfall anomalies relative to the
sample mean, capturing prolonged drought conditions. Heat stress is
measured by the accumulation of positive temperature anomalies,
reflecting persistent warming pressure. These measures are combined
to form a Climate Stress Index (CSI), which is subsequently used to
identify climate stress regimes.
2.3 Preliminary analysis and structural
stability
Descriptive statistics are reported to summarize the central ten-
dency and variability of agricultural and climate variables. Stationarity
properties are examined using the Augmented Dickey–Fuller (ADF)
and KPSS tests to determine the order of integration of each series
(Dickey and Fuller, 1979; Shin, 1992). Given Somalia’s history of insti-
tutional disruption, structural breaks are assessed using the Zivot–
Andrews test, which allows for an endogenously determined break in
both intercept and trend (Zivot and Andrews, 2002). The combination
of unit root and structural break tests informs the choice of economet-
ric framework and validates the use of models capable of handling
mixed integration orders and regime shifts.
2.4 Econometric framework
The empirical analysis employs the Nonlinear Autoregressive
Distributed Lag (NARDL) model within an error-correction frame-
work. The NARDL approach is particularly suitable for small samples
and allows regressors integrated of order I(0) and I(1) while explicitly
modeling asymmetric long-run and short-run relationships (Pesaran
et al., 2001; Shin et al., 2014).
Given the limited annual series, lag selection was guided by the
Akaike Information Criterion to ensure parsimony. The restricted lag
structure preserves degrees of freedom and mitigates overparameter-
ization. Although climate effects may follow threshold or inverted
U-shaped patterns (Tanveer and Kalim, 2025), this study emphasizes
directional asymmetry rather than level-dependent nonlinearity.
NARDL is therefore applied to assess whether increases and decreases
in rainfall and temperature yield distinct short- and long-run effects
and adjustment paths in Somalia’s crop and livestock production
systems.
For each agricultural system (crop and livestock), the NARDL
error-correction specification relates changes in production to lagged
levels of asymmetric rainfall and temperature components, short-run
changes in climate variables, a post-1991 structural dummy, and a
deterministic time trend. Long-run coefficients are derived from the
estimated error-correction representation, while short-run dynamics
capture immediate responses to climate shocks. A baseline linear
ARDL model is estimated as a benchmark to assess whether average
climate effects mask important nonlinearities. The comparison
between linear and nonlinear specifications provides formal justifica-
tion for the asymmetric modeling strategy.
2.5 Dynamic analysis and robustness
Dynamic adjustment paths are examined using dynamic mul -
tipliers, which trace the temporal response of agricultural produc -
tion to positive and negative climate shocks. These multipliers
provide an intuitive representation of persistence, recovery speed,
and asymmetry in climate impacts. Robustness is assessed through
alternative lag structures, exclusion of extreme drought years, and
sensitivity to sample window selection. Additional checks include
incorporating climate stress regimes based on the CSI and verifying
cointegration using the ARDL bounds testing procedure ( Pesaran
et al., 2001 ). Across all specifications, results remain qualitatively
stable, confirming that the findings are not driven by model choice
or outlier years.
The Climate Stress Index (CSI) captures the compound effects of
rainfall deficits and persistent heat, which may jointly intensify evapo-
rative demand and soil-moisture depletion. It combines standardized
rainfall-deficit and heat-stress components relative to Somalia’s his-
torical climate. Rather than replacing rainfall and temperature in the
NARDL models, the CSI complements them by identifying periods of
simultaneous hot–dry stress. It is therefore used to classify climate-
stress regimes and evaluate the robustness of the main results.
3 Results
Table 1 reports descriptive statistics for agricultural production
and climate variables. Crop production has a higher standard devia-
tion than livestock production, indicating greater dispersion around
the mean. Livestock production shows comparatively lower variabil-
ity across the sample period. Annual rainfall displays a wide range
TABLE 1 Descriptive statistics (Somalia, 1989–2021).
Variable N Mean Std.
Dev.
Min Max
Crop
Production
Index (FAO)
33 84.339 16.892 64.221 122.630
Livestock
Production
Index (FAO)
33 79.961 18.099 59.729 110.804
Annual
rainfall (CRU,
mm)
33 279.882 29.522 230.230 348.660
Annual mean
temperature
(CRU, °C)
33 26.933 0.198 26.510 27.343

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 04 frontiersin.org
between minimum and maximum values, reflecting strong interan-
nual fluctuations. In contrast, annual mean temperature varies
within a relatively narrow range, indicating gradual changes over
time rather than abrupt shifts. The identical number of observations
across all variables confirms complete temporal consistency in the
dataset.
Table 2 presents the results of unit root and stationarity tests.
The Augmented Dickey–Fuller test statistics fail to reject the null
hypothesis of a unit root for both crop and livestock production
indices in levels. The KPSS test rejects the null of stationarity for
these series, supporting integration of order one. Rainfall displays
mixed evidence across tests, while temperature exhibits stationarity
under the ADF test but non-stationarity under KPSS. Taken
together, the results indicate a combination of stationary and non-
stationary processes across variables, consistent with mixed orders
of integration.
Figure 1 depicts the evolution of crop and livestock production
indices in Somalia over the period 1989–2021. Both indices exhibit
substantial interannual variation, particularly during the first decade
of the sample. Crop production displays pronounced year-to-year
fluctuations, including several sharp contractions and rebounds,
whereas livestock production follows a comparatively smoother path
with fewer extreme deviations. During the mid-1990s, crop produc-
tion experiences multiple abrupt changes, while livestock production
shows more gradual adjustments. From the early 2000s onward, both
indices exhibit an upward movement, although the degree of volatility
remains consistently higher for crop production. Periods of diver-
gence between the two indices are evident, followed by partial conver-
gence in later years.
