The Journal of Agricultural Inoculant, nitrogen and phosphorus improves 12
Science photosynthesis and water-use efficiency in 3
4
cambridge.org/ags soybean production 5
6
7
C. E. N. Savala1 , A. N. Wiredu1 , J. O. Okoth2 and S. Kyei-Boahen1 8
9
Crops and Soils Research 1International Institute of Tropical Agriculture, P.O. Box 709, Nampula, Mozambique and 2International Institute of 10
Paper Tropical Agriculture, P.O. Box 30258, Lilongwe 3, Malawi 11
12
Cite this article: Savala CEN, Wiredu AN, Abstract 13
Okoth JO, Kyei-Boahen S (2021). Inoculant,
nitrogen and phosphorus improves Soybean yield within the Southern Africa falls below its potential despite similar climatic condi-
14
photosynthesis and water-use efficiency in tions across some agroecologies, replicable agronomic management practices and introduced
15
soybean production. The Journal of improved varieties. Understanding physiological processes and water-use efficiency (WUE) of 16
Agricultural Science 1–14. https://doi.org/ soybean offer information on bridging this yield gap. A field study was conducted in 2017 and 17
10.1017/S0021859621000617 2018 seasons in two agroecologies (Angonia and Ruace) in Mozambique to evaluate the effects 18
Received: 23 April 2021 ofBradyrhizobiumdiazoefficiens strainUSDA110 formerly known asBradyrhizobium japonicum
19
Revised: 28 June 2021 inoculant, nitrogen and phosphorus on nodulation, physiology and yield of non-promiscuous 20
Accepted: 6 July 2021 (Safari) and promiscuous (TGx 1740-2F) soybean varieties. Data on transpiration, photosyn- 21
Keywords: thesis, leaf area index, radiation interception and WUE from the beginning of flowering to
22
Agroecologies; climate change; determinate; maturity were collected. Transpiration rate varied considerably with interaction between
23
yield locations, growth stages, varieties and treatments. At podding, phosphorus-treated soybean at 24
Angonia transpired less (6.3 mmol/m2/s) than check plants (6.6 mmol/m2/s). Photosynthesis 25
Author for correspondence: rate and WUE were distinct with variety, growth stages and inputs within agroecologies. For 26
C. E. N. Savala, E-mail: C.Engoke@cgiar.org
instance, in Angonia 2018 season, phosphorus fertilized TGx 1740-2F photosynthesized more 27
at flowering (25.3 μmol/m2/s) while the lowest was phosphorus-treated Safari at podding 28
with 17.2 μmol/m2/s. At the same site in 2017, inoculated soybean photosynthesized more at 29
22.8 μmol/m2/s leading to better WUE of 3.6 that corresponded to 2894 kg/ha yield. Overall, 30
soybean WUE was higher when inoculated than N-treated, while P application yielded better. 31
Results from this study will complement breeders’ effort in developing phosphorus efficient 32
varieties suited for a wide range of changing climatical conditions. 33
34
35
36
37
Introduction
38
Introduction of new improved soybean (Glycine max L. Merrill) genotypes to smallholder 39
farmers in Southern Africa region aims at increasing productivity. However, yield continue 40
to fall below the varieties production potentials due to climate change regardless of continuous 41
expansion of cultivated land areas, introduction of improved varieties and use of plant nutri- 42
ents inputs for soybean production (Khojely et al., 2018). Vicissitudes in the regional climatic 43
patterns is threatening future production of soybean. Recently, cropping-seasons have been 44
characterized by frequent events of irregular rainfall and drought both mid- and/or end-season 45
(Ngcamu and Chari, 2020) that warrants better water-use efficiency (WUE) for yield improve- 46
ment. Technologies for combating irregular rainfall patterns and drought are numerous, chief 47
being water harnessing in reservoirs (during heavy rains) and irrigation. However, smallholder 48
farmers in sub-Saharan Africa (SSA) rarely adopt these due to a plethora of challenges. 49
Instead, they rely heavily on rainfed production of crops. Despite irrigation being key, other 50
agronomic technologies such as: adjusting or changing planting dates, crop varieties, plant 51
density, nutrient and soil moisture management practices can suffice to improve yield. 52
Judicious management of soil moisture in this changing climate also entails effective and effi- 53
cient water use by crops. It is projected that soybean production in eastern part of Africa spanning 54
from Ethiopia, Kenya, Malawi, Mozambique, Rwanda, Tanzania, Uganda and Zambia could 55
© The Author(s), 2021. This is an Open Access decline by 12% by 2090 due to water stress if climate change is not mitigated (Tatsumi et al., 56
article, distributed under the terms of the 2011). The interaction between water supply, temperature and other environmental conditions 57
Creative Commons Attribution licence (http://
creativecommons.org/licenses/by/4.0/), which affects the relationship between WUE and yield which is not necessarily linear and differ with 58
permits unrestricted re-use, distribution, and crop species (Condon et al., 2002; Irmak and Specht, 2014). Although some studies have reported 59
reproduction in any medium, provided the WUE impacting yield, it depends on the prevailing abiotic conditions at each growth stage and 60
original work is properly cited. the affected plant physiological process. For instance, restricting maximum transpiration rate 61
improved WUE leading to 9–13% increase in sorghum yield (Sinclair et al., 2005). Greater 62
yield losses occur whenWUE is low due towater stress or higher losses coinciding with the repro- 63
ductive stage especially between flowering and full seed in soybean. Anda et al. (2018) reported a 64
yield loss in two soybean varieties, one water stress-tolerant and the other bred for ‘average’ 65
2 C. E. N. Savala et al.
weather conditions due to low WUE in both irrigated and rainfed Nitrogen is a key element in photosynthesis process as it affects 66
systems. When water was supplied at 50% of the requirement at leaf chloroplasts size, number and composition (Bassi et al., 67
vegetative stages, the leaf area index (LAI) reduced hence affecting 2018). Inoculation promotes BNF process that avails more N for 68
bloom and pod-filling of the soybean. Other researchers have plant uptake resulting in better leaf development creating a larger 69
worked on themechanism of plant response towater and heat stress surface area that maximizes the photosynthesis process in plants 70
in relation to WUE. The results from barley, soybean and wheat hence increasing N use efficiency. Therefore, N deficiency in soils 71
have shown that response to water and heat stress which impact could lead to reduced photosynthesis process that affect soybean 72
on WUE is genotypically distinct and can be modified through growth and development (Uprety and Mahalaxmi, 2000). 73
gene manipulation (Anyia et al., 2007; Hufstetler et al., 2007; High yielding drought tolerant soybean genotypes that use 74
Siahpoosh et al., 2011; Rizza et al., 2012). water and nutrient efficiently are required to mitigate effects of cli- 75
Transpiration is important in mass-flow movement of plant mate change. Development and dissemination of drought tolerant 76
nutrients as well as cooling the plant, and some of the elements and specific agroecologically adapted varieties could also help min- 77
are key in photosynthesis. These two processes; transpiration and imize the climate change impact in drought-prone areas (Adhikari 78
photosynthesis, have a bearing on plant WUE. A ratio between et al., 2015). Agronomists and producers use some technologies 79
photosynthesis and transpiration is used to estimate WUE such as nutrient amendments, inoculation, planting time alteration 80
(Medrano et al., 2015). The processes take place in active leaves and plant population that impact on legume response to tempera- 81
depending on atmospheric temperature and soil water content. If ture and available water. However, smallholder farmers in SSA 82
either or both are limiting at any plant development stage, then hardly deploy a fusion of these agronomic technologies in produc- 83
fewer and/or smaller organs are formed impacting on transpir- tion of legumes such as soybean, due to various reasons, example 84
ation, photosynthesis rate and yield. Higher temperatures result resource incapability. In this study, we seek to determine how soy- 85
in increased transpiration to a certain threshold before the stomata bean WUE and production is affected by inoculant, nitrogen and 86
closes. Plants close their stomata to reduce transpiration which phosphorus application. The objectives were to evaluate the effect 87
impedes photosynthesis rate and might cause heat-damage of nutrient application on soybean WUE and yield through evalu- 88
(Lobell and Gourdji, 2012). On the other hand, moisture stress ation of nodulation, physiological processes at three reproductive 89
reduces crop reproductive stage, leaf area and enhances stomatal stages, growth and yield components. 90
closure to minimize water loss. These plant adaptation characteris- 91
tics have a negative effect on crop yield (Thornton et al., 2009). 92
Studies have demonstrated that both water stress and increased Materials and methods 93
heat decrease the length of the growing season and loss of product- Site description 94
ive arable land especially along boundaries of arid and semiarid 95
