Acta agriculturae Slovenica, 117/2, 1–12, Ljubljana 2021 doi:10.14720/aas.2021.117.2.2082 Original research article / izvirni znanstveni članek Maize seedling emergence in response to climatic variability in a tropical rainforest area Chris Adegoke FAYOSE 1,2, * and Morakinyo Abiodun Bamidele FAKOREDE 2 Received February 25, 2021; accepted April 12, 2021. Delo je prispelo 25. februarja 2021, sprejeto 12. aprila 2021. Maize seedling emergence in response to climatic variability in a tropical rainforest area Abstract: Environmental factors causing low seedling emergence often observed in tropical maize (Zea mays L.) are poorly documented. This study was conducted to investigate the effects of weather factors on maize seedling emergence at the Obafemi Awolowo University Teaching and Research Farm (OAU TRF). Five maize varieties sown weekly, in 3-replicate RCBD experiments throughout the 2016 and 2017 cropping seasons, were used to monitor emergence percentage (E %), emergence index (EI) and emergence rate index (ERI). Climatic data were obtained from the automatic weather station located on the farm. Analysis of variance revealed highly significant (P ≤ 0.01) environmental effect for all traits. Soil moisture (Sm), relative humidity, air temperature, heat unit and soil heat flux (SHF) showed significant (P ≤ 0.05) correlation coefficients with all traits, but there was no relationship between the emer- gence traits and grain yield. Stepwise multiple regression and sequential path coefficient analyses indicated that increased Sm, rather than rainfall per se, increased the speed of emer- gence. Minimum air temperature and SHF with direct effects, and heat unit with indirect effect, negatively affected emergence the most. Relatively low Tmin and SHF, along with just enough Sm maximized seedling emergence in the rainforest agro-ecol- ogy of southwestern Nigeria. Keywords: climatology; crop phenology; growth analysis; seedling vigour; Zea mays L. 1 Department of Agricultural Technology, Ekiti State College of Agriculture and Technology, Isan Ekiti, Nigeria. 2 Department of Crop Production and Protection, Obafemi Awolowo University, Ile-Ife, Nigeria. * Corresponding author: +2348060365729, adegoke_chris@yahoo.com Vpliv sprememb podnebja na vznik kalic koruze na območju tropskega deževnega gozda Izvleček: Okoljski dejavniki, ki vplivajo na slab vznik koruze (Zea mays L.) so v tropskih razmerah pogosto opa- ženi a slabo dokumentirani. Ta raziskava je bila izvedena za preučevanje učinkov dejavnikov podnebja na vznik koruze na univerzitetnem učnem in raziskovalnem posestvu Obafemi Awolowo (OAU TRF), Nigerija. Pet sort koruze je bilo pose- janih tedensko s tremi ponovitvami v popolnem naključnem bločnem poskusu (RCBD) v rastnih sezonah 2016 in 2017. Pri tem so bili spremljani odstotek kalitve (E %), indeks vznika (EI) in hitrost vznika (ERI). Klimatski podatki so bili pridobljeni iz avtomatske vremenske postaje na posestvu. Analiza variance je pokazala zelo značilne (P ≤ 0,01) okoljske vplive na vse opa- zovane lastnosti. Vlažnost tal (Sm), relativna vlažnost zraka, temperatura zraka, privzem toplote in tok toplote v tleh (SHF) so imeli značilno korelacijo (P ≤ 0,05) z vsemi opazovanimi la- stnostmi, ugotovljena pa ni bila nobena povezava med lastno- stimi kalitve in pridelkom zrnja. Stopenjska multipla regresija in kasnejša analiza posrednih in neposrednih vplivov na kalitev sta pokazali, da je imela nanjo večji vpliv vlažnost tal, ki je po- večala hitrost kalitve kot pa sama količina padavin. Minimalna temperatura zraka in dotok toplote v tla sta imela največji nepo- sredni vpliv na kalitev. Nizka temperature tal (Tmin) in majhen dotok toplote v tla (SHF) sta ob ravno zadostni vlažnosti tal (Sm) maksimalizirali vznik koruze v agro-ekoloških razmerah deževnega gozda jugozahodne Nigerije. Ključne besede: klimatologija; fenologija poljščin; rastna analiza; vigor sejank; Zea mays L. Acta agriculturae Slovenica, 117/2 – 2021 2 C. A. FAYOSE, M. A. B. FAKOREDE 1 INTRODUCTION The demand for maize (Zea mays L.) grains in sub- Saharan Africa (SSA) has continued to increase because of its importance in human food, livestock feed, and in- dustrial raw material. Maize accounts for more than 15 % of the total calorie intake by the population in SSA, along with its use in animal feed production and as raw ma- terial in some agro-allied industries. As noted by Talabi et al. (2017), better adaptation of maize crop to various agro-ecologies, responsiveness to fertilizer application, and relative ease of cultivation, processing and storage has greatly increased its popularity in West and Central Africa (WCA) over other traditional cereal crops, such as sorghum [Sorghum bicolor (L.) Moench] and pearl millet (Penisetum typhoides L.). There is, therefore, an urgent need for increased maize production in these regions to meet the constant demand for its grains. Environmental