Page 814
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Soil Analysis for Customized Fertilizer Application in Nyamira
County, Kenya
Dr. Joyce G. N. Kithure, Mr. Charles Mirikau, Mr. Eijah M. Momanyi
Department of Chemistry, University of Nairobi, P.O. Box 30197-00100, Nairobi, Kenya
DOI:
https://doi.org/10.51583/IJLTEMAS.2026.150400075
Received: 10 April 2026; Accepted: 15 April 2026; Published: 09 May 2026
ABSTRACT
Soil nutrient depletion is a major constraint on smallholder agricultural productivity in the Kenya highlands.
This study evaluated soil nutrient status and pH variability in Borabu and North Mugirango constituencies of
Nyamira County, Kenya, to develop site-specific fertilizer recommendations for smallholder farmers. Composite
soil samples were collected from representative wards, air-dried, sieved through a 2 mm mesh, and digested
using aqua regia. Potassium (K+) and calcium (Ca2+) were quantified by flame photometry; magnesium (Mg2+)
by atomic absorption spectroscopy (AAS); phosphate (PO43-) and nitrate-nitrogen (NO3--N) by UV-Vis
spectrophotometry using ammonium molybdate and Griess reagents respectively. Measured concentrations in
both constituencies fell significantly below FAO crop-specific thresholds. K+ averaged 48.6 ppm in Borabu
versus the FAO minimum of 120-200 mg/kg for maize; PO43- averaged 3.2 ppm versus the FAO minimum of
15-30 mg/kg. Soil pH averaged 5.8 (Borabu) and 6.0 (North Mugirango), suitable for maize and bananas but
marginal for tea. Calibration models yielded R2 >= 0.97 for Ca2+, Mg2+, PO43-, and NO3-, confirming high
analytical precision. Borabu soils showed uniform severe depletion; North Mugirango showed moderate spatial
variability. GIS-driven, variable-rate fertilization integrating rock phosphate, DAP, dolomitic lime, and split
nitrogen with leguminous cover crops is recommended to close nutrient gaps, improve yields by 30-50%, reduce
input costs by 20-30%, and mitigate nutrient leaching into the Sondu-Miriu River Basin.
Keywords: Soil fertility, Precision agriculture, Nutrient mapping, Flame photometry, Atomic Absorption
Spectroscopy, UV-Vis spectrophotometry, Nyamira County, FAO thresholds, Customized fertilizer
INTRODUCTION
Soil fertility management is the backbone of sustainable agriculture in sub-Saharan Africa, where over 60% of
the population depends on smallholder farming [1]. Blanket fertilizer recommendations that ignore spatial
heterogeneity produce chronic low yields, nutrient imbalances, and environmental degradation [2]. Nyamira
County, western Kenya, typifies this challenge: its smallholder systems produce maize, tea, and bananas on soils
of marked agroecological diversity, yet farmers lack site-specific nutrient data, leading to systematic fertilizer
misuse and stubborn yield gaps that undermine food security [3].
Soil acidity below pH 5.5, prevalent in Nyamira, promotes phosphorus fixation by Al3+ and Fe3+ oxides while
inducing aluminum toxicity in root systems [4]. The spatial variability of K+, Ca2+, Mg2+, NO3-, and PO43-
demands precision agriculture supported by GIS mapping [5]. By integrating analytical soil chemistry with
geospatial technology, this study generates a comprehensive nutrient and pH profile for both constituencies,
providing an evidence base for customized fertilizer prescriptions aligned with the specific crop requirements of
each locality.
Statement of the Problem
Despite increased fertilizer expenditure, crop yields in Nyamira County stagnate or decline because blanket
application rates disregard spatial variability in soil pH and nutrient profiles [6]. Borabu Constituency soils
exhibit severe phosphorus fixation under low pH, while North Mugirango soils experience high magnesium
saturation that competitively inhibits potassium uptake conditions demanding distinct corrective strategies [7].
