1. The study investigated the effect of soil fertility levels on water productivity of maize and potato grown in the Abay Basin in Ethiopia using the AquaCrop model. 2. Simulations showed that soil fertility was the main factor limiting productivity, and biomass and grain yields increased significantly with improved soil fertility. 3. Increasing soil fertility from poor to non-limiting conditions could more than triple maize biomass yields and increase water productivity. Harvesting excess rainfall could allow growing a second crop.

1. FU Berlin
Soil F
Fertility Effect on Water Productivity of Maize and Potato in Abay Basin
y Eff W y f y
Teklu Erkossa and S B. Awulachew
S. B
International Water Management Institute
Introduction: Ab b i (E hi i part of Bl Nil ) i characterized b a mixed crop‐
I t d ti Abay basin (Ethiopian f Blue Nile) is h i d by i d Water Balance: SF levels affected water balance components, except
Water Balance: SF l l ff t d t b l t t
livestock i f d farming system (Fi 1) A i lt l productivity i l
li t k rainfed f i t (Fig. 1). Agricultural d ti it is low d due t hi h
to: high runoff and infiltration. Of the 1450 mm rainfall received during the
runoff and infiltration Of the 1450 mm rainfall received during the
growing period, 60% infiltrates while the rest is lost as runoff. Of the
i i d 60% i fil hil h i l ff Of h
temporal and spatial variation in climate sever land degradation lack of appropriate
climate, degradation,
infiltrated water, 276, 311 and 304 mm percolates under poor, near
infiltrated water 276 311 and 304 mm percolates under poor near
technologies,
technologies poor infrastructure & limited extension services Although rainfall is adequate
services. adequate, optimal and non‐limiting SF conditions, respectively. The balance is either
ti l d li iti SF diti ti l Th b l i ith
water productivity remains low due to poor crop, soil and water management practices. Soil
p y p p, g p used as transpiration (T) or lost as evaporation (E) Improving SF
used as transpiration (T) or lost as evaporation (E). Improving SF
fertility is
f tilit i a prime f t li iti crop water productivity i areas where rainfall i not
i factor limiting t d ti it in h i f ll is t decreased E and i
d d E d increased T (b
d T (beneficial consumption) (Fi 2) Th
fi i l ti ) (Fig. 2). The
limiting (Drechsel et al 2004) This study investigated the effect of soil fertility (SF) levels on
al, 2004). reduction in E is due to better canopy cover under improved fertility,
reduction in E is due to better canopy cover under improved fertility
which attained its maximum in August. This corroborates Cooper et al.
hi h tt i d it i i A t Thi b t C t l
water productivity of maize and potato grown in sequence
sequence.
(1987) who suggested that application of fertilizers enhance crops water
(1987) who suggested that application of fertilizers enhance crops’ water
use efficiency.
ffi i
Max. soil evaporation Actual soil evaporation Max. crop transpiration Actual crop transpiration
Methodology: The FAO AquaCrop model Version 140
3 (R
(Raes et al., 2009 St d t et al., 2009) was used t
t l 2009; Steduto t l d to 120
simulate the grain and biomass productivity of maize 100
A ou (m )
m unt mm
as well as th water b l
ll the t balance of th system. I i ti
f the t Irrigation 80
requirement and water productivity of potato 60
planted f ll i maize was estimated assuming th
l t d following i ti t d i the 40
excess water during the rainy season is harvested for 20
use as supplementary irrigation Climate soil and
irrigation. Climate, 0
r
r
r
er
er
er
t
t
t
ly
ly
ly
ne
ne
ne
be
be
be
us
us
us
Ju
Ju
Ju
ob
ob
ob
crop date obtained from the basin master plan and
Ju
Ju
Ju
m
m
m
ug
ug
ug
ct
ct
ct
te
te
te
A
A
A
O
O
O
ep
ep
ep
S
S
S
research centers located within the basin were input
for model validation and simulation. The Reference Soil fertility Non- li iti
N limiting
Poor Near optimal
N ti l
Evapotranspiration was estimated using the ETo status:
t t
Calculator [Raes et al., 2009]. The Relative Root
[ , ] Fig. 2. Effect of soil fertility status on Evaporation and Transpiration
Fig 2 Effect of soil fertility status on Evaporation and Transpiration
Mean Square Error (RRMSE) Equ 1 and Coefficient of
Equ.1
Fig.
Fig 1 :The major farming systems in
Efficiency (E) Equ.2 were used to examine the Abbay Basin
y ( ) q Water Harvesting: The runoff water can be harvested for use as
robustness of the model
model, supplementary i i i of second crops. P
l irrigation f d Potato planted 10 d
l d days after
f
maize is harvested in October gives 4 8 and 10 t ha‐1 (dry weight) tuber
4,
∑ (O )2
N
1/ 2 − E yield under poor, near optimal and non‐limiting SF respectively (T bl 2)
i ld d i l d li i i SF, i l (Table 2).
⎛ 1 n ⎛ E −O ⎞ 2⎞ i i
E = 1 − i=1
Assuming 65% irrigation efficiency growing potato under near optimal
efficiency,
RRMSE
RRM E = ⎜ ∑⎜ ⎟ ⎟ Equ.1 and
q 2
Equ.2
q
⎜ N i =1⎝ O ⎠ ⎟ N
⎛ _ ⎞
⎝ ⎠ ∑ ⎜ O i− O ⎟ and well‐fertilized conditions requires about 696 and 739 mm i i ti
d ll f tili d diti i b t d irrigation
⎝
i=1
⎠ water , which is about 17 and 25% higher than the potentially harvestable
water. C
t Consequently, either l
tl ith less l d area should b used or crops with
land h ld be d ith
Three soil fertility scenarios were considered:
Three soil fertility scenarios were considered:
less water requirement should be chosen. However, increasing irrigation
1. Poor‐ representing the traditional no fertilizer use
p g
efficiency t 80% can permit growing potato on th same area of l d as
ffi i to it i t t the f land
2.