factors such as poor soil conditions, especially decreasing nitrogen level, drought, increased urbanization, poor or inaccurate agronomic practices, and poor climatic con- ditions have affected maize production, limiting it from attaining the desired production levels, despite maize crop improvements over the years from active breeding programs. FAO (2016) puts maize production in Nigeria at about 11 MT/year compared to 384 MT/year obtained in the USA. A combination of improved varieties, im- proved agronomic practices and production intensifica- tion, including adaptation of maize production to climate through adequate knowledge of crop response to the var- iability of climatic conditions throughout the life cycle of the crop will be necessary to close this huge yield gap to a reasonable level. High emergence percentage followed by high seedling vigour and vigorous vegetative growth are necessary conditions for high final stand of the crop on the field which, in turn, leads to higher grain yield. Maize, for all its positive features, depends on the environment for its survival; the climatic conditions constitute a very important component of the environ- ment. Unfortunately, the information available on maize crop response to climate is very few and fewer still at the seedling stage, especially in the rainforest agro-ecology of Nigeria, and where such information exists, the depth and therefore the accuracy is often in question. This is due to the the use of rather outmoded climatology items of equipment for such studies compared with improved equipment with greater accuracy used in the developed world (Fakorede and Opeke 1985). Furthermore, there are often inadequacies in the number of climatic varia- bles analyzed, or statistical analyses used, or both. Conse- quently, moisture availability has generally been seen as the most important variable influencing maize growth at seedling and flowering stages, even though there is pau- city of information about maize crop response to other climatic variables such as temperature, heat unit and soil heat flux at the seedling stage of maize growth. Awo- sanmi et al. (2016) found significant effect of moisture stress on tropical maize seed yield in the rainforest ecol- ogy of Southwest Nigeria. Oke (2016) found that yield had significant positive correlations with total rainfall and air mean relative humidity and negative correlations with air temperature extremes in similar climate. Tunde et al. (2011) also found maize production to be highly in- flenced by rainfall, air relative humidity, number of rainy days and air temperature at Ilorin, a Southern Guinea sa- vanna location in Nigeria. Ammani et al. (2012) found that drought occuring at about one week before flower- ing to the grain filling stage caused significant reduction in yield when they evaluated the relationship between rainfall and maize crop production in the savannah ecol- ogies. These studies have involved few climatic variables, mostly employed correlation for data analysis and none actually investigated crop response to climatic factors at the seedling stage. The study by Fakorede and Opeke (1985) employed several climatic variables and statistical analysis methods to elucidate the response of maize yield to weather factors in the rain forest of Nigeria, typified by Ile-Ife. They found significant negative correlation coef- ficients of effective rainfall with grain yield and air rela- tive humidity even though the study was limited by the rather outmoded climatological equipment used. Studies conducted in advanced countries, such as that by Jong et al. (1982) concluded that solar radiation was the sin- gle most influencial climatic factor affecting yield and its components in maize even though only air temperature and solar radiation were the focus of analysis in the study. Dimpsey (1995) found 16 °C to 35 °C as the optimum germination range for maize seeds when he studied the effect of air temperature on the germination of different seeds in Australia. Alm et al. (1993) found a positive cor- relation between air temperature (from 10 °C to 25° O C) and seedling elongation of maize and soybean (Glycine max L). Rainfall is the primary climatic factor that deter- mines the timing of most agronomic practices, especially planting operations in Nigeria. False start of rainfall and unpredictable frequent occurrence of short duration drought during the growing season, both of which have become more severe over time, are now features of the rainforest agro-climatic zone of SW Nigeria typified by Ile-Ife, a location at 7 o 28’ N, 4 o 33’ E and 244 m asl in the zone (Fayose and Fakorede, 2021). Such weather anoma- lies have been attributed to climate change (Fakorede and Akinyemiju, 2003). The year 2020 received one of the smallest total rainfall amounts (˂ 700 mm) in recent Acta agriculturae Slovenica, 117/2 – 2021 3 Maize seedling emergence in