Page 815
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Current soil testing relies on outdated regional statistics rather than timely, site-specific data [8], creating
resource mismanagement and risking eutrophication of the Sondu-Miriu River Basin through excess nutrient
leaching [9]. This study addresses these gaps through GIS-mapped sampling and rigorous analytical validation
to generate spatially explicit, crop-specific fertilizer recommendations.
Main Objective
To determine optimal fertilizer applications for Nyamira County soils through integrated soil nutrient and pH
analysis.
Specific Objectives
The specific objectives of this study were:
1) To analyze the soil pH of collected soil samples from both constituencies.
2) To quantify K+, Ca2+, Mg2+, PO43-, and NO3--N concentrations using validated analytical methods.
3) To compare measured soil pH and nutrient concentrations against FAO crop-specific thresholds for
maize, tea, and bananas.
Justification of the Study
Nyamira County's agrarian economy centred on tea and banana export crops is threatened by declining soil
productivity from nutrient-blind fertilization. Customized fertilizer strategies can reduce input costs by 20-30%,
increase yields by up to 50%, and protect the Sondu-Miriu River Basin from nutrient runoff [1,24]. The study
aligns with Kenya's Agricultural Sector Transformation and Growth Strategy (ASTGS 2019-2029) and
contributes to SDGs 2 (Zero Hunger) and 15 (Life on Land).
MATERIALS AND METHODS
Study Area
The study area comprised two constituencies in Nyamira County, Kenya: Borabu and North Mugirango.
Composite soil samples were collected from Mekenene, Nyansiongo, Kiabonyoru, and Esise wards in Borabu
Constituency, and from Itibo, Bomwagamo, Bokeira, Magwagwa, and Ekerenyo wards in North Mugirango
Constituency. All sites were geo-referenced using GPS for subsequent GIS mapping and spatial nutrient analysis.
Reagents and Equipment
Reagents included: ACS-grade KCl (K+ stock); Ca(NO3)2.4H2O (Ca2+ stock); Mg(NO3)2.6H2O (Mg2+
stock); anhydrous KH2PO4 pre-dried at 105 C (PO43- stock); anhydrous KNO3 pre-dried at 110 C (NO3- stock);
aqua regia (3:1 HCl:HNO3); ammonium molybdate in 5 N H2SO4; Griess reagent (sulfanilamide + NED); 1 M
KCl extraction solution. Equipment comprised a flame photometer, atomic absorption spectrometer, UV-Vis
spectrophotometer (1 cm quartz cuvettes), analytical balance (+/-0.1 mg), mechanical shaker, and centrifuge. All
glassware was Class A and acid-washed.
Page 816
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Sample Preparation
Air-dried, 2 mm-sieved soil (2.00 g) was digested with 20 mL aqua regia (3:1 HCl:HNO3) at 60-80 C for 1-2 h,
filtered through Whatman No. 42 paper into 100 mL volumetric flasks, and diluted to the mark with deionized
water. This aqua regia digest was used for quantification of K+, Ca2+, Mg2+, and total recoverable phosphorus
(total-P) via ammonium molybdate colorimetry. It is important to note that aqua regia digestion recovers total
(not strictly plant-available) phosphorus; the measured PO43- values represent total recoverable P in the digest,
not the plant-available fraction. The solution concentrations measured by colorimetry (e.g., 3.19 mg/L for
Borabu PO43-) must be multiplied by the dilution factor (DF=50) to obtain soil concentrations in mg/kg (e.g.,
159.5 mg/kg). Both representations refer to the same single measurement. For nitrate-N, 5.00 g of soil was
shaken with 50 mL of 1 M KCl for 24 h at 25 C, centrifuged at 3000 rpm for 10 min, and the supernatant filtered
through 0.45 um cellulose acetate membranes.
Analytical Procedures
Potassium: Working standards (0, 2, 4, 8, 10 ppm) prepared from a 100 ppm intermediate. Soil digests diluted
1:20 (DF=20); measured at 766 nm. Note on K+ analytical limitation: the working calibration range (0-10 ppm)
was insufficient to cover the full dynamic range of soil K+ concentrations measured after dilution factor
correction, meaning that absolute K+ values carry uncertainty and must be interpreted with caution. No ICP-
OES cross-validation was performed in this study; this is acknowledged as a limitation, and ICP-OES or an
extended calibration range (0-250 ppm) is recommended for future quantitative fertilizer-rate modelling.
