agriculture Article Impact of Extreme Temperature and Soil Water Stress on the Growth and Yield of Soybean (Glycine max (L.) Merrill) Labake Ogunkanmi 1, Dilys S. MacCarthy 2,* and Samuel G. K. Adiku 1 1 Department of Soil Science, University of Ghana, Legon, Accra GA-489-9979, Ghana; loogunkanmi001@st.ug.edu.gh (L.O.); s_adiku@ug.edu.gh (S.G.K.A.) 2 Soil and Irrigation Research Centre, University of Ghana, Kpong EL-0633-5197, Ghana * Correspondence: dmaccarthy@ug.edu.gh; Tel.: +233-0-24-409-0502 Abstract: Climate change is a major environmental stressor that would adversely affect tropical agriculture, which is largely rain-fed. Associated with climate change is an increasing trend in temperature and decline in rainfall, leading to prolonged and repeated droughts. The purpose of this study was to determine the effect of climate variables such as temperature, relative humidity, vapor pressure deficit (VPD), and soil water on the phenology, biomass, and grain yield of soybean crops. A greenhouse experiment was set in a split plot design with three average environmental conditions as the main plots: E1 (36 ◦C, RH = 55%), E2 (34 ◦C, RH = 57%) and E3 (33 ◦C, RH = 44%). Additionally, there were three water treatments: W1 (near saturation), W2 (Field capacity), and W3 (soil water deficit) and two soybean varieties (Afayak and Jenguma). These treatments were replicated nine times. The results showed that high temperatures (E1) accelerated the crop development, particularly at flowering. Additionally, increased atmospheric demand for water under a high temperature environment resulted in high evapotranspiration, leading to high transpiration which probably reduced photosynthetic activity of the plants and thereby contributing to biomass and grain yield loss. Biomass and yield were drastically reduced for the combined effect of high temperature (E1)   and drought (W3) as compared to combined effect of ambient temperature (E3) and well-watered Citation: Ogunkanmi, L.; MacCarthy, condition (W1). Increasing temperatures and erratic rainfall distributions associated with climate D.S.; Adiku, S.G.K. Impact of change poses a potential threat to the soybean production in Ghana. Extreme Temperature and Soil Water Stress on the Growth and Yield of Keywords: drought; smallholders; climate change; soybean; Ghana Soybean (Glycine max (L.) Merrill). Agriculture 2022, 12, 43. https:// doi.org/10.3390/agriculture12010043 Academic Editor: Pascual Romero 1. Introduction Changes in weather components are largely responsible for the frequent yield vari- Received: 26 November 2021 Accepted: 23 December 2021 ability and gaps that have been recorded in the literature, especially in Sub-Saharan Africa Published: 31 December 2021 (SSA) [1,2]. Within the last few decades, the impact of the rapidly increasing tempera- tures and erratic rainfall patterns on crop yields have become more evident in the SSA [3]. Publisher’s Note: MDPI stays neutral An analysis of weather patterns between 1960 and 2000 in Ghana indicated a general tem- with regard to jurisdictional claims in perature rise of about 1 ◦C over the whole country [4]. Associated with this is a decline in published maps and institutional affil- rainfall, which generally decreases from south to north [2]. iations. Indeed, projections show that these trends will continue at least into the near fu- ture [1,5]. These developments spell adverse conditions for rain-fed agriculture, especially due to the inherently low productivity of the soils. The most cultivated and dominant Copyright: © 2021 by the authors. soils in the Guinea savanna