response to climatic variability in a tropical rainforest area times, next to just above 700 mm received in 2017 at the location. Low amount of rainfall was received in June and July when frequent, heavy rainfall is normally expected. Reduced seedling emergence and low vigor, are usually associated with the weather anomalies that characterize maize especially those planted early in the zone. This has often resulted in missing stands, and the situation has been aggravated by climate change in recent times. It is therefore necessary to have adequate knowledge of crop response to climatic variables at all stages in order to be able to cope with the negative impacts of climate change. The primary objective of this study was to identify the climatic factors influencing seedling emergence of maize under the natural climatic conditions of the rain- forest agro-ecology of SW, Nigeria. A secondary objec- tive was to determine the relationship of the seedling emergence traits with grain yield of maize. 2 MATERIAL AND METHODS The study was conducted at the Teaching and Re- search Farm of Obafemi Awolowo University, Ile-Ife (OAU TRF) in the 2016 and 2017 early and late cropping seasons. OAU TRF is located at 7 o 28’ N, 4 o 33’ E and 244 m above sea level in the marginal areas of the rainfor- est agro-ecology of South Western Nigeria. As a typical rainforest agro-ecological location, Ile-Ife is character- ized by two contrasting rainy seasons; the first or early season from about March/April to July and the second or late season from late August to October with a period of short dry spell that occurs in July/August, popularly re- ferred to as the “ August break” or “ August dry spell” . T otal annual range of 740 mm to 2040 mm has been reported at the location, along with a mean maximum tempera- ture range of 29 o C to 32 o C (Fakorede and Akinyemiju, 2003). Five maize varieties adapted to the tropical rain- forest environments were sown in three replicate-rand- omized complete block design. They include four open- pollinated varieties White DT STR SYN1.- TZL Comp. 1 – W, ACR 94 TZE Comp 5 C 3 , TZL Comp. 4 DT F 2 , TZL Comp. 1 C6/DT – SYN – 1 – W all of which were developed at the International Institute of Tropical Ag- riculture (IITA), Ibadan; and a hybrid, ‘Oba Super 1’, obtained from Premier Seeds, Zaria. All the varieties are white-grained, high yielding and have been released for commercial production in Nigeria and several other West and Central African (WCA) countries. The maize was sown weekly from March to Novem- ber each year. However, there were some weeks when sowing could not be done due to some extraneous fac- tors; therefore, data from 54 dates (DOS) were analysed for seedling emergence and vigour and 42 dates were analysed for grain yield. Each plot contained four or six rows, which were 5 m long spaced 0.75 m apart; within row spacing was 0.5 m. Each plot was about 15 m 2 and 22.5 m 2 for the four and six-row plots, respectively. Three seeds were sown per hill. Prior to sowing, the experimen- tal land was ploughed and harrowed, and the seed were treated with thiamethoxam, mefenoxam (metalaxyl-M) and difenoconazole, to control damage by soil-borne diseases and insect pests. Weeds were controlled by the application of post-sowing and pre-emergence of maize with primextra, which contains atrazine (2-chloro-4- (ethyl amino)-6-isopropylamino-s-triazine) and alachlor (N-(methyl-2-methoxy-ethyl)-2-ethyl-8-methyl-chlo- roacetanilide) as active ingredients at the rate of 5 l/ha. Emergence counts were made daily from five to nine days after sowing (DAS); from which emergence percentage (E%), emergence index (EI), and emergence rate index (ERI) were computed as follows (Fakorede and Agbana 1983): Thinning was done immediately after emergence count at 9 DAS to two plants per stand giving an esti- mated plant population density of 53, 333 plants ha -1 . Fertilizer was applied immediately after thinning at the rate of 60 kg ha -1 each for N, P 2 O 5 and K 2 O. Further weed control was done using paraquat (N,N′-dimethyl-4,4′- bipyridinium dichloride), carefully applied as a post- emergence, non-selective and contact herbicide at the rate of 3.0 l ha -1 . Data collected on grain yield were adjusted to 15 % moisture content. Minimum and maximum air tem- peratures and air relative humidity, rainfall, solar radia- tion, net radiation, wind speed, soil temperatures and soil heat flux were monitored at the automatic weather station (AWS), OAU MET Station, of the Atmospheric Physics Research Group located on the OAU TRF. The experimental plots were within 100 – 500 meters of the weather station mostly in plain sight except for the last four environment in 2016 that had a slightly dense veg- etation inbetween. Most sensors (Table 1) at the station were manufactured by Campbell Scientific Inc., USA. Acta agriculturae Slovenica, 117/2 – 2021 4 C. A. FAYOSE, M. A. B. FAKOREDE