Figure 2 depicts annual rainfall and temperature trends over the
study period. Rainfall demonstrates substantial interannual variabil-
ity with alternating high and low values and no persistent monotonic
trend. In contrast, temperature exhibits a smoother and steadily
increasing pattern across time. The differing temporal behavior of
rainfall and temperature highlights the distinct nature of these cli-
mate variables and their potential implications for agricultural
systems.
Figure 3 displays rainfall and temperature anomalies relative to
the long-term mean. Rainfall anomalies alternate frequently between
positive and negative deviations, reflecting recurrent departures from
average conditions. Temperature anomalies, however, become increas-
ingly positive in later years of the sample. Several periods are charac-
terized by simultaneous negative rainfall anomalies and positive
temperature anomalies, indicating the co-occurrence of multiple cli-
mate stressors.
Figure 4 presents cumulative rainfall deficits alongside cumulative
heat stress. The cumulative rainfall deficit series increases during con-
secutive years of below-average rainfall, illustrating the persistence of
drought conditions. Cumulative heat stress rises steadily due to the
sustained presence of positive temperature anomalies. The contrasting
accumulation patterns highlight the temporal persistence of heat
stress relative to episodic rainfall deficits.
Figure 5 illustrates the Climate Stress Index (CSI) and associated
stress regimes. Early years of the sample are dominated by normal and
moderate stress regimes, while high-stress regimes appear with
increasing frequency in later periods. Transitions between regimes
occur intermittently, indicating temporal variability in climate stress
intensity. The duration of high-stress regimes increases toward the end
of the sample.
Figure 6 presents the dynamic asymmetric multipliers of rainfall
shocks on crop production over horizons of up to 5 years. Negative
rainfall shocks generate immediate and sizable cumulative responses,
with the magnitude of the effect increasing progressively across the
forecast horizon. The cumulative response to negative shocks remains
persistently below zero, indicating prolonged adjustment dynamics.
In contrast, positive rainfall shocks produce noticeably smaller cumu-
lative responses, which stabilize more quickly and remain closer to
TABLE 2 Unit root and stationarity tests (1989–2021).
Variable ADF
statistic
ADF
p-value
KPSS
statistic
KPSS p-
value
Crop
Production
Index (FAO)
1.395 0.997 0.877 0.010
Livestock
Production
Index (FAO)
1.311 0.997 0.734 0.010
Annual
rainfall (CRU,
mm)
−2.452 0.128 0.313 0.100
Annual mean
temperature
(CRU, °C)
−4.032 0.001 0.730 0.011
FIGURE 1
Trends in crop and livestock production indices in Somalia (1989–
2021).
FIGURE 2
Annual rainfall and temperature trends in Somalia (1989–2021).

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 05 frontiersin.org
zero throughout the horizon. The divergence between the positive
and negative rainfall paths becomes more pronounced over time,
indicating persistent asymmetry in the dynamic response of crop
production.
Table 3 reports Zivot–Andrews structural break test results. Break
points are endogenously identified near the beginning of the sample
across agricultural and climate variables. Although 1989 is detected as
a break year for all series, this early boundary does not convincingly
capture Somalia’s early-1990s institutional collapse. While instability
justifies the inclusion of structural controls, the inconclusive break
evidence means the post-1991 dummy lacks statistical support.
Relevant test statistics and critical values are reported to clarify this
limitation.
Figure 7 presents corresponding dynamic multipliers for livestock
production. Negative rainfall shocks again produce larger cumulative
responses than positive shocks, though the magnitude and duration
differ from those observed for crops. Livestock responses stabilize
more quickly, suggesting differences in adjustment speed across agri-
cultural systems.
FIGURE 3
Rainfall and temperature anomalies in Somalia (1989–2021).
FIGURE 4
Cumulative rainfall deficit and heat-stress accumulation in Somalia
(1989–2021).
FIGURE 5
Climate stress regimes in Somalia (1989–2021).
FIGURE 6
Dynamic asymmetric multipliers of rainfall shocks on crop
production.
TABLE 3 Structural break tests (Zivot–Andrews, 1989–2021).
Variable Zivot–Andrews
test statistic
Estimated break
year
Crop Production
Index (FAO)
−2.95 1989
Livestock Production
Index (FAO)
−3.46 1989
Annual Rainfall (CRU,
mm)
−7.10 1989
Annual Mean
Temperature (CRU,
°C)
−6.40 1989

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 06 frontiersin.org
Table 4 presents baseline linear ARDL estimates for crop and
livestock production. Lagged dependent variables are positive and
statistically significant in both models, indicating strong persistence
in production dynamics. Coefficients associated with contemporane-
ous and lagged rainfall and temperature variables are generally small
and statistically insignificant. Model fit statistics indicate high
explanatory power, driven primarily by autoregressive components
rather than climate variables.
Table 5 reports long-run asymmetric coefficients from the non -
linear ARDL model for crop production. Coefficients associated with
negative temperature components are larger in absolute magnitude
than those associated with positive temperature components, though
long-run estimates for both rainfall and temperature are imprecisely
FIGURE 7
Dynamic asymmetric multipliers of rainfall shocks on livestock
production.
TABLE 4 Baseline linear ARDL (1,1) estimates for crop and livestock production (1989–2021).