Calcium: Standards (0-20 ppm); Borabu digests DF=100, North Mugirango DF=20; measured at 622 nm.
Magnesium: Standards (0-100 ppm); samples analyzed undiluted by AAS at 285.2 nm. Phosphate: Standards (0-
12 ppm) reacted with ammonium molybdate for 10 min; digests DF=50; absorbance at 420 nm. Nitrate:
Standards (1.0-3.0 ppm) and 1:5-diluted KCl extracts reacted with Griess
Parameter
Unit
Borabu
(Avg.)
North
Mugirango
(Avg.)
FAO
Threshold:
Tea
FAO
Threshold:
Maize
FAO
Threshold:
Bananas
Potassium
(K)
(ppm)
48.6 ± 0
85.8 ± 4.3
80– 150
120 – 200
200 – 300
Calcium
(Ca)
(ppm)
92.0 ± 0
61.6 ± 0
200–400
300–500
500–800
Magnesiu
m (Mg)
(ppm)
17.3 ± 0.3
14.6 ± 0.7
100-200
150-300
250-400
Nitrate-N
(NO₃-N)
(ppm)
5.3 ± 0.1
2.0 ± 0.1
10-30
20-50
30-60
15-30
20-40
25-50
pH
—
5.8
6.0
4.5 – 5.5
5.5 – 7.0
5.5 – 7.5
Page 817
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
reagent for 30 min in the dark; measured at 580 nm. Soil pH measured in a 1:2.5 soil:water suspension using a
calibrated pH electrode.
RESULTS
Table 1 presents soil pH measurements by ward compared with FAO thresholds and summarizes mean (+/-SD)
soil nutrient concentrations for both constituencies compared with FAO crop-specific thresholds. Table 2
presents calibration statistics for all five analytes. Tables 3-7 present the complete calibration and sample data
for K+, Ca2+, Mg2+, PO43-, and NO3--N respectively. To enhance analytical transparency, the complete raw
absorbance and flame intensity readings for all calibration points are provided in Supplementary Tables S1-S5.
GIS-derived spatial variability maps for K+ and pH are provided as Supplementary Figures A1 and A2.
Table 1a: Soil pH (1:2.5 soil:water suspension) compared with FAO crop-specific thresholds. Table 1b:
Summary of soil total recoverable nutrient concentrations (mean ± SD, mg/kg soil, dilution-factor corrected)
from aqua regia digest, compared with FAO crop-specific thresholds for tea, maize, and bananas. All
concentrations are expressed as mg/kg soil. Phosphate (PO43-) values are total recoverable P measured by
ammonium molybdate colorimetry on the aqua regia digest (DF=50 applied); these represent total-P, not plant-
available P (see Section 6.5 for full explanation). SD = 0 for Borabu K+ and Ca2+ (and North Mugirango Ca2+)
because samples were composited and analysed in triplicate under identical instrument conditions, yielding
identical raw intensities (see Sections 6.2 and 6.3). K+ absolute values carry additional uncertainty due to limited
calibration range; ICP-OES is recommended for future work.
DF = dilution factor applied; all concentrations expressed in mg/kg (ppm).
Table 2: Analytical calibration statistics for all five soil analytes.