zone of Ghana which is the hub for soybean cultivation are the Licensee MDPI, Basel, Switzerland. acrisols, which are characteristically low activity clay soils [6] that have a coarse texture as This article is an open access article well as low water holding capacities. In particular, soils in northern Ghana have degraded distributed under the terms and over time, with some losing almost 50% of the top soils and becoming very gravelly and conditions of the Creative Commons shallow [7]. Though soils of the middle belt are inherently deeper, they are also degrading Attribution (CC BY) license (https:// at very fast rates. creativecommons.org/licenses/by/ The combination of rapidly changing weather and the declining soil productivity has 4.0/). adverse consequences on crop growth [8]. First, the crop development rate is accelerated Agriculture 2022, 12, 43. https://doi.org/10.3390/agriculture12010043 https://www.mdpi.com/journal/agriculture Agriculture 2022, 12, 43 2 of 13 when the ambient temperature increases [9], thereby shortening the overall life cycle of the plants. As a consequence, the crop spends less time in the field accumulating biomass, leading to reduction in size, shorter reproductive duration, and reduced overall yield [9]. Second, the plant respiration rate increases with temperature [10], resulting in a reduced net assimilate accumulation [11]. Therefore, even under non-limiting soil water conditions, crop growth would be impaired under increasing temperature conditions. Additionally, increased temperatures would increase the potential evapotranspiration while reducing the vapor pressure, with a resultant increased evaporative demand on the crop [11]. If rainfall reduces as projected under climate change, soil water replenishment and availability would also be adversely affected, especially due to agricultural drought (low water storage capacity). Literature sources have indicated that the combined effect of high temperature and drought has more detrimental effects on yield and grain number as compared to their individual effects [12]. A study carried out by Jumrani and Bhatia [11] showed a significant decline in the seed yield of soybean sown at high temperatures of 38 and 42 ◦C with yield reductions of 42 and 64%, respectively. It was also indicated that the decline was very high at all temperatures (30, 34, 38 and 42 ◦C) when plants were subjected to water stress. An understanding of how plants adapt to the increasing climate change impact is necessary to guide policies and practices that should be developed to minimize the impacts, especially as many sub-Saharan governments seek to expand the cultivation of some crops into their non-traditional growing belt. In Ghana, soybean (Glycine max) is one crop that is promoted out of their traditional growing areas in the south to more savannah locations found in the southern coastal zone and in the northern locations. The desire to achieve this lies in the superior protein and oil content compared with the other traditional legumes (such as cowpea, Phaseolus spp., etc.). The successful introduction of soybean crops into northern Ghana would address the nutritional protein needs that is often a challenge in those locations. Yet, whether or not the changing weather and soil challenges would support the policy drive is yet to be fully understood. Questions of how different varieties will tolerate the increasing heat and water stress still continue to remain open. Presumably, the longer-duration high yielding variety Jenguma (110–115 days), grown in Ghana would better adapt to the stressed environments than the shorter-duration varieties (e.g., Afayak), since the longer natural life cycle of the former could offset the shortening effect of increasing temperatures. This hypothesis is yet to be validated. It is the purpose of this study to investigate, with the aim of gaining further un- derstanding of the responses of the development, the growth and yield of two soybean varieties of contrasting life cycles (namely Jenguma and Afayak) to increased temperature and water stress conditions under greenhouse conditions. 