regions in Africa (Conway, 2009). With the current harsh weather Field studies were conducted in 2017 and 2018 growing seasons 96
conditions resulting from climate change, plant WUE will deter- at two high potential soybean growing areas in Mozambique′ ′ 97
mine its capability to adapt to specific agroecologies through (Angonia 14 ° 32 S, 34 ° 11 E, 1202 metres above sea level′ ′ 98
alterations in the phenological and physiological characteristics. (m a.s.l.) and Ruace 15 ° 21 S, 36 ° 47 E, 772 m a.s.l.). In each sea- 99
Physiological processes within plant tissue that respond to climate son, fields under maize for previous two growing periods were used. 100
change are transpiration, carbon assimilation and photosynthesis According to the Soils Atlas of Africa, the predominant soil type at 101
affecting WUE hence overall yield. For instance, in drier environ- the sites in Angonia and Ruace are Chromic Luvisols (Jones et al., 102
ments, a high WUE increases yield unlike in wetter environments 2013). Six soil samples were taken from 0 to 20 cm soil layer 103
where it leads to proliferation of vegetative parts at the expense of using a soil auger in a W pattern across the field for the trial before 104
reproductive organs. ploughing or harrowing. Soils from each site were combined into 105
Proper utilization of soil and water enhances the effectiveness of a composite sample and four subsamples drawn for chemical and 106
available moisture, fertilizer and improved seeds in crop production. particle-size analysis (Table 1). The pH was determined using a 107
There are organizations involved in the development of soybean high impedance voltmeter on 1:2 soil–water suspension. Total 108
improved seed and accompanying agronomic practices such as organic carbon was determined by Walkley–Black Method, total 109
inoculants, fertilizer regimes especially phosphorus and agroecolo- N by The Kjeldahl method, P by Olsen’s method and K using 110
gically suitable planting times. For instance, inoculation and P fer- ICP-OES after extraction with Mehlich 3. 111
tilizer application increased cowpea grain yield and above-ground 112
plant dry matter at harvest in Mozambique (Kyei-Boahen et al., 113
Experimental layout
2017) and soybean in locations of Kenya (Majengo et al., 2011), 114
Ghana (Akpalu, 2014), Malawi (Phiri et al., 2016) and Rwanda Two early maturity group soybean varieties: TGx 1740-2F (pro- 115
(Rurangwa et al., 2018) among others. Majengo et al. (2011) showed miscuous) locally known as ‘Wamini’ released by the Instituto 116
that inoculation of promiscuous medium-maturity soybean variety de Investigação Agrária de Moçambique (IIAM) (National 117
TGx1740-2F increased drymatter by 27% and average grain yield by Agricultural Research Institute of Mozambique) and Safari (non- 118
15–30% in Bungoma Kenya. These studies focused on the growth promiscuous Seed Co. variety) were used in the experiment. 119
and development characteristics that led to the yield observed fol- Commercial Bradyrhizobium diazoefficiens strain USDA110 for- 120
lowing N and P application but not the physiological, phenological merly known as Bradyrhizobium japonicum (Delamuta et al., 121
and WUE changes due to agronomic practices. Inoculants contain 2013) inoculant product NitroZam was obtained from Zambia 122
effective rhizobia that enhance symbiotic biological nitrogen Agricultural Research Institute (ZARI), Lusaka, Zambia for the 123
fixation (BNF) in soybean (Sanginga et al., 2002; Zimmer et al., study. The experimental design was randomized complete block 124
2016; Savala and Kyei-Boahen, 2020). Similar to P, supply of N with four replications. A factorial arrangement of treatments 125
either through BNF or mineral fertilizer promote plant growth consisted of no inputs (check), inoculation with NitroZam, split 126
that stimulates increased photosynthesis activity in plant leaves. application of 100 kg N/ha as Urea (50 kg N/ha each at planting 127
The Journal of Agricultural Science 3
Table 1. Amount of rainfall for the growth stages of Safari and Wamini soybean varieties recorded in the 2017 and 2018 cropping seasons at Angonia and Ruace 128
129
Angonia rainfall amount (mm) Ruace rainfall amount (mm)
130
Growth stage Safari Wamini Mean Safari Wamini Mean 131
132
Cropping season 2017 133
134
Vegetative 422.2 422.2 422.2 271.5 271.5 271.5
135
Flowering 359.2 359.2 359.2 28.2 28.2 28.2 136
Podding 173.8 173.8 173.8 93.9 55.5 74.7 137
138
Seeding 173.6 173.6 173.6 231.5 231.2 231.4
139
Total 1128.8 1128.8 1128.8 625.1 586.4 605.8 140
Cropping season 2018 141
142
Vegetative 218.9 203.9 211.4 553.0 553.0 553.0
143
Flowering 185.2 215.2 200.2 158.4 48.9 103.6 144
Podding 181.4 175.3 178.3 197.5 251.6 224.5 145
146
Seeding 22.6 13.8 18.2 233.8 272.2 253.0
147
Total 608.1 608.1 608.1 1142.7 1125.7 1134.2 148
149
150
and flowering), application of 40 kg P2O5/ha as single superphos- was determined. Later the biomass was ground to pass a 2-mm 151
phate, and inoculant plus 40 kg P2O5/ha applied together at mesh sieve for plant tissue N, P and K analysis using 152
planting. Plots consisted of seven rows of 8 m in length, 0.50 m Inductively coupled plasma – optical emission spectrometry 153
row-spacing and 0.1 m between plants within rows. Land prepar- (ICP-OES) after extraction with nitric acid and hydrochloric 154
ation was accomplished with disc ploughing followed by two acid. Photosynthesis and transpiration using LCpro+ photosyn- 155
passes with a disc harrow. Both seasons’ experiments were planted thesis system were determined in Angonia and Ruace at flowering, 156
between 12 and 18 December depending on the onset of rains in podding and grain filling. At each growth stage, the photosyn- 157
each location. thesis system was used to take measurements between 09:00 158
To avoid contamination, the non-inoculated and fertilizer and 12:00 h, on one third recently mature fully expanded leaf 159
receiving plots were planted first before the inoculated ones. on three plants in each plot. At maturity, an area of 1 m2 (two 160
Inoculation was done in the field under a shade by weighing 1 m long from the middle rows) was randomly selected and 161
1.0 kg of seeds of each cultivar into separate plastic bags and add- harvested for determination of above ground biomass, pod dens- 162
ing 5 ml of 3% (w/v) gum arabic solution as sticker. The seeds and ity and seed weight. Pods from each plot were threshed manually 163
gum arabic solution were mixed thoroughly and 10 g of the peat- and grain yield was determined. The moisture content of grain 164
based inoculant (according to the manufacturer’s recommendation) samples from each plot was measured using Farmex MT-16 165
was applied to the seeds in each bag and mixed thoroughly until all grain moisture Tester (AgraTronix LLC, Streetsboro, Ohio, 166
the seeds were completely covered with inoculant. The inoculant USA) and grain yield in kg/ha was adjusted to 13% moisture con- 167
was applied to supply approximately 106 rhizobia cells/seed. The tent. After harvesting the pods, the above-ground plant biomass 168
seeds were planted immediately. Planting and weeding (twice) from the 1 m2 plot area were sun-dried to 10% moisture content 169
were done manually, although scheduling was site specific. The for 10 days to determined dry matter yield. Except for land 170
experiment was conducted under rainfed conditions for both sea- preparation, all field activities were done manually. 171
sons (Fig. 1). To control pests, 100ml of Cypermethrin (200 g 172
active ingredient/l) and 50ml of Lambda Cyhalothrin (50 g active 173
Data analysis
ingredient/l) were used in 15 l once at beginning of flowering. 174
Analyses of variance (ANOVAs) were performed using PROC 175
GLM in Statistical Analysis System (SAS)® 9.4 (SAS Institute, 176
Data collection
2018). A combined analysis across locations and cropping seasons 177
Data on growth, physiological, phenological characteristics and was performed. Since location and season effects were dominant, 178
plant tissue content were collected. At R3 (flowering to podding) the variables were combined to form environment. Firstly, a 179
growth stage before second N dose was applied and when pods factorial ANOVA was performed, to evaluate the effects of envir- 180
had reached 10–12 mm long at one of the four uppermost onment, variety, treatment and their interactions. Environments 181
nodes on main stem, five soybean plants were randomly selected effects were significant for all the variables. Secondly, a two-way 182
from each plot and all the roots were excavated using a hoe and ANOVA of varieties and treatments factors considered as fixed 183
spade. The soil was carefully washed out from the roots to ensure effects was performed for each environment. Environment and 184
that all the nodules were recovered as much as possible. The replication were treated as random effects (Moore and Dixon, 185
nodules were plucked from the roots by hands, counted and 2015). Means were determined for treatments, and comparisons 186
subsequently placed in envelopes then dried in an oven at 60 °C done using Tukey adjustment at P≤ 0.05 significance level 187
for 48 h to determine nodule dry weight. The remaining plant based on the standard error of means (S.E.M.) (SAS Institute, 188
biomass were dried in an oven at 60 °C for 72 h and dry weight 2018). Plant growth variables analysed included nodulation, 189