Panel A: Coefficient estimates
Variable Crop
coef
Crop
t-stat
Livestock
coef
Livestock
t-stat
Lagged
production
(t − 1)
1.021 12.08 0.996 26.22
Rainfall (t) −0.011 −0.38 0.012 0.76
Rainfall
(t − 1)
0.003 0.12 −0.003 −0.19
Temperature
(t)
1.832 0.37 −0.402 −0.15
Temperature
(t − 1)
3.436 0.73 3.341 1.34
Constant −139.95 −0.79 −79.87 −0.85
Panel B: Model diagnostics
Statistic Crop Livestock
Observations 31 31
R2 0.948 0.988
AIC 178.15 140.32
Bold values indicate statistical significance at the 1% level.
TABLE 5 NARDL long-run asymmetric effects of rainfall and temperature on crop production
(1989–2021).
Long-run
effect
Coefficient Std.
Error
t-stat
Rainfall (positive) −0.110 0.172 −0.640
Rainfall (negative) −0.178 0.234 −0.759
Temperature
(positive)
36.547 25.987 1.406
Temperature
(negative)
−49.293 32.362 −1.523
Model diagnostics (crop NARDL-ECM): Observations = 31, R2 = 0.622, AIC = 162.79.
TABLE 6 NARDL short-run dynamics and error-correction results for crop production (1989–
2021).
Variable Coefficient Std.
error
t-stat p-value
Constant 12.739 13.894 0.917 0.371
Error-
correction
term (L1_
crop)
−0.300 0.172 −1.743 0.097
Rainfall (+),
long-run level
(L1_rain_pos)
−0.033 0.047 −0.702 0.491
Rainfall (−),
long-run level
(L1_rain_neg)
−0.053 0.058 −0.924 0.367
Temperature
(+), long-run
level (L1_
temp_pos)
10.963 7.405 1.481 0.155
Temperature
(−), long-run
level (L1_
temp_neg)
−14.787 7.474 −1.978 0.063
Rainfall (+),
short-run
change
(drain_pos)
0.009 0.027 0.340 0.738
Rainfall (−),
short-run
change
(drain_neg)
−0.065 0.054 −1.203 0.244
Temperature
(+), short-run
change
(dtemp_pos)
11.396 5.974 1.908 0.072
Temperature
(−), short-run
change
(dtemp_neg)
−16.614 7.223 −2.300 0.033
Post-1991
dummy
−0.108 4.973 −0.022 0.983
Trend −1.650 0.745 −2.216 0.039

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 07 frontiersin.org
estimated, likely reflecting the limited degrees of freedom available
in the annual time series. The direction of effects is nonetheless con-
sistent with theoretical expectations and corroborated by short-run
dynamics. Rainfall-related coefficients are comparatively smaller and
accompanied by larger standard errors. The relative magnitude of
temperature coefficients exceeds that of rainfall coefficients.
Table 6 presents short-run dynamics and error-correction results
for crop production. The error-correction term is negative and statisti-
cally meaningful, confirming adjustment toward long-run equilib-
rium following short-run deviations. Short-run coefficients associated
with negative temperature changes are larger in magnitude than those
associated with positive temperature changes or rainfall variations.
The deterministic time trend is statistically significant, indicating sys-
tematic temporal movement in crop production beyond short-run
fluctuations.
Table 7 reports short-run dynamics and error-correction results
for livestock production. The error-correction coefficient is larger in
absolute magnitude than that estimated for crop production, indicat-
ing faster convergence toward long-run equilibrium. Short-run nega-
tive temperature changes are statistically significant, while
rainfall-related coefficients remain weaker and less consistent across
specifications.
Figure 8 compares long-run negative climate effects across crop
and livestock systems. Negative temperature effects exhibit larger
magnitudes for crop production than for livestock production.
Rainfall-related effects are comparatively smaller for both systems,
with limited differences across agricultural activities.
Figure 9 illustrates dynamic adjustment paths following a negative
temperature shock. Crop production shows larger cumulative declines
and a slower adjustment process across the horizon considered.
Livestock production exhibits a faster convergence toward a stable
cumulative response, with smaller overall magnitudes relative to crops.
4 Discussion
This study examined the effects of climate variability on crop and
livestock production in Somalia over the period 1989–2021 using a
nonlinear autoregressive distributed lag (NARDL) framework. The
findings provide evidence of short-run asymmetric effects of climate
variability on agricultural production, particularly in relation to nega-
tive temperature changes, while the long-run asymmetric estimates
remain comparatively imprecise. Temperature effects receive more
consistent empirical support than rainfall effects across the estimated
models, and crop production exhibits greater sensitivity and slower
adjustment dynamics compared to livestock production. These results
confirm the presence of pronounced nonlinearities in climate–agri-
culture relationships in Somalia, supporting the use of asymmetric
modeling approaches in fragile, climate-exposed settings.
The asymmetric response of agricultural production to climate
shocks represents one of the principal empirical contributions of this
study. The NARDL estimates reveal that negative temperature changes
exert statistically significant short-run effects on both crop and live-
stock production, with coefficients substantially larger in absolute
magnitude than those associated with positive temperature changes.
Dynamic multiplier analysis reinforces this finding, showing that
cumulative production responses to negative rainfall shocks diverge
progressively from those of positive shocks over a five-year horizon.