2. Materials and Methods 2.1. Site Description and Experimental Set Up This study was carried out at the University of Ghana, Soil and Irrigation Research Centre (SIREC)—Kpong (6◦9′ N and 0◦4′ E, 22 m alt) which is located within the coastal savannah zone of Ghana. The soil used is classified as Typic Calciustert [13], locally known as Tropical Black Clay called Akuse series [14]. The soils were potted and placed in three (3) growth chambers constructed by covering a wooden framework with transparent plastic sheets. The dimensions of each chamber were 3.0 m (length), 0.8 m (width), and 1 m (height). The temperatures within the chambers were not controlled to maintain constant values. However, different temperature regimes were realized based on the number and size of windows created on the sides of the chambers. Two of the three chambers, E1 and E2, had two and four window openings respectively, each of size 35 cm by 35 cm, resulting in an average seasonal temperature of 36 ◦C for the chamber with two windows and 34 ◦C for the one with four windows. The third chamber, E3, which served as the control had half of the polythene sheet removed from all sides to allow free wind circulation resulting in an average seasonal temperature of 33 ◦C. The day-to-day variations in the temperature and Agriculture 2022, 12, 43 3 of 13 relative humidity in this chamber reflect conditions that would pertain naturally on crop fields. 2.2. Treatment Structure Each chamber received six treatments replicated nine times comprising two soybean varieties (V1 = Afayak: TGX 1834-5E and V2 = Jenguma: Tax 1445-2E) widely cultivated in northern Ghana [15] and three water regimes: W1 (post-flowering soil water content kept near saturation), W2 (soil water content kept near field capacity throughout the growing period, and W3 (post-flowering drying cycle). The total experimental units were 162 pots with 54 in each growth chamber. The experiment was laid out in a split-plot design, with the temperature chambers as the main plot with the other factors (water and varieties) being the sub-plots randomized within each main plot (chamber). The Jenguma (Tax 1445- 2E) variety has an attractive grain color (cream), high oil content about 20%, is resistant to pod shattering in the field, and is a determinate. Jenguma has an approximate grain yield of 1.7–2.8 t/ha (17–28 bags/ha) [16]. Afayak (TGX 1834-5E) variety is also resistant to shattering, lodging, resistant to pest infestation with a potential yield of 2.0–2.2 t/ha and is also determinate. Both varieties are effective in the control of Striga hermonthica [17]. The pots which had dimensions of 15 cm diameter and 14 cm width were filled with soil to a bulk density of 1.34 g/cm3, resulting in 2 kg of sieved soil per pot. The potted soils were pre-saturated with water and allowed to drain for two days before sowing. Before transferring the pots to the chambers, the emerged seedlings were nursed in a larger screen house for 14 days to ensure uniformity and then thinned to 1 plant/pot. At 15 days after emergence (DAE), the