4 C. E. N. Savala et al.
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
Fig. 1. Colour online. Rainfall and atmospheric temperature during the experiment in (a) Angonia and (b) Ruace. 209
210
211
shoot biomass, N, P and K contents at R3 stage, number of pods/m2, filling stage was 6.8 and 3.8 mmol/m2/s in Angonia and Ruace, 212
number of seed per pod, 100-seed weight, grain yield and plant dry respectively. In the second season, soybean at Angonia transpired 213
matter at harvest. Also, data on days to flowering (R2), podding at 5.7 v. 4.6mmol/m2/s for Ruace between flowering and grain 214
(R3) and maturity (R8) were collected (data not reported). filling stage. When averaged across environments, varieties and 215
Photosynthesis, transpiration, WUE, photoactive radiation inter- inputs, transpiration increased from flowering stage 4.9 to 5.4 216
cepted (PAR) and LAI between flowering and full seed was also mmol/m2/s at podding before declining to 5.0mmol/m2/s during 217
analysed. The ecologies of the locations were different and domin- grain filling stage (Fig. 2) although different trends were observed 218
ant for the variables analysed, thus the effects associated with the at the two locations. At Angonia in 2018, transpiration increased 219
seasons were confounded with the weather within seasons (Moore from flowering to podding then considerably declined towards 220
and Dixon, 2015) making seasons a random effect for the purpose grain filling stage. At all the three growth stages, differences in 221
of estimating variability of treatment differences for the locations transpiration rate were observed either numerically or statistically 222
across seasons. Therefore, location and treatment were considered between the inputs in both locations. In contrast to Angonia 223
fixed effects, whereas cropping season and blocks nested within where a drop in temperature led to a reduced transpiration rate, 224
locations were considered as random effects. at Ruace, the rate and amount of water loss through leaves continu- 225
ally increased from flowering to grain filling stage (Fig. 2). The rate 226
at which soybean was losing water through the leaves at Ruace was 227
Results
significantly higher for the check than all the other inputs at pod- 228
Development of soybean was distinct between the two seasons, ding while at Angonia, between podding (6.9 mmol/m2/s) and 229
sites, varieties and treatments in relation to the weather conditions. grain filling stage (4.2 mmol/m2/s), the magnitude of change in 230
At Ruace, rainfall was high in 2018 notably increasing in frequency transpiration rate indicated by the slope was steep for check unlike 231
and amount towards the end of the season (Table 1) leading to for inoculant, N and P treatments. 232
better yield unlike in 2017. Temperature was comparable between Photosynthesis was distinct at flowering but not podding and 233
the sites but declined towards the end of the seasons in Angonia grain filling stage of the varieties within the environments (Fig. 3). 234
creating favourable conditions for soybean rust (Fig. 1). In 2018, Photosynthesis of both varieties declined at Angonia and Ruace 235
there was a late season attack of soybean rust which led to lower from flowering to podding before slightly increasing as the grains 236
yield than the 2017 season for both varieties in Angonia. filled. Varietal photosynthesis rate was statistically distinct at flow- 237
However, three steps were taken to combat the infestation in our ering and numerically different at podding stage in Angonia. 238
test plots. First a contact fungicide Bravo (Chlorothalonil 500 g/l) Soybean at Angonia photosynthesized more than at Ruace at 239
‘weatherstik’ was sprayed upon early detection. Then 7–14 days flowering and grain filling for both varieties. The relationship 240
later a systemic fungicide Trister EC (triadimenol/trizoles 250 g/l) between photosynthesis and transpiration was used to determine 241
was applied before repeating the contact fungicide after 14 days. WUE (Medrano et al., 2015). WUE is a measure of plant dry 242
These fungicides reduced the severity of the rust damage in our matter or yield production per unit of water used. The WUE 243
experimental plots. was dissimilar at different stages of plant development at both 244
locations (Figs 4a and b). The WUE was not significantly different 245
within growth stages between inputs. A declining trend of WUE 246
Physiological processes changes
was observed between flowering and podding stages in both loca- 247
Transpiration, photosynthesis and WUE tions, which later increased from podding to grain filling stage at 248
Transpiration, photosynthesis and WUE were distinct by variety, Angonia unlike at Ruace where it continued on a reducing trajec- 249
growth stage and input within each environment (Figs 2–5). In tory. However, at grain filling stages, there was better water utiliza- 250
2017 season, mean transpiration rate between podding and grain tion by soybean following application of inputs that contained 251
Fig. 1 - Colour online, B/W in print
The Journal of Agricultural Science 5
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
Fig. 2. Colour online. Transpiration rate at growth 283
stages of soybean in Angonia (a) and Ruace (b) at P
< 0.05 significance level between the treatments at 284
each growth stage. The error bars are ± S.E.M. 285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
Fig. 3. Colour online. Photosynthesis rate of two soy- 300
bean varieties at three growth stages Angonia (dotted 301
lines) and Ruace (solid lines). Significance level at P <
302
0.05 between the varieties at each growth stage. The
error bars are ± S.E.M. 303
304
305
306
inoculant at both locations. At Angonia the WUE increased from at the three stages of growth (Figs 5a and b). Transpiration increased 307
podding to grain filling stage mainly due to reduced transpiration at Angonia from flowering to podding then decreased towards grain 308
that increased at Ruace (Figs 4a and b). The increase in transpiration filling while, the WUE’s response was inverse between the growth 309
at Ruace, on the other hand, led to a reduction in WUE. For both stages for both varieties Safari and TGx 1740-2F. On the contrary, 310
locations, there was an inverse relationship between transpiration photosynthesis had similar response as WUE, although the rate of 311
and photosynthesis that affected the WUE. Similarly, the relation- change between growth stages were distinct when averaged across 312
ship between transpiration and WUE was inverse for both varieties varieties and inputs. For instance, the magnitude of increase or 313
Fig. 3 - Colour online, B/W in print Fig. 2 - Colour online, B/W in print
6 C. E. N. Savala et al.
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
Fig. 4. Colour online. Influence of nutrient amend-
348
ments on WUE of soybean at different growth stages
in (a) Angonia and (b) Ruace at P < 0.05 significance 349
level between the treatments at each growth stage. 350
The error bars are ± S.E.M. 351
352
353
decrease in photosynthesis was different between growth stages. and grain filling stage, LAI of P+-inoculant treatment declined at
354
This signifies that if the magnitude of an increase in photosynthesis a faster rate too (Fig. 6c). The LAI was significantly lower for the
355
is higher than the change in transpiration, then WUE will poten- check than the other treatments at grain filling stage in Angonia.
356
tially improve. At the same site, inoculated soybean LAI insignificantly changed
357
from the phenological stage of flowering to grain filling stage.