These patterns align with theoretical expectations that adverse climate
conditions, such as drought or heat stress, generate nonlinear and
potentially irreversible damage to plant physiology, soil moisture, and
animal health (Hlaváčová et al., 2018; Prasad et al., 2015). The com-
bined effects of drought and heat stress have been shown to be more
detrimental than their individual impacts, affecting physiological and
biochemical processes in ways that reduce productivity beyond what
additive models would predict ( Shubha et al., 2025 ; Swapna et al.,
2023). These findings corroborate research from other regions
employing similar asymmetric approaches. Studies in India, Malaysia,
and the Mekong region using NARDL models have documented
TABLE 7 NARDL short-run dynamics and error-correction results for livestock production.
Variable Coefficient Std.
error
t-stat p-value
Constant 28.079 9.629 2.916 0.009
Error-
correction
term (L1_
livestock)
−0.454 0.152 −2.984 0.008
Rainfall (+),
long-run level
(L1_rain_pos)
0.008 0.030 0.274 0.787
Rainfall (−),
long-run level
(L1_rain_neg)
0.070 0.038 1.857 0.079
Temperature
(+), long-run
level (L1_
temp_pos)
2.332 4.365 0.534 0.599
Temperature
(−), long-run
level (L1_
temp_neg)
−7.436 5.114 −1.454 0.162
Rainfall (+),
short-run
change
(drain_pos)
−0.014 0.016 −0.869 0.396
Rainfall (−),
short-run
change
(drain_neg)
0.050 0.033 1.516 0.146
Temperature
(+), short-run
change
(dtemp_pos)
3.094 3.583 0.864 0.399
Temperature
(−), short-run
change
(dtemp_neg)
−9.097 4.216 −2.158 0.044
Post-1991
dummy
1.810 2.912 0.622 0.542
Trend 0.858 0.528 1.624 0.121

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 08 frontiersin.org
comparable asymmetries, with negative climate deviations exerting
disproportionate effects on agricultural yields (Mujtaba et al., 2023;
Thann et al., 2025; Zhang et al., 2023). In Sri Lanka, both positive and
negative changes in climate variables significantly affect rice yields,
with negative changes often conferring benefits that positive changes
do not (Samarasinghe et al., 2025). The consistency of asymmetric
responses across diverse agricultural contexts strengthens confidence
in the generalizability of these patterns and the appropriateness of
nonlinear frameworks for climate–agriculture analysis.
The finding that temperature effects are more consistently sup-
ported than rainfall effects warrants particular attention. The esti -
mated long-run coefficients suggest that negative temperature shocks
may be associated with production declines; however, these estimates
should be interpreted cautiously because the long-run coefficients are
imprecisely estimated. Additionally, although rainfall and temperature
coefficients cannot be compared directly in magnitude because they
are expressed in different units, temperature effects are more consis-
tently supported across the estimated models in terms of their direc-
tion, statistical significance, and persistence, particularly for negative
short-run temperature changes. This pattern may appear counterintui-
tive given the critical importance of water availability in arid and
semi-arid production systems; nevertheless, it is consistent with accu-
mulating evidence from Somalia and comparable environments. Prior
studies in Somalia have documented that rising temperatures consis-
tently exert negative impacts on crop production in both short and
long run, attributable to the country’s high susceptibility to climate
change and the limitations of predominantly rain-fed systems (Abdi
et al., 2025; Abdullahi et al., 2025; Ali et al., 2023; Mohamed et al.,
2024; Omar et al., 2025; Samatar, 2024; Warsame et al., 2021, 2022).
Research focusing on sorghum production in Somalia similarly iden-
tifies temperature and political instability as significant constraints,
while rainfall and cultivated area enhance output ( Warsame et al.,
2022). The temperature sensitivity observed in this study is also con-
sistent with findings from dryland agriculture elsewhere, where yields
of major rainfed crops increase up to a threshold of approximately
32 °C, beyond which they decline sharply due to direct heat stress and
induced water stress (Vargas Zeppetello et al., 2025). In the Greater
Horn of Africa, projected warming under 1.5 °C and 2 °C global
warming scenarios is expected to alter precipitation patterns and fur-
ther affect agricultural productivity (Osima et al., 2018).
A further contribution of this study lies in its comparative analysis
of crop and livestock systems. The error-correction coefficient for live-
stock production is larger in absolute magnitude than that estimated
for crops, indicating faster convergence toward long-run equilibrium
following short-run deviations. Dynamic multiplier paths confirm
that livestock production stabilizes more quickly after climate shocks,
whereas crop production experiences larger cumulative declines and
slower recovery. These differential adjustment dynamics are consistent
with the structural characteristics of pastoral and agro-pastoral sys-
tems. Livestock systems possess inherent buffering mechanisms,
including herd mobility, dietary flexibility, and the capacity to rebuild
populations over successive seasons, which can moderate the persis-
tence of climate shocks (Kebede et al., 2024; Thornton et al., 2011).
Evidence from the Sahel indicates that cattle populations can recover
following drought, although herd structures may remain disrupted for
extended periods (Lesnoff et al., 2012). In Botswana, delayed rainfall
sharply reduces livestock numbers, yet controlled harvesting and
adaptive management can mitigate losses (Nketsang et al., 2025). By
contrast, crop production in Somalia is predominantly rain-fed, with
limited irrigation infrastructure and minimal capacity to substitute
inputs across growing seasons. The absence of effective buffering
mechanisms renders crop systems more vulnerable to persistent cli-
mate stress, a pattern observed in studies of dryland agriculture in
Sub-Saharan Africa and elsewhere (Gachene et al., 2014; Martinez-
Feria and Basso, 2020; Zachariah et al., 2020).
