INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue III, March 2025
www.ijltemas.in Page 270
The results of the regression analysis indicated that the model's Sum of Squares (SS) for rainfall variability was 0.000373 with 1
degree of freedom (df), resulting in a Mean Square (MS) of 0.000373211. The statistical insignificance of the model was evident
from the F-statistic value of 0.000 and a corresponding probability (Prob > F) of 0.946, indicating no meaningful relationship
between rainfall variability and maize yields. Further, the regression results showed an R-squared value of 0.000, signifying that
the model explained none of the variability in maize yields. The adjusted R-squared value was -0.007, reinforcing that the
inclusion of rainfall variability as a predictor did not enhance the model's explanatory power. The study showed a negative but
insignificant effect of rainfall variability on maize yields among farming households in Machakos County (Ξ²=-0.020, p=0.946).
This implies that, holding other factors constant, a one-unit increase in rainfall is associated with a 0.020 decrease in maize yields
(measured in metric tons per hectare). Thus, rainfall variability has severe implications on the yields of maize.
Effects of Rainfall anomalies on Maize yields in Machakos County
Rainfall mean data shows considerable variability over the years, with fluctuations ranging from a low of 17.9 mm in 2022 to a
high of 96.2 mm in 2006. The mean rainfall over the entire period is 59.9 mm. Maize production has varied significantly, with the
highest production was 143,825 MT in 1994, and the lowest was 55,300 MT in 2005. There appears to be a general increase in
production in recent years, with notable peaks in 2009 and 2015. In years with higher rainfall (e.g., 1994, 2006, 2015), maize
production also tended to be higher, suggesting that adequate rainfall supports better crop yields. Conversely, years with lower
rainfall (e.g., 2000, 2005, and 2020) often show lower maize production, indicating that insufficient rainfall adversely impacts
yields. Periods of excessive rainfall can lead to flooding, for instance, the high rainfall in 2006 coincided with substantial maize
production but can also be attributed to potential risks of over-saturation. Insufficient rainfall often leads to drought conditions,
which significantly impact maize yields. For instance, the low rainfall in 2000 and 2005 aligns with reduced maize production,
highlighting the adverse effects of drought on crop yields. When considering rainfall anomalies, both the mean rainfall (r=0.0540)
and deviations in rainfall from the mean (r=0.0619) show very weak positive correlations with maize yields. These low
correlations suggest that variations in rainfall have a minimal effect on maize yields. This could imply that other factors may have
a more pronounced impact on maize yields, or that the crop is relatively resilient to variations in rainfall within the observed
range. Generally, moderate to high rainfall correlates positively with maize yields, provided the rainfall is well-distributed and
occurs at critical growth stages. Both excessive and deficient rainfall can lead to anomalies in maize yields. Excessive rainfall
may result in waterlogged fields, while deficient rainfall can cause drought stress, both leading to lower yields. In 1994, higher
rainfall was associated with high maize production (143,825 MT). In contrast, 2005 experienced low rainfall and correspondingly
lower maize production (55,300 MT), highlighting the detrimental effect of drought conditions.
Figure 7: Rainfall mean vs Production in MT
III. Conclusions and Recommendations
The study on the effect of extreme temperature and rainfall anomalies on maize yields among farming households in Machakos
County from 1993 to 2023 has yielded several key conclusions. The relationship between cultivated area and production has been
volatile, with production showing significant variability despite the cultivated area remaining within a certain range. For instance,
in 2000, despite a cultivated area of 162,000 hectares, production plummeted to 58,320 metric tonnes. Similarly, in 2015, the
cultivated area was 125,652 hectares, yet production surged to 121,682 metric tonnes. Rainfall variability has a significant impact
on agricultural yields across the sub-counties. The years 1993, 1995, 1996, 1999, 2000, 2001, 2003, 2004, 2007, 2008, 2009,
2011, 2012, 2013, 2014, 2016, 2017, 2021, and 2022 experienced positive deviations from the mean rainfall, indicating higher-
than-average rainfall. Conversely, the years 1994, 1997, 1998, 2002, 2005, 2006, 2010, 2015, 2018, 2019, 2020, and 2023
witnessed negative deviations from the mean, indicating lower-than-average rainfall. The study found a statistically significant
negative effect of temperature variability on maize yields, with a coefficient (Ξ²) of -0.054 and a p-value of 0.000. This suggests
that as temperature becomes more unpredictable, maize yields suffer considerably. In contrast, the study found a negative but
statistically insignificant effect of rainfall variability on maize yields, with a coefficient (Ξ²) of -0.020 and a p-value of 0.946. This
implies that, holding other factors constant, a one-unit increase in rainfall variability is associated with a 0.020 decrease in maize
yields.