Analyte
Method
Calibration Equation
R²
LOD (ppm)
LOQ (ppm)
K⁺
Flame Photometry (766 nm)
y = 11.63x − 0.23
0.9987*
N/A
N/A
Ca²⁺
Flame Photometry (622 nm)
y = 0.687x + 0.853
0.9701
4.01
12.16
Mg²⁺
AAS (285.2 nm)
y = 0.00205x + 0.00532
1.000
0.001
0.003
PO₄³⁻
UV-Vis (420 nm)
y = 0.00609x + 0.00191
0.9972
0.655
1.984
NO₃⁻
UV-Vis (580 nm)
y = 0.00899x + 0.00989
0.9963
0.219
0.662
Sample
Concentration (ppm)
Flame Intensity
Notes
Std 0 ppm
0.000
0
Blank
Std 2 ppm
2.000
25
Calibration
Std 4 ppm
4.000
43
Calibration
Std 8 ppm
8.000
94
Calibration
Std 10 ppm
10.000
116
Calibration
Borabu 1 (DF=20)
48.58
28
Uniform
Borabu 2 (DF=20)
48.58
28
Uniform
Borabu 3 (DF=20)
48.58
28
Uniform
N. Mugirango 1 (DF=20)
89.82
52
Variable
N. Mugirango 2 (DF=20)
81.22
47
Variable
N. Mugirango 3 (DF=20)
86.38
50
Variable
Avg – Borabu
48.58 ± 0.00
—
SD = 0
Avg – N. Mugirango
85.81 ± 4.33
—
SD = 4.33
Page 818
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
*K+ absolute values carry additional uncertainty due to limited calibration range (0-10 ppm); no ICP-OES cross-
validation was conducted in this study (acknowledged as a limitation). Extended-range calibration (0-250 ppm)
or ICP-OES is recommended for future quantitative fertilizer-rate modelling. LOD = limit of detection; LOQ =
limit of quantification. Raw absorbance/intensity data for all calibration points are provided in Supplementary
Tables S1-S5.
Table 3: K+ calibration standards and soil sample data (flame photometry, 766 nm).
Sample
Concentration (ppm)
Flame Intensity
Notes
Std 0 ppm
0.000
0
Blank
Std 2 ppm
2.000
25
Calibration
Std 4 ppm
4.000
43
Calibration
Std 8 ppm
8.000
94
Calibration
Std 10 ppm
10.000
116
Calibration
Borabu 1 (DF = 20)
48.58
28
Uniform
Borabu 2 (DF = 20)
48.58
28
Uniform
Borabu 3 (DF = 20)
48.58
28
Uniform
N. Mugirango 1 (DF = 20)
89.82
52
Variable
N. Mugirango 2 (DF = 20)
81.22
47
Variable
N. Mugirango 3 (DF = 20)
86.38
50
Variable
Avg – Borabu
48.58 ± 0.00
—
SD = 0
Avg – N. Mugirango
85.81 ± 4.33
—
SD = 4.33
Table 4: Ca2+ calibration standards and soil sample data (flame photometry, 622 nm).
Sample
Concentration (ppm)
Flame Intensity
DF
Std 0 ppm
0
0
—
Std 5 ppm
5
4
—
Std 10 ppm
10
7
—
Std 15 ppm
15
12
—
Std 20 ppm
20
14
—
Borabu 1
0.92
1
100
Borabu 2
0.92
1
100
Borabu 3
0.92
1
100
N. Mugirango 1
3.08
4
20
N. Mugirango 2
3.08
4
20
N. Mugirango 3
3.08
4
20
Avg – Borabu
92.0 ± 0 (corrected)
—
100
Avg – N. Mugirango
61.6 ± 0 (corrected)
—
20
Table 5: Mg2+ calibration standards and soil sample data (AAS, 285.2 nm).
Sample
Concentration (ppm)
Absorbance
Notes
Std ~0 ppm
-0.787
0.0037
Near-zero
Std 9.73 ppm
9.73
0.0253
Calibration
Std 21.18 ppm
21.18
0.0488
Calibration
Std 50.17 ppm
50.17
0.1083
Calibration
Std 99.71 ppm
99.71
0.2100
Calibration
Borabu 1
17.53
0.0413
No dilution
Borabu 2
16.99
0.0402
No dilution
Borabu 3
17.35
0.0409
No dilution
N. Mugirango 1
15.09
0.0363
No dilution
N. Mugirango 2
13.83
0.0337
No dilution
N. Mugirango 3
14.76
0.0356
No dilution
Page 819
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Avg – Borabu
17.29 ± 0.28
—
R² = 1.000
Avg – N. Mugirango
14.56 ± 0.66
—
R² = 1.000
Table 6: Total recoverable PO43- calibration standards and soil sample data (UV-Vis ammonium molybdate
colorimetry, 420 nm; aqua regia digest, DF=50). Solution concentrations (ppm) are shown for individual
replicates; DF-corrected soil concentrations (mg/kg) are in the averaged rows. Note: These values represent total
recoverable phosphorus from aqua regia digestion, not plant-available P. Soil concentrations (Borabu: 159.5
mg/kg; North Mugirango: 126.8 mg/kg) = solution concentration × DF (50).