358
Photosynthetically active radiation and LAI Soybean that received inoculant and N at Angonia resulted in a 359
The PAR values declined in Angonia as the plant developed with higher LAI at grain filling stage. The rate at which the LAI increased 360
significant inputs differences between flowering and grain filling between flowering and podding before declining towards grain 361
stage (Fig. 6a). While at Ruace PAR values slightly increased to pod- filling stage at Ruace was different (Fig. 6d). Notably at this site, 362
ding before declining at grain filling stagewith numerical differences P+-inoculant resulted in a denser canopy at grain filling stage. 363
observed between inputs and growth stages (Fig. 6b). The percent The leaf canopy LAI results suggest that perhaps soybean develop- 364
PAR values representing the light intercepted by soybean that ment at Angonia is more restricted by the availability of N while at 365
received P fertilizer at podding stage were significant from the Ruace both P+inoculant is required to improve growth and product- 366
other inputs while those with N and inoculant were only numeric- ivity through increased leaf surface area that captures more light 367
ally different from the check at Angonia. At this location, percent necessary for the photosynthesis process. 368
PAR intercepted for the inputs decreased from flowering to grain 369
filling stage at different magnitudes. For instance, the rate of 370
Biomass quality at R3 stage
decrease in percent PAR intercepted between podding and grain fill- 371
ing stage by P+-inoculated field was faster than all other inputs. The Besides nodule characterization at R3 stage, samples were also col- 372
change in PAR intercepted signify a decrease in the amount of leaves lected from both sites Angonia and Ruace to ascertain the quality 373
developed or change in the leaf angle orientation as the soybean of biomass through determination of N, P and potassium (K) 374
plant matures. Similar to PAR intercepted trend between podding amount accumulated in the plant tissue. The amount of N (kg/ 375
Fig. 4 - Colour online, B/W in print
The Journal of Agricultural Science 7
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
Fig. 5. Colour online. Transpiration (dotted lines) and 411
WUE (solid lines) of Safari and TGx 1740-2F at three 412
growth stages in (a) Angonia and (b) Ruace at P < 413
0.05 significance level between the varieties at each 414
growth stage. The error bars are ± S.E.M.
415
416
417
ha) in the plant tissue at R3 stage was numerically higher than all accumulated in the tissue of TGx 1740-2F while more K was 418
the other inputs with application of N fertilizer (Fig. 7). Also, at stored in plant tissues of Safari at Angonia. 419
this stage for all treatments except P, the amount of plant tissue 420
N content was significantly higher for TGx 1740-2F than Safari. 421
Nodulation, yield components and yield
The varietal difference in the N content for P input treatment 422
was numerically higher for TGx 1740-2F than Safari but not sig- Formation of nodules in soybean requires an association of the 423
nificant. This signifies that TGx 1740-2F had accumulated more plant with the appropriate rhizobia strain. Our results showed 424
N/ha than Safari. In both varieties, the amount of N in plant tis- that inoculant and P fertilizer applied either singly or in combin- 425
sue showed that use of inoculant, N and P fertilizer was signifi- ation promoted nodulation leading to higher nodule numbers per 426
cantly higher than the check. When R3 plant tissue N is related plant than the check for the two soybean varieties (Table 3). 427
to the production (Fig. 8), Safari yielded numerically more than Inoculating soybean at Angonia led to more nodule numbers 428
TGx 1740-2F meaning that there is no direct relationship with per plant than when combined with P. The increase in the num- 429
the amount of nutrient in the plant tissue at one stage with the ber of nodules per plant upon application of P+inoculant was 430
yield. In addition, use of inputs in both varieties yielded signifi- statistically higher than inoculant or P alone for TGx 1740-2F 431
cantly higher than the check with the combination of P+inoculant at Ruace in 2018 while in 2017 significant differences were 432
having overall synergetic effect on production. The amount of between individual input rather than the combinations. On the 433
P and K in plant tissue at R3 stage was also analysed. other hand, N application seemed to depress formation of 434
Accumulation of P and K in plant tissue at R3 stage was signifi- nodules in both varieties and locations. There was no direct rela- 435
cant with variety within each site but not with treatment tionship between the number and the weight of the nodules 436
(Table 2). In both locations, there was higher N and P although a similar trend as that observed with nodule population 437
Fig. 5 - Colour online, B/W in print
8 C. E. N. Savala et al.
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
Fig. 6. Colour online. Photosynthetically active radiation (PAR) intercepted at (a) Angonia and (b) Ruace and LAI of soybean at (c) Angonia and (d) Ruace at different 487
growth stages as influenced by nutrient amendments at significance level of P < 0.05 between the treatments at each growth stage. The error bars are ± S.E.M.
488
489
was depicted with the weights except at Angonia where there were lowest seeds/pod (2.4) among the inputs (Table 4). At the same 490
heavier nodules following application of P fertilizer in 2017 location, application of P+ inoculant on Safari resulted in more 491
season (Table 3). The nodules from P treatments plots in the seeds/pod and the highest yield of 3615 kg/ha. The number of 492
two sites were relatively larger compared to the other inputs. pods/m2, seeds/pod and 100 seed weight were distinct by location 493
Data were collected on three yield components namely; and treatment and different from the check in all sites, Safari had 494
number of pods/m2, seeds/pod and 100 seed weight. These com- generally heavier seeds than the TGx 1740-2F and there exists a 495
ponents varied among inputs both within and between sites. slight relationship of the number of seeds per pods being inverse 496
Results indicate that the pods/m2 and seeds/pod increased when to 100 seed weight for Safari at Angonia (Table 4). There is no 497
inputs were used (Table 4). For instance, at Angonia 2017, appli- direct relationship between and individual yield component and 498
cation of N on Safari had the highest pods/m2 (2182) with the yield since their interaction is additive or counteractive in a 499
Fig. 6 - Colour online, B/W in print
The Journal of Agricultural Science 9
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
Fig. 7. Colour online. The effect of nutrient amend- 522
ments on N in the biomass at soybean flowering
523
stage at the two sites-Angonia and Ruace. Means are
significantly different at P < 0.05 between the treat- 524
ments and varieties. The error bars are ± S.E.M. 525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
Fig. 8. Colour online. The interaction of grain yield at
the two sites-Angonia and Ruace. Means are signifi- 547
cantly different at P < 0.05 between the treatments 548
and varieties across the seasons. The error bars are 549
± S.E.M. 550
551
552
complex manner to result into yield. An increase in one compo- season and a site with high initial P (Table 5), fields of TGx 553
nent could lead to a reduction or complimentary effect in another 1740-2F soybean that received inoculant gave higher yield than 554
during plant growth. Safari unlike the check, N and P alone that posted opposite results 555
Regardless of location, all treatments resulted in a higher yield between the varieties. Our results show that there is a potential to 556
than the check for both varieties (Fig. 8). At Angonia, Safari a deter- increase soybean yield through application of inoculant, N and P 557
minate variety had a numerically higher yield than TGx 1740-2F in over the check. The yield increase potential ranges were N (108– 558
both seasons for all treatments but not for P+-inoculant. At the 927 kg/ha), inoculant (12–1145 kg/ha), P (114–955 kg/ha) and P+ 559
same location in 2018 season, due to the rust attack, both varieties inoculant (16–1401 kg/ha). These potential ranges derived from 560
had a comparable yield for each treatment. At Ruace in the same the two season’s results for Angonia and Ruace locations in 561