Table 1: Weather tracking sensors and their level of accuracy and output units at the automatic weather station of the OAU MET Station within the OAU TRF. Weather variable Tracking sensor Accuracy Output unit Solar radiation CS 300 Pyranometer ± 5 % W m -2 Net radiation NR-LITE Radiometer ± 5 % W m -2 Air temperature/ Relative Humidity HMP45 Temperature and Humidity Probe ± 0.4 o C ± 2 – 3 % o C (Temperature) % (Relative humidity) Soil temperature Model 108 Soil Temperature Probe ± 0.3  o C to 0.7  o C o C Soil heat flux Model HFP01 Soil Heat Flux Plate -15 to 5 % Wm -2 Wind speed A100L Cup Anemometer ± 0.1 m s -1 m/s Rainfall TE525 Tipping Bucket Rain Gauge ± 1 % (for up to 1 inch per hour) -3 to 0 % (for 1 to 2 inches per hour) -5 to 0 % (for 2 to 3 inches per hour) mm They are popular for designing weather tracking systems that operate in vast array of climates including those specific for the tropical hot and humid climate with high level of sensitivity and precision. Heat unit was computed from the minimum and maximum tempera- ture per day as given below: where X i L is the daily minimum air temperature ( o C), X i H is the daily maximum air temperature ( o C) (X i H = 30 if X i H > 30 o C, X i H = X i H if X i H ≤ 30 o C), and 10°C is the base temperature (Abasi et al., 1985). Prior to analysis, percentage data were transformed using the square root method; while soil moisture data were subjected to log (log e ) transformation. Analysis of variance was done on all data using PROC GLM of Sta- tistical Analysis System (SAS, 2000). The linear additive model for the ANOV A was: Y ijk = µ + α i + β j(i) + ƛ k + αƛ (ik) + ɛ ijk , in which Y ijk is the observed measurement of the k th genotype grown in the j th rep under the i th environment; µ is the grand mean; α i is the main effect of the i th environ- ment, i = 1,2,…..,42 or 54; β j(i) is the effect of the j th repli- cation nested within the i th environment, j = 1,2,3; ƛ k is the effect of the k th genotype, k = 1,2,….5; αƛ (ik) is the first order interaction of the i th environment with the k th geno- type, and ɛ ijk is the random error (residual) term. Further- more, correlation and regression analyses were done to determine the trends in maize seedling response to weather variables. Emergence traits were regressed on weather variables in linear model: Y = a + bX and poly- nomial model: Y = a + b 1 X 1 +....bnX n , Y and X are the emergence traits and weather variable(s), respectively; a and b are the intercept and regression coefficient, respec- tively; n is the order of the polynomial (quadratic, cubic, quartic). Correlation analysis was done where the coeffi- cient . Path analysis is an extension of multiple regression, particularly step-wise multiple regression analysis which, along with sequential path diagrams, was employed to elucidate the cause and effect relationships among traits.  The Statistical Package for the Social Sciences (SPSS Inc, 2007) was used for the stepwise regression analy- ses to provide information on the path coefficients and the causal relationships required for the path diagrams. The procedure, which was described by Mohammadi et al. (2003), has been used by Badu-Apraku et al. (2014) and Talabi et al. (2017). The predictor variables which, in this case, were weather factors, were organized into first, second, and third order, based on their contributions to the total variation in emergence traits with minimized multi collinearity. To perform the stepwise regression analysis, emergence traits were regressed on climatic variables to identify the ones with significant contribu- Acta agriculturae Slovenica, 117/2 – 2021 5 Maize seedling emergence in response to climatic variability in a tropical rainforest area Table 2: Mean squares from the ANOV A of seedling emergence traits of five maize varieties evaluated in 54 environments at the OAU TRF in 2016 and 2017 seasons. ***- F statistic significant at 0.001 level of probability. CV- Coefficient of variation R 2 - Coefficient of determination † - E %_5 = emergence percentage at 5 days after sowing (DAS), E%_7 = emergence percentage at 7DASS, E %_9 = emergence percentage at 9DAS, EI = emergence index and ERI = emergence rate index. tions to the variation in the traits at p ≤ 0.05, and they were categorized as first order variables. The first-order variables thereafter were each regressed on other climatic variables which were not in the first order category, to identify the climatic variables with significant contribu- tions to emergence traits through the first-order varia- bles. These variables were classified as second order vari- ables. The same procedure was repeated to identify third order variables(s) and so on. The path coefficients were obtained from the standardized b-values of the stepwise regression analysis. The path coefficients were tested for significance using the standard errors at 0.05 probability level. Only traits having significant path coefficients were retained in each order. 