The integration of cumulative climate stress indicators and struc-
tural break controls distinguishes this analysis from prior work on
Somalia’s agricultural sector. The Climate Stress Index constructed in
this study reveals that high-stress regimes have become more frequent
in recent decades, with durations of elevated stress increasing toward
the end of the sample period. This temporal pattern is consistent with
observational evidence that drought frequency and severity have
intensified in the Horn of Africa, contributing to recurrent crop fail-
ures and livestock losses (Funk et al., 2018). The Zivot–Andrews tests
identify structural breaks at the beginning of the sample period, coin-
ciding with Somalia’s institutional collapse in the early 1990s. The
inclusion of a post-1991 structural dummy in the NARDL specifica-
tions controls for these regime shifts, ensuring that estimated climate
effects are not confounded with broader structural changes unrelated
to climate variability.
Several methodological strengths support the validity of the
findings. The NARDL framework accommodates mixed orders of
integration and is well-suited to the small sample sizes typical of
country-level climate–agriculture studies ( Pesaran et al., 2001; Shin
et al., 2014). The use of partial sum decompositions enables direct
testing of asymmetric responses without imposing restrictive func -
tional forms. Dynamic multipliers provide transparent and intuitive
representations of adjustment paths, facilitating comparison across
agricultural systems and climate variables. Robustness checks,
including alternative lag structures, exclusion of extreme drought
years, and sensitivity to sample window selection, yield qualitatively
consistent results. The use of FAO production indices and CRU cli-
mate data ensures comparability with other studies and minimizes
concerns about data quality in a fragile-state context. These findings
carry implications for agricultural policy and adaptation planning
in Somalia and comparable climate-exposed settings. The pro -
nounced asymmetry in climate impacts underscores the importance
of prioritizing protective measures against adverse shocks rather
than relying on recovery under favorable conditions. Investments in
heat-tolerant crop varieties, improved water management, and agri-
cultural infrastructure are consistent with recommendations from
previous regional studies ( Abdullahi et al., 2025 ; Omar et al., 2025 ;
FIGURE 8
Comparison of long-run climate impacts on crop and livestock
production.

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 09 frontiersin.org
Samatar, 2024 ). However, evidence from semi-arid agricultural
regions indicates that the adoption of water-saving technologies is
influenced not only by their technical availability but also by house-
hold characteristics, institutional support, and farmers’ perceptions
(Mohan et al., 2024). The faster and more statistically robust adjust-
ment of livestock production suggests that pastoral livelihoods may
possess relative resilience under climate variability, although extreme
climatic stress can still exceed their adaptive capacity ( Tofu et al.,
2025). Policymakers may benefit from supporting adaptive strategies
such as herd mobility, diversification, and access to water resources,
while also addressing non-climatic constraints including political
instability and infrastructure deficits ( Mohamed et al., 2024 ;
Warsame et al., 2022). Moreover, because instability in crop and live-
stock production may have broader consequences for food avail -
ability and access, future assessments should connect agricultural
production responses with appropriate food-security indicators and
measurement frameworks (Reddy, 2016).
The contribution of these findings should be interpreted in rela-
tion to the direction, persistence, and potential reversibility of the
estimated effects. Rather than showing only whether a climatic vari-
able is positively or negatively associated with agricultural production
on average, the NARDL framework allows positive and negative
changes to be examined separately in both the short and long run. A
significant negative component accompanied by a weaker or insignifi-
cant positive component may suggest that production losses associ-
ated with climatic deterioration are not fully reversed when conditions
improve. Such a pattern may reflect biological damage, depletion of
soil and water resources, loss of productive assets, or delayed invest-
ment and recovery. Conversely, a stronger positive component may
suggest that agricultural production responds more effectively to
favorable climatic conditions when complementary resources and
production capacity are available. The findings therefore contribute to
the ARDL/NARDL literature by providing evidence on direction-
specific adjustment patterns in the climate–agriculture relationship.
The study nevertheless has several methodological limitations.
Although the NARDL framework captures direction-specific
responses, it does not explicitly estimate climatic thresholds or
inverted U-shaped relationships. Furthermore, decomposing rainfall
and temperature into positive and negative components increases the
number of estimated parameters. Together with the structural dummy
and deterministic trend, this reduces the available degrees of freedom
and limits the statistical power to identify long-run relationships. This
limitation is reflected in the relatively large standard errors and gener-
ally weak statistical significance of the long-run coefficients.
Consequently, their signs and relative patterns should be regarded as
indicative rather than definitive, while greater confidence can be
placed in the statistically significant short-run negative-temperature
effects and estimated adjustment dynamics. Future research using
longer or higher-frequency datasets should examine climatic thresh-
olds and reassess the stability and precision of the long-run direction-
specific relationships.
5 Conclusion
This study examined the effects of climate variability on crop and
livestock production in Somalia during 1989–2021 using a nonlinear
autoregressive distributed lag (NARDL) framework. The findings
reveal statistically supported short-run asymmetries in climate–agri-
culture relationships, particularly for negative temperature changes,
while the long-run estimates provide only indicative evidence because
of their limited precision. The estimated temperature effects are gener-
ally larger in magnitude than the rainfall effects, although this com-
parison is more clearly supported in the short-run results than in the
imprecisely estimated long-run relationships. Crop production exhib-
its slower and less statistically robust adjustment, whereas livestock
production converges more rapidly toward equilibrium, potentially
reflecting buffering mechanisms such as herd mobility and dietary
flexibility. The Climate Stress Index further indicates that high-stress
climatic conditions have become more frequent and persistent in
recent decades. These findings highlight the importance of protecting
FIGURE 9
Dynamic adjustment to negative temperature shocks under the NARDL framework.