Fig. 8 - Colour online, B/W in print Fig. 7 - Colour online, B/W in print
10 C. E. N. Savala et al.
Table 2. The amount of total N, P and K in biomass at R3 growth stage 562
563
Nitrogen Phosphorus Potassium Yield
Site Variety (kg N/ha) (kg P/ha) (kg K/ha) (kg/ha) 564
565
Angonia Safari 78 4.8 33 2788 566
TGx 1740-2F 100 6.4 45 2626 567
568
S.E.M. 4.7 0.36 1.9 104.1 569
Ruace Safari 104 8.3 51 2051 570
TGx 1740-2F 108 9.2 56 2059 571
572
S.E.M. 5.5 0.49 2.9 122.3 573
Means compared using S.E.M. of variety per site at P≤ 0.05 significant level between varieties for each variable per site 574
575
576
Table 3. Nutrient amendments effect on the number of nodules and weight at cowpea that were either inoculated or received P fertilizer 577
Angonia and Ruace (Kyei-Boahen et al., 2017) and soybean (Savala and 578
Angonia Ruace Angonia Ruace Kyei-Boahen, 2020). Accumulation of N and K in TGx 1740-2F 579
2017 2018 2017 2018 could be related to elevated transpiration. The transpiration rate 580
of TGx 1740-2F was higher than that of Safari, except at grain fill- 581
Nodule weight ing stage in Angonia (Fig. 5a). Transpiration-driven mass-flow 582
Input Nodule/plant (mg/plant) has a correlation with plant nutrient accumulation (Cramer 583
et al., 2008; Matimati et al., 2014; Houshmandfar et al., 2018). 584
Safari
Similar to N, P must be supplied to biological N fixing plants 585
Check 36.2 4.7 133.8 56.8 to maintain nodule tissue and provision of energy ATP required 586
Inoculant (Inoc) 56.8 20.0 395.3 238.1 for biochemical processes. As a result, roots and nodules become 587
an important P sink in N fixing plants (Jakobsen, 1985; Hart, 588
Nitrogen 19.9 3.0 90.7 36.75
1990). BNF imposes a high phosphorus demand and therefore 589
Phosphorus (Phos) 37.8 10.1 465.9 135.3 can only occur when the nutrient is sufficient in soil (Tefera 590
Phos + Inoc 55.5 22.0 431.7 247.4 et al., 2010). Supplying N and P to soybean is important for can- 591
opy development that captures more radiation for the photosyn- 592
TGx 1740-2F
thesis process. The nutrients are found in varied amounts in plant 593
Check 32.9 8.2 154.2 74.6 tissue to facilitate essential biochemical processes. Nitrogen and K 594
Inoculant (Inoc) 52.5 39.0 392.5 350.5 accumulated more in the above ground matter than P which has 595
been reported to be stored more in the roots. Results from our 596
Nitrogen 27.5 4.7 236.7 59.35 study show low P values in the above ground biomass at R3 597
Phosphorus (Phos) 49.2 24.3 508.8 270.6 stage. Perhaps the P element accumulated in the plant roots 598
Phos + Inoc 47.9 40.1 233.9 360.8 and nodules tissues (not analysed in our study) unlike in the 599
above ground biomass. Therefore, it is important to provide N, 600
S.E.M. 0.10 0.05 44.80 22.51 P and K nutrients in soybean production to improve productivity. 601
Input and variety differences within each environment compared using S.E.M. at P≤ 0.05 Inoculation and application of N and P fertilizer in soybean pro- 602
significant level between inputs and across varieties for each variable within the
603
environment (location × season). duction improves plant development and biochemical processes
that improved WUE especially during podding which leads to 604
an increase in the final yield. The key in realizing the benefits 605
Mozambique indicate increase in yield one could get when using of supplying these nutrients lies in the timing and judicious appli- 606
the different amendments in soybean production. It is evident cation of inoculants, organic and inorganic fertilizers in soybean 607
that handling of inoculant is crucial in determining the final production. Other activities that improve nutrient exploration and 608
soybean yield. There is a wide potential yield increase range for uptake by plants roots from the soil such as mychorrizal associa- 609
the treatments that contained inoculant over singly applying N or tions (not evaluated in this study) could offer additional agro- 610
P fertilizer. nomic information on input utilization. Improving soybean 611
growth characteristics and enhancing efficient N and P assimila- 612
tion through formation of mychorrizal associations that promote 613
Discussion
biochemical processes could be attained through breeding. In 614
Nutrient demand by soybean is elevated during the reproductive future, information on related mychorrizal associations that 615
stage. Above ground biomass analysis showed more N and K promote P extraction from the soil by soybean could also be eval- 616
accumulation in the plant tissue at R3 stage. Since both N and uated alongside inoculants to select for agroecologically suitable 617
K are essential elements for photosynthesis and transpiration, fungi. These improved varieties can be adapted to a wide range 618
they are expected to be in high amounts in biomass of actively of changing climatical conditions within numerous agroecologies 619
growing soybean. From our experiment, promiscuous variety and be adopted by farmers who will increase production of 620
TGx 1740-2F accumulated more N and K in the above ground soybean per unit area. 621
biomass at R3 stage than the non-promiscuous, Safari. Inputs that supply N and/or P promoted soybean development 622
Accumulation of N and K was also documented at R3 stage in and improves yield production. Our study shows that using inputs 623
The Journal of Agricultural Science 11
Table 4. Yield and yield components of soybean varieties under nutrient amendments 624
625
Angonia 2017 Ruace 2018
626
100 Seed 100 Seed 627
Input Pods/m2 Seed/pod weight (g) Yield kg/ha Pods/m2 Seed/pod weight (g) Yield kg/ha 628
629
Safari 630
Check 1373 2.2 17.4 2394 999 2.3 15.8 2216 631
632
Inoculant (Inoc) 1997 2.4 16.3 3347 1573 2.4 17.6 2568
633
Nitrogen 2182 2.4 16.2 3284 1379 2.7 22.2 2805 634
Phosphorus (Phos) 1770 3.0 15.7 3349 1647 2.6 24.0 2668 635
636
Phos + Inoc 2070 2.5 18.4 3314 1394 2.4 21.8 2885
637
TGx 1740-2F 638
Check 2290 2.7 16.0 2316 1424 2.3 12.6 1853 639
640
Inoculant (Inoc) 1940 2.4 14.2 3029 1690 2.3 17.0 2998
641
Nitrogen 2392 2.0 14.9 3243 2105 2.3 17.3 2573 642
Phosphorus (Phos) 1813 2.6 12.0 3222 1727 2.3 18.8 2688 643
644
Phos + Inoc 1812 2.5 13.3 3615 1534 2.3 20.0 3253 645
S.E.M. 285.2 0.18 1.65 215.4 164.2 0.06 1.66 162.3 646
The variety and input differences denoted by letters at P≤ 0.05 significant level between inputs and across varieties for each variable 647
648
649
650
Table 5. Soil properties (0–20 cm) and texture at Angonia and Ruace study sites 651
652
Total org. Total N CEC (mmol Sand Silt Clay
Location pH (1:2 H20) C (%) (%) P (mg/kg) K (mg/kg) EC (uS/cm) (+)/kg) (%) (%) (%)
653
654
Angonia 2017 5.7 1.96 0.10 2.6 163.5 79.0 13.9 51.8 7.1 41.1 655
Angonia 2018 5.4 2.03 0.14 1.6 94.6 84.0 15.3 47.5 7.1 45.5 656
657
Ruace 2017 6.0 2.10 0.16 25.5 272.5 55.5 13.7 58.4 11.1 30.6 658
Ruace 2018 6.2 1.66 0.14 17.3 226.0 57.5 16.4 55.3 13.1 31.7 659
S.E.M. 0.11 0.153 0.013 3.97 27.49 6.75 1.63 3.00 1.25 3.16 660
661
662
663
in production of soybean in different sites leads to distinct yield the other hand, yielded higher at Ruace especially for TGx 664
increase potentials. The yield increase potentials are dependent 1740-2F. This signifies that P levels in the soil at this location 665
on the availability of nutrients in an agroecological zone, climatic were also high (Table 1) and that a lower application rate could 666
conditions favouring a variety and the accompanying soybean be sufficient for soybean production. Application of these nutri- 667
agronomic management practices. Differences in the yield ent to soil act together with specific location conditions impacting 668
increase potential signifies variation of soybean variety suitability on soybean development such as nodulation, number of pods and 669
in a particular agroecology. Bekere and Hailemaria (2012) reported seed weight that contribute towards yield in a complex inter- 670
increase in soybean yield when singly inoculated (256–480 kg/ha) action. Since inoculant could easily be affordable among most 671
and with P fertilizer application (264–801 kg/ha) while Majengo smallholder farmers, their soybean yield could increase by 61% 672
et al. (2011) found out that inoculation increased soybean yield (1145 kg/ha). However, combining P with the inoculant would 673
between 0.5 and 1.0 t/ha in Western Kenya. Production of soy- improve yield more by 76% (1401 kg/ha). Of course, this yield 674
bean in Mozambique and Southern Africa in general can be increase on P application and inoculation does not negate the 675
improved through selection of varieties suitable for specific importance of supplying N in low quantities as a starter fertilizer 676
regions, application of cost-effective inputs and marching them for soybean production where necessary in depleted soils. When 677
with good agronomic management practices. In a good season P is adequate in the soil, inoculation will promote soybean root 678
such as at Angonia in 2017, use of inoculant, N or combining development enhancing P scavenging ability (He et al., 2017). For 679
P+ inoculant was advantageous for both varieties. In this same such locations, breeding for soybean varieties with better P scaven- 680
location, use of P alone was not beneficial but performed better ging abilities or promoting soil mychorrizal associations would 681
when combined with inoculant. High yield was realized when boost the benefits of inoculation. 682
P+inoculant was applied on both varieties. Similar results of syn- Inoculation of soybean and other legumes is economically jus- 683
ergetic effect of combining P+inoculant were reported for soybean tifiable in many SSA fields due to its low cost in comparison to 684
production in Pakistan (Afzal et al., 2010). Inoculation alone on inorganic fertilizers (Kyei-Boahen et al., 2017). However, selection 685