3 RESULTS Results from the combined ANOVA for grain yield and seedling emergence traits revealed highly significant effects (p ≤ 0.01) for all sources of variation; that is, en- vironment, variety and E x V interaction (Table 2). The effect of replication within environment was also highly significant for all traits except ERI. The coefficient of var - iation (CV) ranged from about 7 % for EI to about 81 % for ERI. Fairly high R 2 value was observed for grain yield Sources DF E %_5 † E %_7 E %_9 EI (Days) ERI (Days) DF Yield (t ha -1 ) Env (E) 53 2404.88*** 4527.06*** 3256.58*** 3.55*** 941.65*** 41 7.61*** Rep/Env 108 100.42*** 324.17*** 268.92*** 0.57*** 127.98 84 0.49*** Variety (V) 4 7684.37*** 15343.14*** 14332.64*** 7.33*** 6680.45*** 4 4.76*** E x V 212 201.47*** 619.56*** 607.57*** 0.37*** 475.86*** 164 0.45*** Error 432 39.02 65.54 69.59 0.21 116.46 336 0.22 Total 809 282.58 613.08 516.49 0.56 298.69 629 0.83 CV ,% 30.01 12.89 11.97 6.90 81.11 30.67 R 2 0.93 0.94 0.93 0.80 0.79 0.86 (86 %) and higher still for E % at 5, 7 and 9 DAS (93 to 94 %), while ERI had the lowest R 2 value. Rainfall had no significant correlation coefficients (r-values) with seedling emergence traits, unlike soil moisture (Sm) and air relative humidity (RH) which, in most cases, had statistically significant r-values with the traits (Table 3); the higher the Sm and RH, the higher the emergence percentage, and the faster the rate of emer- gence. Similarly, radiation had no significant r-values with seedling vigour traits, whereas air temperature and heat units (HU) were significantly correlated with the emergence traits; the higher the air temperature and HU, the lower the emergence percentage, and the slower the rate of emergence. A third group of sharp contrasts in- volved soil temperatures which, except in only two cases, were not significantly correlated with the emergence traits whereas the higher the soil heat flux (SHF), the lower the emergence percentage and rate of emergence (Table 3). As expected, wind speed had no relationship with emergence traits, although it showed a surprising significant negative relationship (r = -0.38 at p ≤ 0.05) with ERI. Highly significant correlations were also ob- served among all seedling emergence and vigour traits but none had significant correlation with grain yield (Ta- ble 4). Acta agriculturae Slovenica, 117/2 – 2021 6 C. A. FAYOSE, M. A. B. FAKOREDE Table 3: Correlation coefficients of seedling emergence traits of five maize varieties with climatic variables obtained for 54 environ- ments at the OAU TRF in 2016 and 2017. Table 4: Correlation coefficients of seedling emergence traits of five maize varieties with grain yield of 40 environments at the OAU TRF in 2016 and 2017. *,**- Significance at 0.05 and 0.01 level of probability, respectively. † - See Table 1. *,**- Significance at 0.05 and 0.01 level of probability, respectively. † - See Table 1. global solar radiation, net radiation and soil heat flux are in W m -2 ; temperature is in o C, soil temperatures 1 – 5 are soil temperatures at different levels (2 cm, 5 cm, 10 cm, 20 cm and 50 cm for soil tem- peratures 1, 2, 3, 4 and 5, respectively), soil moisture is in m 3 /m 3 , rainfall is in mm, windspeed is in m/s, relative humidity is in %. E%_5 † E%_7 E%_9 EI ERI Soil temperature 1 0.16 -0.11 -0.05 0.06 -0.03 Soil temperature 2 0.08 -0.22 -0.16 0.13 0.11 Soil temperature 3 0.09 -0.26 -0.19 0.11 0.06 Soil temperature 4 0.06 -0.33* -0.26 0.15 0.08 Soil temperature 5 0.03 -0.37* -0.30 0.17 0.09 Soil heat flux -0.49** -0.64** -0.77** 0.33* 0.44** Soil moisture 0.38* 0.62** 0.54** -0.43** -0.48** Rainfall 0.01 0.22 0.20 0.01 -0.24 Wind speed 0.04 0.19 0.27 -0.06 -0.38* Mean air temperature -0.41** -0.79** -0.76** 0.38* 0.43** Minimum air temperature -0.48** -0.82** -0.80** 0.38* 0.44** Maximum air temperature -0.39* -0.67** -0.64** 0.37* 0.33* Heat unit -0.47** -0.81** -0.80** 0.36* 0.44** Mean relative humidity 0.30 0.66** 0.65** -0.30* -0.42** Minimum humidity 0.04 0.35* 0.31* -0.16 -0.26 Maximum humidiy 0.48** 0.67** 0.67** -0.3 -0.34* Net radiation 0.13 -0.07 -0.09 0.07 -0.06 Mean global radiation 0.09 -0.16 -0.09 0.03 -0.07 Total global radiation 0.09 -0.15 -0.10 0.02 -0.06 Mean E_5 † Mean E_7 Mean E_9 EI ERI Yield (t ha -1 ) Mean E_5 1 0.70** 0.63** -0.83** -0.68** 0.21 Mean E_7 1 0.90** -0.70** -0.78** 0.24 Mean E_9 1 -0.43** -0.80** 0.26 EI 1 0.58** -0.10 ERI 1 -0.16 Yield (t/ha) 1 Acta agriculturae Slovenica, 117/2 – 2021 7 Maize seedling emergence in response to climatic variability in a tropical rainforest area Table 5: Regression coefficients (b-values), coefficients of determination (R 2 ) and change in R 2 (ΔR 2 ) from the stepwise multiple regression of maize seedling emergence traits on climatic variables at the OAU TRF in 2016 and 2017. † - See