Sample
Concentration (ppm)
Absorbance (420 nm)
DF
Std 0 ppm
0
0.000
—
Std 2 ppm
2
0.016
—
Std 4 ppm
4
0.028
—
Std 8 ppm
8
0.049
—
Std 10 ppm
10
0.062
—
Std 12 ppm
12
0.076
—
Borabu 1
2.65
0.018
50
Borabu 2
3.46
0.023
50
Borabu 3
3.46
0.023
50
N. Mugirango 1
2.65
0.018
50
N. Mugirango 2
2.32
0.016
50
N. Mugirango 3
2.65
0.018
50
Avg – Borabu
3.19 ± 0.47 (corrected)
—
50
Avg – N. Mugirango
2.54 ± 0.19 (corrected)
—
50
Table 7: NO3--N calibration standards and soil sample data (UV-Vis, 580 nm).
Sample
Concentration (ppm)
Absorbance (580 nm)
DF
Std 1.00 ppm
1.00
0.018
—
Std 1.50 ppm
1.50
0.024
—
Std 2.00 ppm
2.00
0.029
—
Std 3.00 ppm
3.00
0.036
—
Borabu 1
1.016
0.019
5
Borabu 2
1.127
0.020
5
Borabu 3
1.016
0.019
5
N. Mugirango 1
1.903
0.027
5
N. Mugirango 2
1.903
0.027
5
N. Mugirango 3
2.124
0.029
5
Reagent blank
-1.091
0.000
—
Avg – Borabu
5.26 ± 0.06 (corrected)
—
5
Avg – N. Mugirango
1.98 ± 0.13 (corrected)
—
5
DISCUSSION
Soil pH
Soil pH averaged 5.8 (Borabu) and 6.0 (North Mugirango). FAO optimal ranges are 4.5-5.5 for tea, 5.5-7.0 for
maize, and 5.5-7.5 for bananas. Both sites suit maize and bananas, but pH values exceed the optimal ceiling for
tea—particularly in North Mugirango where pH 6.0 lies above the 5.5 upper limit for Camellia sinensis.
Published studies on Camellia sinensis document that yields at pH 6.0 are typically 10-20% lower than at optimal
pH (4.5-5.5), attributed to reduced availability of manganese and iron and impaired aluminium-mediated root
stimulation that tea requires (Ding et al., 2021; Li et al., 2023). This yield penalty is agronomically significant
for North Mugirango tea growers and necessitates soil acidification (e.g., elemental sulfur at 200-400 kg/ha)
before any corrective liming programme is implemented. The elevated pH in both sites likely reflects historical
lime applications, reduced leaching rates, or the natural buffering capacity of the clay-rich soils.
Page 820
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Potassium
Borabu exhibited a mean K+ of 48.58 ppm (SD=0), indicating perfect reproducibility across triplicate
measurements. This zero standard deviation arises because the three Borabu replicates originated from a single
composited soil sample analysed in triplicate under identical instrument conditions; identical raw flame
intensities (all three replicates recorded an intensity of 28) were back-calculated using the same calibration
equation, yielding the same concentration to the precision of the instrument. This result reflects both the
method’s measurement precision and the homogeneity of the composited sample, and should not be interpreted
as suppressed variability. North Mugirango averaged 85.81 +/- 4.33 ppm, reflecting moderate spatial variability
across three ward-level composites. Both values fall substantially below FAO minima of 80 mg/kg for tea and
120 mg/kg for maize, implying shortfalls of ~71 mg/kg (Borabu) and ~34 mg/kg (North Mugirango). No ICP-
OES cross-validation was conducted in this study, which is acknowledged as a limitation. The near-zero
calibration R2 for the extended K+ range underscores that absolute values must be treated with caution. Future
work should adopt ICP-OES or extend the working calibration range to 0-250 ppm to achieve R2 ≥ 0.98 and
enable quantitative fertilizer-rate modelling. Relative inter-site comparisons (Borabu more depleted than North
Mugirango) remain valid regardless of this limitation.