2 Near optimal‐ representing use recommended rates of N and P
maize.
3. Non‐limiting‐ representing use of recommended rates of N and P + other macro and
l g p g f d d f d h d
micronutrients Table 2. Productivity and Water Requirement of Potato grown after Maize
Table 2 Productivity and Water Requirement of Potato grown after Maize
Productivity (ton ha-1) NIR (mm)
( ) GIR ( 65% IE)
(at )
Soil fertility status (mm)
Results: The RRMSE percentage was low (8.1%) and the model efficiency (E) was close to
R l p g ( %) y( ) Dry biomass
y Yield
+ 1 indicating that the model is robust enough to predict maize productivity in the area
1, area. Poor 5.1
51 3.8
38 414 637
Near optimal 10.5
10 5 7.9
79 452 696
Non limiting
N li iti 13.1
13 1 9.8
98 480 739
Simulated Productivity and Water Balance
Simulated Productivity and Water Balance
Productivity of maize: Moisture availability was not limiting, but soil fertility was the
Productivity of maize: Moisture availability was not limiting, but soil fertility was the Conclusion: A C
AquaCrop can reasonably predict th productivity of
bl di t the d ti it f
main source of variation in productivity. About 39%, 75% and 100% of the potentially
i f i ti i d ti it Ab t 39% 75% d 100% f th t ti ll maize grown in Abbay river basin Soil fertility and the use of high yielding
basin.
achievable biomass yield can be obtained under the poor, near optimal and non limiting soil
achievable biomass yield can be obtained under the poor near optimal and non‐limiting soil varieties can significantly i
i ti i ifi tl increase water productivity of maize and potato
t d ti it f i d t t
fertility conditions, respectively while the grain yield ranges between 2.5 t ha‐1 and 9.2 t ha‐1
f tilit diti ti l hil th i i ld b t 25th 1 d92th grown in the area. The increased water productivity of maize is achieved
under poor and non‐limiting soil fertility conditions, with a corresponding increase in
under poor and non limiting soil fertility conditions, with a corresponding increase in mainly b reducing evaporation and i
i l by d i ti d increasing t
i transpiration. H
i ti Harvesting
ti
biomass and grain water productivity (Table 1). Assuming only 50% of the 3.03 million ha of
biomass and grain ater prod cti it (Table 1) Ass ming onl 50% of the 3 03 million ha of the excess water during the main can allow the growing of a second crop
Nitisols in maize based farming system in the basin is planted to hybrid maize, about 10
Nitisols in maize based farming system in the basin is planted to hybrid maize, about 10 with supplemental irrigation
irrigation.
million ton and 14 million ton grain can be obtained with near optimal and non‐limiting soil
million ton and 14 million ton grain can be obtained with near optimal and non limiting soil
fertility conditions. That is, only row planting of hybrid seeds and applying recommended
fertility conditions. That is, only row planting of hybrid seeds and applying recommended Reference:
rates of nitrogen and phosphorus as practiced by research centres can increase productivity
rates of nitrogen and phosphorus as practiced by research centres can increase productivity Drechsel, P., Giordano, M. and Gyiele, L. 2004. Valuing nutrition in soil and
h l d d y l l g l d
by t ee o d as co pa ed to t e cu e t
by three fold as compared to the current. water: concepts and techniques with examples from IWMI studies in
developing world. IWMI R
d l i ld Research P
h Paper 82 I82. International W
i l Water
Table 1. Water productivity of maize as affected by soil fertility status
T bl 1 W d i i f i ff db il f ili Management Institute (IWMI) Colombo Sri Lanka
(IWMI), Colombo,
Cooper P J M G
C P. J. M., Gregory. P J K i
P. J., Keatinge, J D H and B
J. D. H. d Brown S C 1987
S. C.1987.
Yield (t ha-1) % Biomass in reference to Water productivity (kg m-3) Effects of fertilizer variety and location on barley production under
fertilizer,
Soil fertility rainfed conditions i northern S i Fi ld C
i f d diti in th Syria. Field Crops R 16 67 84
Res. 16, 67–84
Biomass Grain well watered well fertilized Biomass Grain Raes, D., Steduto, P., Hsiao,
Raes D Steduto P Hsiao Th and Fereres E 2009 AquaCrop—The FAO
Fereres, E. 2009. AquaCrop The
Poor 7.5 2.5 100 39 5.1 1.7 crop model f predicting yield response t water: II M i algorithms and
d l for di ti i ld to t II. Main l ith d
Near
N optimal
i l 14.3
14 3 6.4
64 100 75 5.3
53 2.4
24 soft ware description Agron J 101:438–447
description. Agron. J. 101:438 447
Steduto, P., Hsiao, T.C., Raes, D., Fereres, E., 2009. AquaCrop—the FAO
St d t P H i T C R D F E 2009 A C th FAO
Non limiting 19.2
19 2 9.2
92 100 100 5.4
54 2.6
26
crop model to simulate yield response to water. I. Concepts. J. Agron. 101,
crop model to simulate yield response to water. I. Concepts. J. Agron. 101,
426–437
426 437
For
F more i f
information contact: (
ti t t (e-mail)
il)
Adress:
d
t.erskossa@cgiar.org
http://www.iwmi.cgiar.org/africa/east_africa/
p g g _