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 10 frontiersin.org
agricultural systems against adverse climate shocks. Investments in
heat-tolerant crop varieties, improved water management, and adap-
tive support for pastoral systems should form central components of
Somalia’s agricultural adaptation strategy. Nevertheless, the limited
sample size and exclusion of broader institutional and socioeconomic
variables constrain the scope of the conclusions. Future research using
longer or higher-frequency data should examine how climatic, insti-
tutional, and socioeconomic factors jointly shape agricultural produc-
tion and food security in fragile contexts.
Data availability statement
The original contributions presented in the study are included in
the article/supplementary material, further inquiries can be directed
to the corresponding author.
Author contributions
AA: Conceptualization, Data curation, Formal analysis,
Investigation, Methodology, Project administration, Software,
Writing  – original draft. AM: Software, Writing  – original draft,
Writing – review & editing. KA: Data curation, Writing – original
draft, Writing  – review & editing. IA: Writing  – original draft,
Writing – review & editing. AH: Methodology, Writing – original
draft, Writing  – review & editing. AO: Writing  – original draft,
Writing – review & editing. MA: Software, Validation, Writing – origi-
nal draft, Writing – review & editing. SA: Formal analysis, Writing –
original draft, Writing – review & editing.
Funding
The author(s) declared that financial support was not received for
this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the
absence of any commercial or financial relationships that could be
construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the
creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this
article has been generated by Frontiers with the support of artificial
intelligence and reasonable efforts have been made to ensure accuracy,
including review by the authors wherever possible. If you identify any
issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated organizations,
or those of the publisher, the editors and the reviewers. Any product
that may be evaluated in this article, or claim that may be made by its
manufacturer, is not guaranteed or endorsed by the publisher.
References
Abdi, A. H., Ahmed, A. M., and Khalif, A. A. (2025). Agricultural market integration:
investigating the impacts of trade openness and climate change on food production in
Somalia. Cogent Food Agric. 11, 2596345. doi: 10.1080/23311932.2025.2596345
Abdullahi, M. A., Muse, A. H., Muse, Y . H., Hassan, M. A., and Nadarajah, S. (2025). Harvests
under heat: quantifying the long-run impacts of climate and fertilizers on Somalia cereal yields:
an ARDL analysis. Earth Syst. Environ. 10, 6599–6610. doi: 10.1007/s41748-025-00972-w
Ahmed, A. H., and Ali, I. I. (2024). The impact of drought on food security in Somalia: a
comprehensive review. Current Res. Environ. Sci. Ecol. Letters 1, 01–09. doi: 10.33140/
cresel.01.02.05
Alasow, A. A., Hamed, M. M., and Shahid, S. (2024). Spatiotemporal variability of drought
and affected croplands in the horn of Africa. Stoch. Env. Res. Risk A. 38, 281–296. doi:
10.1007/s00477-023-02575-1
Alasow, A. A., Hersi, A. A., and Nadarajah, S. (2025). Climate rhythms in a fragile region:
decoding Somalia’s 120-year rainfall patterns through spatial econometrics. Earth Syst.
Environ. 9, 2045–2060. doi: 10.1007/s41748-025-00738-4
Ali, I. I., and Ahmed, A. H. (2024). Agricultural supply chains: challenges and opportuni-
ties for agribusiness in Afgoye, lower Shabelle, Somalia. Scientific Res. Communications
4, 61–72. doi: 10.52460/src.2024.006
Ali, D. A., Dahir, A. M., and Yusuf, S. M. (2023). Autoregressive distributed lag model-
ing of climate and non-climatic determinants affecting cereal production: empirical
evidence from Somalia. Int. J. Energy Econ. Policy 13, 577–584. doi: 10.32479/
ijeep.14553
Burke, M., Hsiang, S. M., and Miguel, E. (2015). Global non-linear effect of temperature
on economic production. Nature 527, 235–239. doi: 10.1038/nature15725
Dell, M., Jones, B. F ., and Olken, B. A. (2014). What do we learn from the weather?
The new climate-economy literature. J. Econ. Lit. 52, 740–798. doi: 10.1257/
jel.52.3.740
Dickey, D. A., and Fuller, W . A. (1979). Distribution of the estimators for autoregressive
time series with a unit root. J. Am. Stat. Assoc. 74, 427–431.
FAO (2021) In Brief to The State of Food Security and Nutrition in the World 2021 .
Rome: FAO.
FAO, G. D. P (2018). Climate change and the Global Dairy cattle sector–The role of the
Dairy sector in a low-carbon Future. Rome: Food and Agriculture Organization of the
United Nations and Global Dairy Platform Inc.
Food and Agriculture Organization of the United Nations (FAO) and World Bank. (2018).
Rebuilding Resilient and Sustainable Agriculture in Somalia. Rome: FAO.
Funk, C., Harrison, L., Shukla, S., Pomposi, C., Galu, G., Korecha, D., et al. (2018).