pots were transferred to the growth chambers. The pots continued to receive watering to maintain the soil water at or near field capacity until flowering time when water treatments were imposed. The pots were weighed every other day and topped up with water for those treatments that required so. For W1, water was applied to saturate the pots with a head of 2 cm which was allowed to drain, transpire or evaporate before the re-watering to saturation. For W2, the soil water continued to be maintained at field capacity with no ponding. In the case of W3, watering frequency was reduced to achieve a longer drying cycle until maturity. In total W1, W2 and W3 received 21,900, 14,800 and 11,100 mL, respectively, by the end of the growing period. The water content in the pots under W1 and W2 were between 0.35–0.4 gg−1 and 0.25–0.3 gg−1, respectively. For W3, the water content declined from 0.30 gg−1 at the onset of water stress imposition to about 0.1 gg−1 at maturity. 2.3. Weather Variables Measurements The weather variables (temperature in ◦C and relative humidity in %) were measured in each growth chamber five times daily (6 am, 9 am, 12 noon, 3 pm and 6 pm) throughout the growth period using a combined temperature and humidity meter (BioTemp 1 × 1.5 V AAA). As the measurements were manual, no data could be collected in the night. The average temperatures and relative humidity in each of the growth chambers were used to estimate the daily vapor pressure deficit (VPD) following [18]: 100− RH VPD = × SVP (1) 100 7.5T SVP = 610.7× 10 (2) 237 + T where SVP is the standard vapor pressure (Pa), RH is the relative humidity (%), and T is the temperature (◦C). The potential evapotranspiration in each chamber was measured on daily basis by measuring the decrease in the level of water in measuring beakers that were placed in the chambers. Agriculture 2022, 12, 43 4 of 13 2.4. Plant Development The plant development was determined as the number of days to (i) 50% emergence, (ii) 50% flowering, (iii) 50% podding, and (iv) 50% physiological maturity. The time to reach each developmental stage was also expressed as the growing day degrees (GDD) or cumulative thermal time TT, defined as GDD = TT = ∑(Tav − Tb) t (3) where Tav (◦C) is the average daily temperature in a given chamber, T (◦b C) is the base temperature and, t is time. The value of T ◦b = 10 C was taken from the literature [19]. Additionally, data were collected on plant height, number of leaves, and rate of node appearance using selected plants that were tagged soon after emergence. 2.5. Plant Growth and Yield Plants were harvested sequentially during the growth period. Four (4) dry matter harvests were carried out at vegetative (28 DAE), flowering (35 DAE), pod formation (50 DAE) and at maturity stages. The dry matter was determined after oven drying for three days at 70 ◦C. At physiological maturity, the matured pods (i.e., color turned yellow to brown) were harvested to determine yield parameters such as pod number, seed number and seed weight. The undamaged pods were detached from the plants and counted, manually threshed and the undamaged seeds counted after which the seeds were oven dried and the seed dry weight estimated. 