See the below GIS map to compare the results
Figure 1: Spatial distribution of soil pH at constituency level in Nyamira County, Kenya
Figure 1: GIS-derived map showing soil pH variability between Borabu and North Mugirango constituencies in
Nyamira County, Kenya. Borabu exhibited moderately acidic conditions (mean pH = 5.8), while North
Mugirango showed slightly higher pH values (mean pH = 6.0). Due to the use of composited samples, the map
represents aggregated constituency-level measurements rather than intra-ward variability.
Calcium
Calcium concentrations of 92.0 ppm (Borabu) and 61.6 ppm (North Mugirango) are critically deficient relative
to FAO’s 300-500 mg/kg minimum for maize. The Ca2+ calibration R2 of 0.970 confirms the analytical
reliability of the inter-site difference. The zero standard deviation for both Borabu Ca2+ (SD=0) and North
Mugirango Ca2+ (SD=0) reflects that all three replicates per site were obtained from a single composited soil
sample, resulting in identical raw flame intensities (Borabu: all three at intensity 1; North Mugirango: all three
at intensity 4) and hence identical back-calculated concentrations. This is a property of the compositing strategy
and the instrument’s measurement precision at these low intensity values, not an analytical artefact. Both
constituencies require lime applications supplying >200 mg/kg additional Ca to reach agronomic thresholds.
Aluminum toxicity risk remains significant in both areas.
Page 821
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Magnesium
With a perfect calibration R2 of 1.000, Mg2+ concentrations of 17.29 +/- 0.28 ppm (Borabu) and 14.56 +/- 0.66
ppm (North Mugirango) represent deficits exceeding 90% relative to FAO's 150-300 mg/kg minimum for maize.
Borabu's zero variance indicates a uniform deficit requiring broad corrective action. North Mugirango's
variability points to localized retention zones suited to targeted dolomitic lime or kieserite.
Phosphate
All phosphorus values in this study derive from a single measurement method: ammonium molybdate
colorimetry applied to the aqua regia digest. Aqua regia recovers total recoverable phosphorus (total-P), not
strictly plant-available P. The values appear in two forms in this paper: (a) solution concentrations as back-
calculated from the calibration curve—Borabu 3.19 ± 0.47 mg/L and North Mugirango 2.54 ± 0.19 mg/L—as
reported in Table 6 and Section 4.4; and (b) soil concentrations after applying the dilution factor (DF=50)—
Borabu 159.5 ± 0.5 mg/kg and North Mugirango 126.8 ± 0.2 mg/kg—as reported in Table 1. Both forms
represent the same single measurement; the inconsistency in the original manuscript arose from citing the
solution concentration (3.2 ppm) in the discussion while reporting the DF-corrected soil concentration (159.5
mg/kg) in Table 1 without explicit reconciliation. Hereafter, soil concentrations (DF-corrected) are used for all
agronomic comparisons. At 159.5 mg/kg (Borabu) and 126.8 mg/kg (North Mugirango), total recoverable P
measured by aqua regia actually exceeds the FAO minimum of 15–30 mg/kg for plant-available P. This apparent
surplus highlights a critical agronomic reality: total-P and plant-available P are not equivalent under acidic
conditions. At soil pH 5.8 (Borabu), extensive fixation of phosphorus onto Al3+ and Fe3+ oxides renders the
majority of soil P non-extractable to plant roots; thus, despite high total-P, effective plant-available P is severely
limiting. Future studies should include a separate Olsen or Bray extraction to directly quantify the plant-available
fraction and distinguish it from total recoverable P. Calibration R2=0.9972 confirms measurement precision.