12 C. E. N. Savala et al.
of appropriate input should be done after soil analysis, which is for promiscuous varieties have been reported when evaluating 686
seldom among smallholder farmers in Africa, to determine the TGx 1835-10E, TGx 1904-6F and TGx 1448-2F at Samaru 687
most limiting and other deficient nutrients before deciding on Zaria in the northern Guinea savanna and Samaru-Kataf in the 688
the rate. Safari responded to inoculant better than TGx 1740-2F southern Guinea savanna (Kamara et al., 2014) and five TGx var- 689
at Angonia, perhaps due to reduced ability of introduced rhizobia ieties against non-promiscuous Bossier in Tanzania (Lyimo et al., 690
to form association or a result of competition from ineffective 2017). Although TGx 1740-2F variety is popular, breeders have a 691
indigenous strains with the latter variety. Inoculating with the challenge to increase the seed size without compromising the pod 692
right rhizobia strain results in more N assimilation hence seed density. 693
increased yield (Giller et al., 2011). Use of inoculants improves The process of transpiration, photosynthesis and WUE are 694
yield of both determinate and indeterminate varieties especially affected by both biotic and abiotic factors. Biotic factors that affect 695
in soils with no previous history of soybean production these processes are plant canopy characteristics (leaf angle, size, 696
(Sanginga et al., 2002; Zimmer et al., 2016). Therefore, inoculant colour and stomata) and root development while the abiotic fac- 697
and P are essential for both the host plant and rhizobia in the tors include temperature, soil moisture, sunlight and wind. These 698
BNF process. The symbiotic relationship between the nitrogen factors interact to determine the rate of the processes. Agronomic 699
fixing legumes and the rhizobia strain leads to transformation management have an effect on plant characteristics and soil mois- 700
of atmospheric N into forms that can be absorbed by the plant ture availability. Transpiration increased from flowering to pod- 701
while the bacteria benefits by acquiring assimilates from the ding at different rates when inputs were applied. By the time of 702
host plant. grain filling stage, most of the soybean had reached canopy clos- 703
Inoculation provides the right rhizobia strain for symbiotic ure covering the soil hence reducing air current movement lead- 704
association for both the promiscuous and non-promiscuous var- ing to comparable transpiration rate for all the inputs. The 705
ieties leading to formation of more nodules on soybean roots. increase was occasioned by leaf development in size and number 706
There were significantly more nodules per plant in inoculated which created a large surface area and more active stomata. As 707
treatments of TGx 1740-2F than Safari at Ruace. Perhaps at plant grows, leaf density in relation to the ground area known 708
Ruace, apart from the introduced strains, TGx 1740-2F could as LAI increases affecting PAR intercepted. As more radiation is 709
have formed additional associations with resident strains that intercepted by the actively photosynthesizing leaves, the percent 710
promoted more nodulation. Our findings concur with nodulation PAR increases. However, there is a caveat to an increase in percent 711
results between promiscuous and non-promiscuous varieties PAR and the plant productivity. A higher PAR does not necessar- 712
where nodule numbers per plant at podding stage for TGx ily translate to a more productive plant canopy but rather only 713
1893-10F were significantly higher than Gazelle when inoculated indicate the amount of radiation intercepted. The productivity 714
in Meru, Kenya (Gitonga et al., 2010). The relationship between will depend on the amount of actively photosynthesizing leaves 715
the nodule numbers and their weights was not clear in all the in the canopy. Since the PAR was determined between flowering 716
sites despite the existence of a trend possibly because of the vari- and grain filling stage when most of the leaves are photosynthesiz- 717
ation in nodule size i.e. those formed at the crown are bigger than ing, the values could be an indicator of productivity. Coupled with 718
those on the lateral roots. It was evident that from the variability environmental factors such as temperature, wind and soil mois- 719
in nodule weight, that the size of the nodules also varied between ture, soybean lost water through the leaves to help cool the 720
site. Some treatments had large nodules that resulted in a higher plant and transport nutrients by mass-flow from the roots to 721
weight. For instance, comparable nodule numbers at Angonia above ground tissues. There was a relatively low transpiration 722
between P and P+inoculant corresponded to different weights. rate with inoculated fields perhaps due to some of the imbibed 723
Several studies have reported high nodule number when inoculant water being retained in the root system to maintain turgidity 724
is applied together with P fertilizer. A high nodule number and and BNF activity of the nodules. At flowering, fewer leaves had 725
increased BNF process was reported with inoculated soybean developed at Ruace than Angonia. However, by the time of pod- 726
that received supplemental P fertilizer at Sumbrungu, Ghana ding, soybean in both locations had a comparable canopy with 727
(Akpalu, 2014) and Bvumbwe, Malawi (Phiri et al., 2016). LAI between 3.5 and 4.5. It has been reported that excess increase 728
Although the number of nodules and weight could be an indicator in the LAI at flowering stage could lead to increased microclimate 729
of high BNF process, it is not always the case unless the nodule temperatures that cause flower abortion in soybean, hence 730
activity is determined. lowering the yield. Inoculated soybean also had a low LAI than 731
Interaction of yield components such as pods per plant, seeds P treatments between flowering and podding stages. Inoculation 732
per pod and seed weight affects the final production of soybean. reduced transpiration of TGx1485-1D and TGx1440-1E unlike 733
Increase in number of pods, seeds per pod and heavier seeds in TGx1448-2E, TGx1740-2F and TGx1445-3E where it led to 734
if considered singly are likely to positively correlate with higher elevated stomatal conductance that resulted in more water loss 735
yield. As observed in some locations, more seeds per pod corre- (Pule-Meulenberg et al., 2011). Therefore, it is possible for tran- 736
sponded to high number of seeds while less seeds per pod led spiration rate to be different among genotypes of a particular 737
to a high weight of 100 seeds. The complex interaction and differ- crop with respect to the environment and management practices 738
ences in yield components emphasize plant compensation in such as nutrient supply. However, high transpiration rates due to 739
response to input application and management (Pedersen and increased mesophyll over stomatal conductance has been found 740
Lauer, 2004). Varietal differences play a role in the resultant not to affect carboxylation that is necessary for photosynthesis 741
seed weight. Safari had generally heavier seeds than TGx in some soybean varieties (Buezo et al., 2019). 742
1740-2F. There were also some varietal differences in the number There were also distinct transpiration trends between podding 743
of pods/m2; mostly, the indeterminate variety TGx 1740-2F had and grain filling stages with respect to varieties and sites. At 744
more pods/m2. This could be as a result of the TGx 1740-2F Angonia for instance, the rate declined between these stages per- 745
variety having longer reproductive period (60 days) v. Safari haps due to cooling of the atmospheric temperatures that started 746
(56 days) in Ruace (data not reported). Comparable pods/m2 in the month of April which could have reduced the activity of 747
The Journal of Agricultural Science 13
many biochemical processes. The declining temperatures coin- Conclusion 748
cided with the period between the stages. Temperature seems to 749
Yield of both promiscuous and non-promiscuous soybean var-
have been the main factor driving the transpiration process 750
ieties in Mozambique can be improved by use of inoculants and
since the other factors such as soil moisture were not limiting 751
application of supplemental P fertilizer. Use of inoculant that is
due to available rains. As temperature declines, the transpiration 752
economically justifiable due to its low cost in relation to the inor-
rate also decreases because the vapour pressure gradient between 753
ganic fertilizers, N and P either singly or in combination led to an
the leaves and the surrounding air become low (Fletcher et al., 754
increase in soybean yield in all the sites. Soybean plant processes
2007). Contrarily, the rate increased between the stages at 755
such as transpiration, photosynthesis, WUE, PAR and LAI vary
Ruace. At Ruace, the temperature did not reduce (Fig. 1) at the 756
depending on the type of nutrient applied. As a result of increased
period between podding and grain filling, as air remained warmer, 757
WUE, soybean will utilize water efficiently to produce more bio-
creating a steep vapour pressure gradient. High air temperatures 758
mass especially grain production that lead to an increase in yield
and low humidity conditions lead to an increase in the transpir- 759
per unit area.