2.6. Statistical Analysis The data were analyzed with the analysis of variance (ANOVA) using GenStat statisti- cal software (12th edition, 2009, VSN International Ltd., Hemel Hempstead, UK). Means were separated using the Duncan Multiple Range Test and compared at 5% level of signifi- cance. Microsoft Excel (Office 2013, Microsoft Corporation, Redmond, WA, USA) was used for data entry and graphical representation of data were with Sigma Plot (2006 version, SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Weather Conditions in the Chambers and Watering Regimes Temperature variations during the growth period showed that the daily average temperature range for E1 (highest temperature) was from 29.7 to 41 ◦C, (Figure 1a) but the mean was 36 ◦C, giving a variability (CV) = of 7%. The temperature range for E2 were between 29.6 and 38 ◦C while E3 (ambient) were between 29.1 and 37.5 ◦C. Environment E2 and E3, had means of 34 and 33 ◦C, respectively, with both having CVs of 6%. On hourly time scale, the ranges were far higher (not shown). Indeed, the daily cycle data indicated that by early morning, the temperatures in all the chambers were similar but the greatest differences were observed at 3 pm. The effect of night time temperatures were not measured. Contrary to the temperature patterns, the relative humidity was lowest for E3 and highest for E2. In general, the RH declined over time (Figure 1b). The average relative humidity over the growing period in the chambers were 54, 57 and 44% for E1, E2, and E3, suggesting a drier condition for E3 than E1 and E2, respectively. Although the plants in E1 and E2 were under higher relative humidity, the higher temperatures in these environments increased the stress on the plants. Agriculture 2022, 12, x FOR PEER REVIEW 5 of 14 Agriculture 2022, 12, 43 5 of 13 45 (a) (b) E1 70 E2 40 E3 60 35 50 40 30 30 25 Date Fiigurre 1. (a) Tempeerraatturree ((◦°CC))a anndd( b(b) )r erlealtaitvieveh uhmumidiidtyity(% (%) i)n itnh ethger ogwrotwh tch acmhabmerbsetrhsr othurgohuoguhtotuhte tdhuer dautiroantioofnt hoef tehxep eexripmereinmt.ent. 33..22.. EEffffeecctt ooff TTeemmppeerraattuurree oonn PPllaanntt DDeevveellooppmmeenntt TThhee vvaarriieettyy AAffaayyaakk aattttaaiinneedd 5500%% fflloowweerriinngg ((aavveerraaggee 3388 DDAAEE)) eeaarrlliieerr tthhaann tthhee JJeenngguummaa vvaarriieettyy ((4400 DDAAEE) )ono nchcrhornoonloolgoigcaicl atlimtiem beasbias.s iHs.owHeovwere, vwerh,ewn hexepnreesxsperde sisne tdheinrmthale trimmael utinmites,u thneit sd,iftfheerednicffeesr ewnecrees nwote rseignnoifticsaignnt i(fiTcaabnlet 1(T).a Dblieff1er).enDciefsf eirne ntecmespienratteumrep ienr athtuer de iifn- ftehreendti fcfehraemntbcehrsa minbfleuresnincefldu epnlcaendt pdleavnetlodpemvelo◦ent p. mPelannt.t Pdleavnetlodpevmeelonpt, meespnet,ceiaslplye ctioal lmy ato- tmuraittuyr, iwtya, sw saosmsoewmhewath faatstfears tuenrduenrd Ee1r E(316 (°3C6) tCh)atnh tahnet choeocloero leenrveinrovnirmonemntesn ints tihnet hcaescea osef AoffaAyafaky. Iank. geInnegreanl,e trhael ,AthfaeyaAkf vayaarkietvya raipeptyeaarpepde taor ebde mtoobree smenosrietivseen tsoi ttievme ptoertaetmurpee draiftfuerre- ednifcfeesr ethnacens tthhea JnenthguemJean gvuamrieatvya. riety. Table 1. Chronological days and thermal time for developmental stages. Table 1. Chronological days and thermal time for developmental stages. Flowering Podding Maturity Environment VarieEtynviron- F◦lowering Podding Maturity CT (DAE) TT ( Cd) CT (DAE) TT (◦Cd) CT (DAE) TT (◦Cd) ment Variety ◦ CT (DAE) TT (°Cd) CT (DAE) TT (°Cd) CT (DAE) TT (°Cd) E1 (36 C) Afayak 37 894 48 1176 97 2416 E1 (36 ◦C) JenguEm1a (36 °C) A3f9ayak 93278 89542 48 1265 1176 101 97 25204616 E2 (34 ◦C) AfayEak1 (36 °C) Jen3g8uma 83995 92580 52 1194 1265 104 101 25235006 E2 (34 ◦C) Jenguma E3 (33 ◦C) AfayEak2 (34 °C) A 40 4f0ayak 9 83 3 98 5 3 89 5 55 2 0 50 1234 101 2429 1134 1194 97 104 22275130 E3 (33 ◦C) JenguEm2a (34 °C) Jen4g1uma 94103 9352 52 1174 1234 101 101 23264029 Environment (Envt) E3 (33 °C) A0.0fa0y1ak 40 809.312 3 50 1134 0.001 97 2271 Variety (Var) 0.003 0.001 0.001 Envt × Var E3 (33 °C) Je0n.6g0u4 ma 41 901.317 8 52 1174 0.001 101 2360 CTE, cnhvroirnoonlo-gical time ; TT, therma0l.t0im01e ; DAE, days after emer0g.e1n2c3e. 0.001 ment (Envt) VarieWtyi t(hVarer)g ard t o plant he0i.g0h0t3, both vari eties (Afa0y.0a0k1a nd Jengu ma) pro0d.u00ce1d the hig hest mean heights in E1 (36 ◦C) with Afayak showing a more rapid increase than Jenguma (FEingvutr e× 2Vaa)r. Simila rly, envir0o.n6m04e nt E1 had the 10 n0o.1d7e8s formed by 48 D0A.0E01fo r the A fayak CvTa,r icehtryonwohloegriecaals ttihmee;s TaTm, ethneurmmabl etirmoef; nDoAdEe, sdaaypsp aefaterre demoenrg5e0nDceA. E for Jenguma (Figure 2b). Thus, the Afayak variety showed greater sensitivity to increasing air temperature with regarWd ittoh nreogdaerdap tpoe palraanntc he.eiJgenhgt,u bmoaths hvoawrieetdielsi t(tAlefaoyrank oanddif fJeernegnucmeai)n ptrhoednuucmedb tehreo hf ingohdeests mfoeramne hdeiinghatlsl itnh rEe1e (e3n6v °iCro) nwmitehn Atsfapyraiko rshtoowfloinwge ar imngo.re rapid increase than Jenguma (Figure 2a). SAimltihlaorulyg,h enthveirroenwmaesntn Eo1d hiaffde rtehnec 1e0i nnotdhees tfootraml endu bmyb 4e8r DofAnEo fdoer sthfoe rAmfaeydakf ovrarbieottyh wvahreieretiaess tihnea lslatmhree enuenmvbireorn omf ennotds,etsh aeprpaetearoefdn oond e5a0p DpAeaEr afnocre Jednifgfuemread .(FTihgeurteo t2abl )n. uTmhubes,r tohfel eAafvayeaskf voarrAieftayy askhoinwEed1, gEr2eaatnerd sEen3switievriety3 5to, 2in5carenadsi3n2g, areirs pteemctpiveerlayt,uarte 5w0itDhA reEg,awrdh itloe nJeondgeu mapapheaadra3n7c,e3. 5Jeanngdum41a fsohroEw1e, dE 2liattnled oEr3 n, roe dspifefcetrievneclye. in the number of nodes formed in all three environments prior to flowering. Temperature (°C) 11-Sep 18-Sep 25-Sep 02-Oct 09-Oct 16-Oct 23-Oct 30-Oct 06-Nov 13-Nov 20-Nov 27-Nov 04-Dec 11-Dec 11-Sep 18-Sep 25-Sep 02-Oct 09-Oct 16-Oct 23-Oct 30-Oct 06-Nov 13-Nov 20-Nov 27-Nov 04-Dec 11-Dec Relative humidity (%) AAggrriiccuullttuurree 22002222,, 1122,, 4x3 FOR PEER REVIEW 6 6ooff 114 3 45 (a) E1 Afayak Jenguma 40 E2 E3 35 30 25 20 15 10 10 20 30 40 50 10 20 30 40 50 Days After Emergence (DAE) (b) 12 E1 Afayak Jenguma E2 10 E3 8 6 4 2 0 10 20 30 40 50 10 20 30 40 50 Days After Emergence (DAE) (c) 40 E1(36°C & 55%) Afayak Jenguma E2 (34°C & 57%) E3 (33°C & 44%) 30 20 10 0 10 20 30 40 50 10 20 30 40 50 Days After Emergence (DAE) Figure 2. Diiffferrencess iin (a) plant height and (b) node appearance (c) leaf appearance for the two variietiies undeerr vvaarryyiinngg eennvviriroonnmmeennt t(E(E11: :363 6°C◦,C R, HR:H 5:55%5;% E;2E: 324: °3C4, ◦RCH, :R 5H7%: 5; 7a%nd; aEn3d: 3E33 :°C3,3 R◦HC:, 4R4H%:)4. 4%). 3.3. PAaltttheronusgohf Bthioemrea sws aAsc ncuom duiflfaetiroennce in the total number of nodes formed for both vari- eties Pinla anltls tihnreEe2 erencvoirdoendmtehneths,i gthes rtamtee aonf nt otdale darpypweaerigahntce(T dDifWfe)re(pd<. T0h.0e5 t)ootfa2l .n3u9mg/bpelra onft laetatvhees efonrd Aoffaythake ivne gEe1t,a Eti2v eansdta gEe3 wheirle 3th5,o s2e5 ianndE 132re, croersdpedctiaveTlyD,W at o5f0 1D.5A1Eg,/ wplhainlet J(eTnagbulem2a) .hAadt t3h7e, 3fl5o awnedr i4n1g fsotra gEe1,, En2 vainrodn Em3e, nretsEp3echtaivdetlhy.e highest total dry biomass with mean weights of 5.22 and 6.04 g/plant for both Afayak and Jenguma varieties, respectively 3(T.3a.b Pleat2t)e.rTnsh eofT BDioWmaosfs tAheccAufmayualkatvioanr iety in E3 was significantly higher than those in E1 and E2 buPtlannotss iginn iEfi2c arnetcodridffeedre nthcee shwigehreesot bmseeravne dtobteatlw dereyn wEe1iganhdt (ET2DwWh)i l(epJ e