Table 1. ɸ - See Table 2 Climatic variable b-value R 2 ΔR 2 E%_5 † Soil heat flux ɸ -0.43 0.24 0.24 E%_7 Minimum air temperature -20.92 0.67 0.67 Soil moisture 34.22 0.71 0.04 E%_9 Minimum air temperature -6.52 0.62 0.62 Soil heat flux -0.56 0.66 0.04 EI Soil moisture -8.06 0.2 0.2 ERI Soil moisture -154.57 0.23 0.23 Wind speed -13.96 0.37 0.14 Average radiation -0.07 0.43 0.07 Minimum humidity -1.41 0.59 0.16 Rainfall 0.08 0.63 0.04 Results from the stepwise multiple regression (Table 5) showed that only SHF had a significant effect on E % at 5 DAS. The simple linear regression of E % at 5 DAS on SHF produced the prediction equation: Ŷ = 19.44 – 0.46X; (R 2 = 0.24). This means that for every unit in- crease in SHF, emergence decreased by 0.46 % with a rather low, though statistically significant R 2 value. Step- wise multiple regression analysis showed minimum air temperature (X 1 ), followed by Sm (X 2 ) as the most impor- tant climatic factors controlling E % at 7 DAS. The multi- ple linear regression produced the prediction equation Ŷ = 34.61 – 20.92X 1 + 34.22X 2 ; (R 2 = 0.71) with minimum temperature accounting for 67 % of the 71 % variations captured by the regression model (Table 3). Emergence % at 9 DAS was significantly influenced by minimum air temperature (X 1 ) and SHF (X 2 ) with the prediction equa- tion Ŷ = 169.37 – 4.12X 1 – 0.51X 2 ; R 2 = 0.66. Stepwise multiple regression also revealed that Sm was the only environmental factor that interacted with EI (Ŷ = 7.89 – 7.91X, R 2 = 0.2); while Sm (X 1 ), windspeed (X 2 ), average global radiation (X 3 ), minimum RH (X 4 ) and rainfall (X 5 ) significantly influenced ERI according to the equation Ŷ = 198.79 – 158.25X 1 – 9.86X 2 – 0.29X 3 – 1.64X 4 – 0.08X 5 (R 2 = 0.63). The summary of the influence of the climatic variables on the seedling emergence traits as obtained from stepwise multiple regression is contained in Table 5. Although Sm had direct effect on E % at 7 DAS, tem- perature and the heat changes in the soil (SHF) seemed to have exerted greater influence on E % than soil mois- ture availability (Figures 1 – 4), because SHF had a direct effect on E % at 5 DAS, while minimum air temperature influenced E % at 7 DAS alongside Sm. This is the only significant effect (direct or indirect) of Sm on E %. Fur- thermore, E % at 9 DAS was again directly influenced by SHF and minimum air temperature. The secondary cli- matic variables that interacted with the primary variables to affect E % were mostly air temperature and radiation. Acta agriculturae Slovenica, 117/2 – 2021 8 C. A. FAYOSE, M. A. B. FAKOREDE Figure 1: Sequential path-coefficient analysis diagram of climatic variables affecting emergence percent at 5 DAS. One directional arrows indicate direct effects while double arrows are correlation coefficients. Figure 2: Sequential path-coefficient analysis diagram of climatic variables affecting emergence percent at 7 DAS. One directional arrows indicate direct effects while double arrows are correlation coefficients. Figure 3: Sequential path-coefficient analysis diagram of climatic variables affecting emergence percent at 9 DAS. One directional arrows indicate direct effects while double arrows are correlation coefficients. Acta agriculturae Slovenica, 117/2 – 2021 9 Maize seedling emergence in response to climatic variability in a tropical rainforest area Figure 4: Sequential path-coefficient analysis diagram of climatic variables on emergence index. One directional arrows indicate direct effects while double arrows are correlation coefficients. Figure 5: Response of emergence percentage to minimum air temperature for five maize varieties evaluated in 54 environments at the OAU TRF in 2016 and 2017 cropping seasons. Figure 6: Response of emergence percentage to soil heat flux for five maize varieties evaluated in 54 environments at the OAU TRF in 2016 and 2017 cropping seasons. Acta agriculturae Slovenica, 117/2 – 2021 10 C. A. FAYOSE, M. A. B. FAKOREDE However, Figure 4 showed that EI, which represents the speed of emergence, was directly influenced by Sm. The secondary (indirect) effect of SHF (P = -0.310) was again observed on EI alongside mean RH (P = 0.332) and rainfall (P = 0.322) (Figure 4). Emergence was found to occur at minimum air temperature range of 20 o C to 26 o C; and SHF range of -25 to 16 W m -2 but the optimum range was 21 to 22 o C for Tmin and -25 to 0 W m -2 for SHF (Figures 5 and 6). 