Examining the role of unusually warm Indo-Pacific Sea-surface temperatures in recent
African droughts. Q. J. R. Meteorol. Soc. 144, 360–383. doi: 10.1002/qj.3266
Gachene, C. K. K., Karuma, A. N., and Baaru, M. W . (2014). “Climate change and crop
yield in sub-Saharan Africa, ” in Lal, R., Singh, B. R., Mwaseba, D. L., Kraybill, D.,
Hansen, D. O., and Eik, L. O. Eds., Sustainable Intensification to Advance Food Security
and Enhance Climate Resilience in Africa, (Cham: Springer), 165–183.
Hlaváčová, M., Klem, K., Rapantová, B., Novotná, K., Urban, O., Hlavinka, P ., et al. (2018).
Interactive effects of high temperature and drought stress during stem elongation, anthe-
sis and early grain filling on the yield formation and photosynthesis of winter wheat. Field
Crop Res. 221, 182–195. doi: 10.1016/j.fcr.2018.02.022
Kebede, H. Y ., Mekonnen, A. B., Emiru, N. C., Mekuyie, M., and Ayal, D. Y . (2024).
Climate variability and indigenous adaptation strategies by Somali pastoralists in
Ethiopia. Theor. Appl. Climatol. 155, 7259–7273. doi: 10.1007/s00704-024-04993-9
Lesnoff, M., Corniaux, C., and Hiernaux, P . (2012). Sensitivity analysis of the recovery
dynamics of a cattle population following drought in the Sahel region. Ecol. Model. 232,
28–39. doi: 10.1016/j.ecolmodel.2012.02.018
Lobell, D. B., Hammer, G. L., McLean, G., Messina, C., Roberts, M. J., and Schlenker, W .
(2013). The critical role of extreme heat for maize production in the United States. Nat.
Clim. Chang. 3, 497–501. doi: 10.1038/nclimate1832
Lobell, D. B., Schlenker, W ., and Costa-Roberts, J. (2011). Climate trends and global crop
production since 1980. Science 333, 616–620. doi: 10.1126/science.1204531

Alasow et al. 10.3389/fsufs.2026.1828365
Frontiers in Sustainable Food Systems 11 frontiersin.org
Martinez-Feria, R. A., and Basso, B. (2020). Unstable crop yields reveal opportunities for site-
specific adaptations to climate variability. Sci. Rep. 10, 2885. doi: 10.1038/s41598-020-59494-2
Mohamed, A. A., Omar, M. M., and Abdulle, A. Y . (2024). Foreign aid’s role in Somali
agriculture: a detailed empirical study. Int. J. Sustain. Dev. Plan. 19, 3613–3621. doi:
10.18280/ijsdp.190929
Mohamednur, A. K. H., and Ali, I. I. (2025). The effects of climate change on agricultural
sustainability in Somalia. Current Res. Environ. Sci. Ecol. Letters 2, 01–08.
Mohan, G., Perarapu, L. N., Chapagain, S. K., Reddy, A. A., Melts, I., Mishra, R., et al.
(2024). Assessing determinants, challenges and perceptions to adopting water-saving
technologies among agricultural households in semi-arid states of India. Curr. Res.
Environ. Sustainability 7:100255. doi: 10.1016/j.crsust.2024.100255
Mujtaba, A., Pawar, S., Joshi, D. P . P ., and Gopinathan, R. (2023). Examining asymmetric
impact of climatic factors on crop yields in India. Appl. Econ. Lett. 30, 1488–1495. doi:
10.1080/13504851.2022.2064416
Nketsang, T. S., Kassa, S. M., Kgosimore, M., and Tsidu, G. M. (2025). Understanding the
impacts of rainfall variability on natural forage–livestock dynamics in arid and semi-arid
environments. Appl. Sci. 15, 3918. doi: 10.3390/app15073918
Omar, A. O., Alasow, A. A., Farah, A. A., and Shahid, S. (2024). Spatiotemporal analysis
of agricultural drought severity and hotspots in Somaliland. Int. J. Sustain. Dev. Plan. 19,
4135–4146. doi: 10.18280/ijsdp.191104
Omar, M. M., Mohamed, A. A., Abdulle, A. Y ., and Alasow, A. A. (2025). Weathering the
storm: the role of climate variability and production inputs in shaping Somalia’s agricul-
tural output. Int. J. Sustain. Dev. Plan. 20, 709–719. doi: 10.18280/ijsdp.200220
Osima, S., Indasi, V . S., Zaroug, M., Endris, H. S., Gudoshava, M., Misiani, H. O., et al.
(2018). Projected climate over the greater horn of Africa under 1.5 °c and 2 °c global
warming. Environ. Res. Lett. 13:065004. doi: 10.1088/1748-9326/aaba1b
Pesaran, M. H., Shin, Y ., and Smith, R. J. (2001). Bounds testing approaches to the analysis
of level relationships. J. Appl. Econ. 16, 289–326. doi: 10.1002/jae.616
Prasad, P . V . V ., Staggenborg, S. A., and Ristic, Z. (2015). “Impacts of drought and/or heat
stress on physiological, developmental, growth, and yield processes of crop plants, ” in
Ahuja, L. R., Reddy, V . R., Saseendran, S. A., and Yu, Q., Eds., Response of Crops to Limited
Water: Understanding and Modeling Water Stress Effects on Plant Growth Processes ,
(Madison, WI: ASA, CSSA, SSSA), 301–355.
Reddy, A. A. (2016). Food security indicators in India compared to similar countries.
Curr. Sci. 111, 632–640. doi: 10.18520/cs/v111/i4/632-640
Rojas-Downing, M. M., Nejadhashemi, A. P ., Harrigan, T., and Woznicki, S. A. (2017).