ation due to increased stomatal conductance (Habermann et al., 760
2019). Following the two transpiration rate scenarios between Acknowledgements. The authors greatly acknowledge financial support 761
the two locations, it was also noted that soybean dried in a shorter from the Consortium of International Agricultural Research Centers 762
period (98 days after emergence) at Ruace than Angonia (over 108 (CGIAR) through the Research Program on Grain Legumes and Dryland 763
days after emergence). Use of N and P inputs in soybean produc- Cereals (CRP-GLDC) in Mozambique. Thanks to the IITA technical staff at 764
tion reduced the transpiration rate in relation to photosynthesis Angonia and Gurue stations in Mozambique for managing the trials and 765
rate unlike for the check. The check took a longer period for collecting of field-related data. 766
the canopy to close coupled with less nutrient uptake at flowering Financial support. This work was supported by the Consortium of 767
stage and decreasing LAI between podding and grain filling stages International Agricultural Research Centers (CGIAR) through the Research 768
leading to open large exposed intra-row soil surface. Open spaces Program on Grain Legumes and Dryland Cereals (CRP-GLDC). 769
between rows could have led to more wind currents that resulted 770
in increased transpiration. Conflict of interest. The authors declare that the research was conducted in 771
Photosynthesis differences were prominent between varieties the absence of any commercial or financial benefits that could be construed as 772
and sites than with the inputs (Fig. 3). There were distinct differ- a potential conflict of interest. 773
ences in the photosynthesis rates between the varieties at Angonia Ethical standards. Not applicable. 774
and Ruace. At Ruace the rate did not change significantly with the 775
TGx 1740-2F variety from flowering to podding through to grain 776
filling stage. TGx 1740-2F continually develops new leaves which 777
have longer life span than Safari. However, significant differences References 778
in photosynthesis between Safari and TGx 1740-2F were observed Adhikari U, Nejadhashemi AP and Woznicki SA (2015) Climate change and 779
at flowering in Angonia and at podding in Ruace. The increase in Eastern Africa: a review of impact on major crops. Food and Energy Security 780
the photosynthesis rate between podding and grain filling stage 4, 110–132. 781
seem to have a bearing on yield. At grain filling stage, photosyn- Afzal A, Asghari B and Mussarat F (2010) Higher soybean yield by inocula- 782
thesis rate between Safari (21.3 μmol/m2/s) and TGx 1740-2F tion with N-fixing and P-solubilizing bacteria. Agronomy for Sustainable 783
(20.6 μmol/m2/s) was slightly numerically different but statistic- Development 30, 487–495. 784
Akpalu M (2014) Phosphorus application and rhizobia inoculation on growth
ally insignificant and corresponded to a yield of 2520 and 2247 and yield of soybean (Glycine max L. Merrill). American Journal of 785
kg/ha respectively at Angonia. Nagasuga (2019) reported that soy- Experimental Agriculture 4, 674–685. 786
bean yield could be improved through elevated photosynthesis Anda A, Simon B, Soós G, Teixeira da Silva JA and Kucserka T (2018) 787
that increases pod and seed numbers. Comparable soybean sea- Crop-water relation and production of two soybean varieties under different 788
sonal carbon dioxide exchange rate values that estimate photosyn- water supplies. Theoretical and Applied Climatology 137, 1515–1528. 789
thesis between R1 and R6 (De Bruin et al., 2010) R2 and R4 (Gai Anyia AO, Slaski JJ, Nyachiro JM, Archambault DJ and Juskiw P (2007) 790
et al., 2017) have also been documented similar to our findings Relationship of carbon isotope discrimination to water use efficiency and 791
between R3 and R6 growth stages. productivity of barley under field and greenhouse conditions. Journal of 792
The ratio of photosynthesis to transpiration was used to deter- Agronomy and Crop Science 193, 313–323. 793
mine WUE that estimated plant productivity per unit of water Bassi D, Menossi M and Mattiello L (2018) Nitrogen supply influences 794
photosynthesis establishment along the sugarcane leaf. Scientific Reports
spend at particular stage depended on variety and input applied. 8, 1–13. 795
Treatments that received fertilizer or inoculant-based N had a Bekere W and Hailemaria A (2012) Influences of inoculation methods and 796
higher WUE. Nitrogen is a major component of leaf chlorophyll phosphorus levels on nitrogen fixation attributes and yield of soybean 797
and amino acids that are protein and nucleic acids (deoxyribo- (Glycine max L.) at Haru, Western Ethiopia. American Journal of Plant 798
nucleic acid-DNA) building blocks. Availability of N which leads Nutrition and Fertilization Technology 2, 45–55. 799
to increased photosynthesis was reported in China (Gai et al., Buezo J, Sanz-saez Á, Moran JF, Soba D and Aranjuelo I (2019) Drought 800
2017). The study found out that as N was applied at 0, 25 and tolerance response of high-yielding soybean varieties to mild drought: 801
50 kg N/ha, the photosynthesis rate of soybean at R2 stage increased physiological and photochemical adjustments. Physiolgia Plantarum 166, 802
(23.54, 23.59 and 36.42 μmol CO2/m
2/s), respectively. Therefore, 88–104. 803
inputs that avail more N to plants such as inoculants and organic Condon AG, Richards R.A, Rebetzke GJ and Farquhar GD (2002) 804
Improving intrinsic water-use efficiency and crop yield. Crop Science 42,
fertilizers lead to increased photosynthesis hence increased WUE. 122–131. 805
This indicates that use of inoculant, timely application of N and P Conway G (2009) The science of climate change in Africa. Grantham Institute 806
fertilizers in soybean production provide important elements neces- Discussion Papers 1, 1–24. 807
sary for photosynthesis, better WUE could translate into improved CramerMD, Hoffmann V and VerboomGA (2008) Nutrient availability mod- 808
soybean yield inMozambique and the larger southernAfrica region. erates transpiration in Ehrharta calycina. New Phytologist 179, 1048–1057. 809
14 C. E. N. Savala et al.
De Bruin JL, Singer JW, Pedersen P and Rotundo JL (2010) Soybean photo- microbial inoculants on soybean production in Bungoma, Kenya. African 810
synthetic rate and carbon fixation at early and late planting dates. Crop Crop Science Conference Proceedings 10, 117–119. 811
Science 50, 2516–2524. Matimati I, Verboom GA and Cramer MD (2014) Nitrogen regulation of 812
Delamuta JRM, Ribeiro RA, Ormeño-Orrillo E, Melo IS, Martínez-Romero transpiration controls mass-flow acquisition of nutrients. Journal of
813
E and Hungria M (2013) Polyphasic evidence supporting the reclassifica- Experimental Botany 65, 159–168.