4 DISCUSSION The objective of this study was to investigate the ef- fect of climatic factors, which are important factors of the environment, on maize seedling emergence in the rainforest agro-ecology of Southwestern Nigeria as typi- fied by OAU TRF. Results of the ANOV A, showed highly significant mean squares for the environment source of variation for all seedling traits, accounting for up to 56 % of the total sum of squares in some cases. This suggests that the environment, which greatly influenced seedling emergence of maize varieties in this and other studies (see inter alia Fakorede, 1984; Fakorede and Agbana, 1983) needs serious research attention. Emergence % in arable crops generally respond negatively to higher temperature, heat unit and soil heat flux (SHF); and positively to soil and air humidity. In the present study, E % had significant positive correlation with soil moisture but not with rainfall. This was con- trary to expectation because rainfall was the only source of moisture at the location of this study and, therefore, the only source of the moisture available in the soil. How- ever, other climatic factors with considerable influence on soil moisture such as temperature, humidity and heat flux as well as soil properties, such as the soil structure and water holding capacity, may have played a part in the observed results. It is not the amount of water received through rainfall, but the amount of soil moisture avail- able for plant growth that is important. It would seem, therefore, that climatic factors that influence the amount of water retained in the soil, such as humidity and tem- perature, are more important to crops than the amount of water that enters the soil through rainfall. Emergence index and ERI generally showed an opposite trend to that observed in E %. This is expected because the negative correlations of EI and ERI with E % has been widely re- ported in the literature (Fakorede and Ojo, 1981; Fakore- de and Agbana, 1983; Fayose and Fakorede, 2014). The relatively weaker correlations (compared to temperature variables) and the vastly low R 2 value where there was significant effect of Sm on seedling growth, suggests that Sm is not the most important environmental factor af- fecting maize seedling emergence in the rainforest. Con- trary to the general belief, temperature and heat transfer in and around the seedlings could be more important if they reach a threshold (minimum air temperature above 23 °C, and soil heat flux above 0 W m -2 in this study) ir- respective of the amount of moisture available in the soil. Spurious correlation is also common where climatic variables are involved because of the complex interac- tions that exist among them. Fakorede and Opeke (1985) described spurious correlation as a relationship between two variables, a and b, in which r ab results largely from the fact that a varies along with some other variable c which, indeed, is the true predictor of b. This is an aspect where most of the previous studies fell short as they de- pended largely on correlation for data analysis and inter- pretation. Therefore in this study, path coefficient anal- ysis was used in addition to correlation and regression analyses to determine the true effect of climatic factors on maize seedling emergence. Results of path analyses implicated minimum air temperature and soil heat flux as influencing emergence much more than soil moisture per se. The two air temperature variables directly influ- enced E % negatively at 9 DAS and, in addition, soil heat flux had negative direct impact on E % at 5 DAS. This suggest that heat transferred at night caused an effect on maize growth at the early stage. Perhaps, the soil tem- perature gradient created by low air temperature caused the soil to radiate the heat accumulated during the day, thereby causing an effect on the maize seedlings in the process. The thermal exchange processes at the soil sur- face is dominated by the meteorological conditions and occur by radiation, conduction, and convection, with or without phase changes. The thermal soil properties are strongly dependent on water content and many processes occurring in soils are strongly influenced by temperature (Tokoro et al., 2016). Soil moisture had a direct positive impact on E % at 7 DAS even though the negative effect of minimum air temperature was even more pronounced. Results also revealed that, although soil moisture had a negative direct effect on EI, the ambit of influence might be limited because of the low R 2 (about 20 %), with soil heat flux negatively influencing EI at the secondary level of interaction. This suggests that increased soil moisture might speed up the rate of emergence if other variables were not interacting. It is a known fact that moisture is necessary for germination. In the present study, seeds sown in March for the first environment in 2017 stayed in the soil for up to three weeks and only emerged following the first rainfall. Unfortunately, climatic variables are in con- stant interaction; therefore, the benefits of moisture for seedling growth is seemingly masked when it is not at a critical level. Also, 2017 had poorer seedling growth Acta agriculturae Slovenica, 117/2 – 2021 11 Maize seedling emergence in response to climatic variability in a tropical rainforest area performance compared to 2016, a result attributable to the air