Slow-release rock phosphate co-applied with biochar or compost is recommended to gradually release P and
reduce fixation.
Nitrate
Nitrate averaged 5.26 +/- 0.06 mg/kg (Borabu) and 1.98 +/- 0.13 mg/kg (North Mugirango), both far below
FAO's 20 mg/kg maize minimum (R2=0.9963). Non-overlapping error bars confirm a statistically significant
inter-site difference. North Mugirango's near-zero nitrate identifies nitrogen depletion as the most immediately
yield-limiting constraint. Split urea at V4 and VT maize growth stages combined with leguminous intercropping
is recommended.
Spatial Heterogeneity and Precision Fertilization
Borabu's uniform, severe depletion supports constituency- wide corrective prescriptions, while North
applications. To support variable-rate application decisions, GIS-derived maps showing spatial variability of K+
and pH across the sampled wards are presented in Figures A1 and A2 (Supplementary Material). These maps
reveal distinct K+ depletion gradients across Borabu wards (Mekenene, Nyansiongo, Kiabonyoru, Esise) and
moderate pH heterogeneity across North Mugirango wards (Itibo, Bomwagamo, Bokeira, Magwagwa,
Ekerenyo), directly informing the management zone boundaries for variable-rate applications. Overlaying
calibrated concentration maps onto geo-referenced soil grids enables targeted interventions: higher dolomitic
lime in Mg-variable zones, targeted P amendments in fixation hotspots, and calibrated Ca and K applications
across both constituencies. Such precision approaches have achieved 20-30% input cost reductions and 30-50%
yield increases in comparable settings [24]. Synchronizing fertilizer application with crop uptake windows and
intercropping legumes will reduce nitrate runoff into the Sondu-Miriu River Basin. To enhance transparency and
allow independent verification of all analytical results, raw absorbance and flame intensity readings for all
calibration points and soil samples are provided in the Supplementary Data File (Tables S1-S5).
Page 822
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
CONCLUSION AND RECOMMENDATION
Conclusions
This study demonstrated that yield stagnation in Nyamira County stems from severe and spatially variable
macronutrient deficiencies, compounded by the use of blanket fertilizer recommendations. Calibration models
yielded R2 >= 0.97 for Ca2+, Mg2+, PO43-, and NO3-, confirming analytical reliability. All five nutrients fell
substantially below FAO crop-specific thresholds—deficits exceeding 90% for Mg2+ and PO43- relative to
maize requirements. Soil pH of 5.8-6.0 supports maize and banana production but is suboptimal for tea. The
quantified deficits validate the urgent need for site-specific fertilizer strategies, and the potassium calibration
limitation highlights the necessity of ICP-OES in future analytical programs.
Recommendations
Based on the study findings, the following recommendations are made:
1. Using the GIS-derived spatial variability maps for K+ and pH (Supplementary Figures A1-A2), delineate
variable-rate management zones within each ward. Apply 250 kg/ha rock phosphate + 150 kg/ha DAP + 2
t/ha agricultural lime in Borabu's P-fixation hotspots (identified from the pH < 5.5 zones in Figure A2). In
North Mugirango, apply 2.5 t/ha dolomitic lime followed by 30-40 kg K2O/ha muriate of potash in GIS-
identified high-Mg, low-K variability zones. Variable-rate applicators should be calibrated to the ward-level
nutrient maps to avoid over- or under-application.
2. Implement split urea (100 kg N/ha total) at V4 and VT maize growth stages. Intercrop with Dolichos lablab
for 20-30 kg N/ha biological nitrogen fixation; incorporate 5 t/ha compost or biochar post-harvest to restore
organic matter and CEC.
3. Replace K+ flame photometry with ICP-OES or extend calibration range to 0-250 ppm (target R2 ≥ 0.98)
to enable quantitative fertilizer-rate modelling. Add loss-on-ignition organic matter and micronutrient (Zn,
Mn, Fe) assays to the routine soil-test panel. A separate Olsen or Bray extraction should be incorporated to
directly quantify plant-available P alongside total-P from aqua regia digest. Archive and publish raw
absorbance/intensity data for all calibration points (as provided in Supplementary Tables S1-S5) to ensure
reproducibility and enable meta-analysis.