Climate change and livestock: impacts, adaptation, and mitigation. Clim. Risk Manag. 16,
145–163. doi: 10.1016/j.crm.2017.02.001
Samarasinghe, B. K. D. J. R., Zhu, Y ., Abeynayake, N. R., Zeng, X., Mathavan, B.,
Wanninayake, R. W . W . M. P . K., et al. (2025). Climate change and rice production in Sri
Lanka: short-run vs. long-run symmetric and asymmetric effects. Front. Sustain. Food
Syst. 9, 1592542. doi: 10.3389/fsufs.2025.1592542
Samatar, E. H. (2024). The influence of climate change on crop production: evidence from
Somalia. Edelweiss Applied Sci. Technol. 8, 1428–1447. doi: 10.55214/25768484.v8i4.1627
Shin, Y . (1992). Testing the null hypothesis of stationarity against the alternative of a unit
root. J. Econ. 54, 159–178.
Shin, Y ., Yu, B., and Greenwood-Nimmo, M. (2014). “Modelling asymmetric cointegra-
tion and dynamic multipliers in a nonlinear ARDL framework, ” in Sickles, R. C., and
Horrace, W . C., Eds., Festschrift in honor of Peter Schmidt: Econometric Methods and
Applications, (New Y ork, NY: Springer), 281–314.
Shubha, K., Dubey, R., Kumar, S., Kumari, M., and Das, A. (2025). “Potential impacts of heat
and drought on vegetable production and quality, ” in Dubey, R., Shubha, K., Kumar, S.,
Rakshit, A., and Das, A., Eds., Drought and Heat Stress in Agriculture: Implications, Mitigation
and Policy Approaches, (Singapore: Springer Nature Singapore). 55–68.
Swapna, B., Kameswaran, S., Ramakrishna, M., and Chandrasekhar, T. (2023). “Impact
of heat coupled with drought stress on plants, ” in Chen, J.-T., Ed., Molecular and
Physiological Insights into Plant Stress Tolerance and Applications in Agriculture ,
(Singapore: Bentham Science Publishers). 200–216.
Tanveer, Z., and Kalim, R. (2025). An empirical analysis of climate transition: a global
outlook of agriculture productivity. J. Econ. Stud. 52, 1025–1042. doi: 10.1108/
JES-07-2024-0466
Thann, O. H., Yuhuan, Z., Ahad, M., Oo, A. T., and Naing, S. Y . (2025). Asymmetric effects
of climate variability: temperature, rainfall, and carbon emissions on agricultural production
in the Mekong region. J. Asia Pac. Econ. 31, 2198–2227. doi: 10.1080/13547860.2025.2587668
Thornton, P . K., Rufino, M. C., Karanja, S., Jones, P . G., Mutie, I., Herrero, M. (2011)
Genesis Reversed: Climate change Impacts on Agriculture and Livelihoods in Mixed crop-
Livestock Systems of East Africa. Nairobi: International Livestock Research Institute (ILRI).
Tofu, D. A., Dilbato, T., Fana, C., Dirbaba, N. B., and Tesso, G. (2025). Analysis of vulnerability,
its drivers, and strategies applied towards reducing the pastoral and agro-pastoral livelihood
vulnerability to climatic shocks. Sci. Rep. 15:2567. doi: 10.1038/s41598-024-79165-w
Vargas Zeppetello, L. R., Proctor, J., and Huybers, P . (2025). Is water stress the root cause
of the observed nonlinear relationship between yield losses and temperature? AGU Adv.
6, e2025AV001704. doi: 10.1029/2025AV001704
Warsame, A. A., Sheik-Ali, I. A., Ali, A. O., and Sarkodie, S. A. (2021). Climate change
and crop production nexus in Somalia: an empirical evidence from ARDL technique.
Environ. Sci. Pollut. Res. 28, 19838–19850. doi: 10.1007/s11356-020-11739-3
Warsame, A. A., Sheik-Ali, I. A., Jama, O. M., Hassan, A. A., and Barre, G. M. (2022). Assessing
the effects of climate change and political instability on sorghum production: empirical evi-
dence from Somalia. J. Clean. Prod. 360:131893. doi: 10.1016/j.jclepro.2022.131893
World Bank. (2023). Somalia Country Climate and Development Report. Washington, DC:
World Bank Group.
Yusuf, H. A. N., Alasow, A. A., and Mennan, H. (2024). Significant weed species, density
and frequency in sesame cultivated areas of middle Shabelle Province in Somalia. Acta
Agric. Scand. Sect. B Soil Plant Sci. 74:2392525. doi: 10.1080/09064710.2024.2392525
Zachariah, M., Mondal, A., Das, M., Achutarao, K. M., and Ghosh, S. (2020). On the role
of rainfall deficits and cropping choices in loss of agricultural yield in Marathwada, India.
Environ. Res. Lett. 15:094029. doi: 10.1088/1748-9326/ab93fc
Zhang, Q., Akhtar, R., Saif, A. N. M., Akhter, H., Hossan, D., Alam, S. M. A., et al. (2023).
The symmetric and asymmetric effects of climate change on rice productivity in Malaysia.
Heliyon 9:e16118. doi: 10.1016/j.heliyon.2023.e16118
Zivot, E., and Andrews, D. W . K. (2002). Further evidence on the great crash, the
oil-price shock, and the unit-root hypothesis. J. Bus. Econ. Stat. 20, 25–44. doi:
10.1198/073500102753410372

论文原文与出版信息

Frontiers in Sustainable Food Systems
查看论文原文

继续探索