814
tion of Bradyrhizobium japonicum group Ia strains as Bradyrhizobium dia- Medrano H, Tomás M, Martorell S, Flexas J, Hernández E, Rosselló J, Pou
zoefficiens sp. nov. International Journal of Systematic and Evolutionary A, Escalona JM and Bota J (2015) From leaf to whole-plant water use effi- 815
Microbiology 63, 3342–3351. ciency (WUE) in complex canopies: limitations of leaf WUE as a selection 816
Fletcher AL, Sinclair TR and Allen LH (2007) Wilting and commercial target. Crop Journal 3, 220–228. 817
soybean transpiration responses to vapor pressure deficit in well watered Moore KJ and Dixon PM (2015) Analysis of combined experiments revisited. 818
‘slow-wilting’ and commercial soybean. Environmental and Experimental Agronomy Journal 107, 763–771. 819
Botany 61, 145–151. Nagasuga K (2019) Soybean seed production and canopy photosynthesis in 820
Gai Z, Zhang J and Li C (2017) Effects of starter nitrogen fertilizer on soybean cultivation. In Kasai M (ed.), Soybean – Biomass, Yield and Productivity. 821
root activity, leaf photosynthesis and grain yield. PLoS ONE 12, 1–15. London, UK: IntechOpen, pp. 41–58. 822
Giller KE, Murwira MS, Dhliwayo DKC, Mafongoya PL and Mpepereki S Ngcamu BS and Chari F (2020) Drought influences on food insecurity in
823
(2011) Soyabeans and sustainable agriculture in Southern Africa. Africa: a systematic literature review. International Journal of
824
International Journal of Agricultural Sustainability 9, 50–58. Environmental Research and Public Health 17, 1–17.
Gitonga MN, Gatheri GW, Cheruiyot R, Gitonga NM and Maingi J (2010) Pedersen P and Lauer JG (2004) Response of soybean yield components to 825
Nodulation and nitrogen fixation in promiscuous and non promiscuous management system and planting date. Agronomy Journal 96, 1372–1381. 826
soybean (Glycine max (L.) Merrill) varieties in Eastern Kenya. Journal of Phiri A, Kausiwa D, Omollo J, Muindi E and Yegon R (2016) Inoculated 827
Tropical Microbiology and Biotechnology 6, 3–8. soybean yields response to nitrogen and phosphorus application. 828
Habermann E, Dias de Oliveira EA, Contin DR, San Martin JAB, Curtarelli International Journal of Plant and Soil Science 10, 1–7. 829
L, Gonzalez-Meler MA and Martinez CA (2019) Stomatal development Pule-Meulenberg F, Gyogluu C, Naab J and Dakora FD (2011) Symbiotic N 830
and conductance of a Tropical forage legume are regulated by elevated nutrition, bradyrhizobial biodiversity and photosynthetic functioning of six 831
[CO2] under moderate warming. Frontier in Plant Science 10, 1–17. inoculated promiscuous-nodulating soybean genotypes. Journal of Plant 832
Hart, A. L. (1990) Nodule activity and nodule phosphorus in white clover Physiology 168:540–548.
833
(Trifolium repens) in the presence of mineral nitrogen. Plant Nutrition Rizza F, Ghashghaie J, Meyer S, Matteu L, Mastrangelo AM and Badeck FW
834
Physiology and Applications 41, 189–192. (2012) Constitutive differences in water use efficiency between two durum
He J, Du YL, Wang T, Turner NC, Yang RP, Jin Y, Xi Y, Zhang C, Cui T, wheat cultivars. Field Crops Research 125, 49–60. 835
Fang XW and Li FM (2017) Conserved water use improves the yield per- Rurangwa E, Vanlauwe B and Giller KE (2018) Agriculture, ecosystems and 836
formance of soybean (Glycine max (L.) Merr.) under drought. Agricultural environment benefits of inoculation, P fertilizer and manure on yields of 837
Water Management 179, 236–245. common bean and soybean also increase yield of subsequent maize. 838
Houshmandfar A, Fitzgerald GJ, O’Leary G, Tausz-Posch S, Fletcher A and Agriculture, Ecosystems and Environment 261, 219–229. 839
Tausz M (2018) The relationship between transpiration and nutrient uptake Sanginga N, Okogun J, Vanlauwe B and Dashiell K (2002) The contribution 840
in wheat changes under elevated atmospheric CO2. Physiologia Plantarum of nitrogen by promiscuous soybeans to maize based cropping the moist 841
163, 516–529. savanna of Nigeria. Plant and Soil 241, 223–231. 842
Hufstetler EV, Boerma HR, Carter Jr TE and Earl HJ (2007) Genotypic vari- SAS Institute Inc. (2018) SAS/STAT® 15.1 User’s Guide. Cary, NC: SAS
843
ation for three physiological traits affecting drought tolerance in soybean. Institute Inc.
844
Crop Science 47, 25–35. Savala CEN and Kyei-Boahen S (2020) Potential of inoculant and phosphorus
Irmak S and Specht JE (2014) Soybean yield, evapotranspiration, water prod- application on soybean production in Mozambique. Universal Journal of 845
uctivity, and soil water extraction response to subsurface drip irrigation and Agricultural Research 8, 46–57. 846
fertigation. Transactions of the ASABE 57, 729–748. SiahpooshMR, Dehghanian E andKamgar A (2011) Drought tolerance evalu- 847
Jakobsen I (1985) The role of phosphorus in nitrogen fixation by young pea ation of bread wheat genotypes using water use efficiency, evapotranspiration 848
plants (Pisum sativum). Physiologia Plantarum 64, 190–196. efficiency, and drought susceptibility index. Crop Science 51, 1198–1204. 849
Jones A, Breuning-Madsen H, Brossard M, Dampha A, Deckers J, Dewitte Sinclair TR, Hammer GL and van Oosterom EJ (2005) Potential yield and 850
O, Gallali T, Hallett S, Jones R, Kilasara M, Le Roux P, Micheli E., water-use efficiency benefits in sorghum from limited maximum transpir- 851
Montanarella L, Spaargaren O, Thiombiano L, Van Ranst E, Yemefack ation rate. Functional Plant Biology 32, 945–952. 852
M and Zougmoré R (eds) (2013) Soil Atlas of Africa. Luxembourg: Tatsumi K, Yamashiki Y, Valmir da Silva R, Takara K, Matsuoka Y,
853
European Commission, Publication Office of the European Union. 176pp. Takahashi K, Maruyama K and Kawahara N (2011) Estimation of poten-
854
Kamara AY, Ewansiha SU, Boahen S and Tofa AI (2014) Agronomic tial changes in cereals production under climate change scenarios.
response of soybean varieties to plant population in the Guinea savannas Hydrological Processes 25, 2715–2725. 855
of Nigeria. Agronomy Journal 106, 1051–1059. Tefera H, Kamara AY, Asafo-Adjei B and Dashiell KE (2010) Breeding pro- 856
Khojely DM, Ibrahim SE, Sapey E and Han T (2018) History, current status, gress for grain yield and associated traits in medium and late maturing pro- 857
and prospects of soybean production and research in sub-Saharan Africa. miscuous soybeans in Nigeria. Euphytica 175, 251–260. 858
Crop Journal 6, 226–235. Thornton PK, Jones PG, Alagarswamy G and Andresen J (2009) Spatial 859
Kyei-Boahen S, Savala CEN, Chikoye D and Abaidoo R (2017) Growth and variation of crop yield response to climate change in East Africa. Global 860
yield responses of cowpea to inoculation and phosphorus fertilization in dif- Environmental Change 19, 54–65. 861
ferent environments. Frontiers in Plant Science 8, 1–13. Uprety D and Mahalaxmi V (2000) Effect of elevated CO2 and nitrogen nutri- 862
Lobell DB and Gourdji SM (2012) The influence of climate change on global tion on photosynthesis, growth and carbon-nitrogen balance in Brassica
863
crop productivity. Plant Physiology 160, 1686–1697. juncea. Journal of Agronomy and Crop Science 184, 271–276.
864
Lyimo LD, Tamba MR and Madege RR (2017) Effects of genotype on yield Zimmer S, Messmer M, Haase T, Piepho H, Mindermann A, Schulz H,
and yield component of soybean (Glycine max (L) Merrill). African Journal Habekuß A, Ordon F, Wilbois K and Heß J (2016) Effects of soybean var- 865
of Agricultural Research 12, 1930–1939. iety and Bradyrhizobium strains on yield, protein content and biological 866
Majengo CO, Okalebo JR, Lesueur D, Pypers P, Ng’etich W, Misyoki MW nitrogen fixation under cool growing conditions in Germany. European 867
and Mburu M (2011) Interaction between nitrogen and phosphorus Journal of Agronomy 72, 38–46. 868
869
870
871