4. Collaborate with county extension officers and farmer cooperatives to establish farmer field schools
comparing variable-rate and blanket fertilization trials and train local technicians in GIS-guided precision
agronomy.
5. Design multi-season crop response trials and establish water quality monitoring in the Sondu-Miriu Basin
to evaluate nitrate and phosphate leaching and refine application timing.
REFERENCES
1. Sanchez, P.A. (2015). En route to plentiful food production in Africa. Nature Plants, 1(1), 1-2.
2. Tittonell, P. & Giller, K.E. (2013). When yield gaps are poverty traps. Field Crops Research, 143, 76-90.
3. Otwori, J. et al. (2024). Soil fertility assessment in smallholder farms of Nyamira County. African Journal
of Agricultural Science.
4. Kisinyo, P.O. et al. (2014). Soil acidity constraints to crop production in western Kenya. Journal of
Agricultural Science, 6(3), 23-35.
5. Basso, B. et al. (2016). Environmental and economic benefits of variable rate N fertilization. Science of
the Total Environment, 545, 227-235.
6. Vanlauwe, B. et al. (2011). Agronomic use efficiency of N fertilizer in maize-based systems. Plant and
Soil, 339, 35-50.
7. Mucheru-Muna, M. et al. (2014). Enhancing maize productivity in central Kenya. Experimental
Agriculture, 50(2), 250-269.
Page 823
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
8. Okalebo, J.R. et al. (2006). Fertilizer use manual. TSBF-CIAT, Nairobi.
9. Carpenter, S.R. et al. (1998). Nonpoint pollution of surface waters. Ecological Applications, 8(3), 559-
568.
10. Turcios, A.E. et al. (2021). Potassium improves water use efficiency. Journal of Agronomy and Crop
Science, 207(4), 618-630.
11. Xi, S. et al. (2023). Effect of potassium on tea yield and quality: meta-analysis. European Journal of
Agronomy, 144, 126767.
12. Muindi, E.M. et al. (2015). Lime-Al-P interactions in acid soils of Kenya Highlands. American Journal
of Experimental Agriculture, 9(4).
13. Kour, J. et al. (2023). Calcium's multifaceted functions in plant stress. South African Journal of Botany,
152, 247-263.
14. Asiedu, D.D. & Miedaner, T. (2025). Genomic tools for Fusarium stalk rot resistance in maize. Plants,
14(5), 819.
15. Ishfaq, M. et al. (2022). Physiological essence of magnesium in plants. Frontiers in Plant Science, 13,
802274.
16. [16] Ngetich, F.K. et al. (2023). Nitrogen dynamics and maize response to nitrate fertilization. Nutrient
Cycling in Agroecosystems.
17. Mutua, J. et al. (2022). Phosphorus response of maize on low-P soils, western Kenya. Field Crops
Research.
18. Havlin, J.L. et al. (2017). Soil fertility and fertilizers (8th ed.). Pearson.
19. Nguyen, T. et al. (2021). Atomic absorption spectrometry in soil analysis. Analytical Methods in
Agriculture.
20. Kenya Ministry of Agriculture (2023). Agricultural Soil Management Policy. MoALF, Nairobi.
21. FAO (2021). Voluntary Guidelines for Sustainable Soil Management. FAO, Rome.
22. Mairura, F.S. et al. (2021). Farm factors influencing soil fertility management in Upper Eastern Kenya.
Environmental Challenges, 100409.
23. MoALF (2019). Agricultural Sector Transformation and Growth Strategy 2019-2029. Government of
Kenya.
24. Tamene, L. et al. (2017). Soil fertility management and crop response in Ethiopia. CIAT Publication No.
449.
25. Aliyu, K.T. et al. (2021). Nutrient imbalances in maize using the DRIS approach. Scientific Reports,
11(1), 16018.