temperature/soil heat flux interaction, because of the smaller rainfall in 2017 (675 mm in 2017 vs more than 1000 mm in 2016 ) and consequently, higher air temperature (average value of 28.5 o C in 2017 vs 25.8 o C in 2016). One could theorize, therefore, that soil heat transfer occurring at night raises night air temperature around maize seedlings, thereby negatively affecting the seedlings, even though Sm remains at an adequate level; or that seedling growth will proceed at an optimum level with Sm level that is just sufficient, as long as soil heat and heat transfer are minimized especially in the early hours of the day. It is also noteworthy that most of the variables that affected E % indirectly via the first order variables were air temperature and radiation. Even though there was no direct relationships of the seedling emergence and vigour traits with yield, in agreement with some ear- lier studies (Fayose and Fakorede, 2014; Fakorede and Agbana, 1983; Fakorede and Ojo, 1981), it is however incontrovertible, that the seedling traits strongly influ- ence yield, albeit, indirectly. For instance, severe and prolonged water stress at the seedling stage may damage the structure of the photosynthetic membrane resulting in lower chlorophyll content and thus, low radiation use efficiency (RUE, Song et al., 2019). Maize plant with such damage often did not show meaningful recovery irre- spective of the amount of moisture supplied at the latter stages. Therefore, unrecoverable yield loss could occur if adequate attention is not paid to growth at the seedling stage. Also, high final stands, tolerance to infestation by pests and diseases, less likelihood of lodging, increased efficiency of interception of solar radiation for photosyn- thesis, and reduced need for weed control (when cano- pies touch) are some of the strong benefits of ensuring that climatic conditions are favourable for maize at the seedling stage that have been well documented in the lit- erature (Li et al., 2015; Fayose et al., 2021). 5 CONCLUSIONS Despite the wide recognition of the importance of soil moisture to seed germination and seedling emer- gence, soil heat flux, minimum air temperature and the interactions between them were more important, al- though Sm favours higher speed of emergence. Emer- gence occurred at minimum air temperature range of 20 o C to 26 o C; and SHF range of -25 to 16 W m -2 but the optimum air Tmin and SHF for emergence were 21 to 22 o C and -25 to 0 W m -2 , respectively. Also, soil moisture, rather than rainfall, is the important factor for maize seedling emergence, despite the general recogni- tion of rainfall as an important factor for agriculture in the rainforest ecologies. Consequently, any agronomic practices that would conserve soil moisture would be of benefit to maize at the seedling stage. 6 REFERENCES Abasi, L., Fakorede, M. A. B. and Alofe, C. O. (1985). Compari- son of heat units and calendar days for predicting silking dates in maize in a tropical rainforest location. Maydica, 30, 15–30. https://doi.org/10.1017/S0014479700012217 Alm, D.M., Stoller, E.W. and Wax, L.M. (1993). An index model for predicting seed germination and emergence rates. Weed Technology, 7, 560–69. https://doi.org/10.1017/ S0890037X00037349 Ammani, A. A., Ja’afaru, A. K., Aliyu, J. A. and Arab, A. I. (2012). Climate change and maize production: empirical evidence from Kaduna State, Nigeria. Journal of Agricultural Extension, 16(1), 1-8. https://doi.org/10.4314/jae.v16i1.1 Awosanmi, F. E., Ajayi, S. A. and Menkir, A. (2016). Impact of moisture stress on seed yield in tropical maize. Interna- tional Journal of Agricultural Innovation and Research, 4(6), 1033-1038. Badu-Apraku, B., Akinwale, R. O. and Oyekunle, M. (2014). Ef- ficiency of secondary traits in selecting for improved grain yield in extra-early maize under Striga-infested and Striga- free environments. Plant Breeding. doi:10.1111/pbr.12163. https://doi.org/10.1111/pbr.12163 Dimpsey, R. (1995). Sweet corn production. Agriculture notes. State of Victoria, Department of Primary Industries, Aus- tralia. Fakorede, M.A.B. (1985). Response of maize to planting dates in a tropical rainforest location. Experimental Agriculture, 21, 19-30. https://doi.org/10.1017/S0014479700012217 Fakorede, M. A. B. and Akinyemiju, O. A. (2003). Climate change: effects on maize production in a tropical rainfor- est location. P . 272-282. In Badu-Apraku, B., Fakorede, M. A. B., Ouedraogo, M., Carsky, R.J. and Menkir, A. (Eds.). Maize revolution in West and Central Afica. Proceedings of a Regional Maize Workshop, IITA-Cotonou, Benin Repub- lic. 14-18 May 2001. WECAMAN/IITA. Fakorede, M. A. B. and Ojo, D. K. (1981). Variability for seed- ling vigour in maize. Experimental Agriculture, 17, 195-201. https://doi.org/10.1017/S0014479700011455 Fakorede, M. A. B. and Agbana, S. B. (1983). Heterotic effects and association of seedling vigor with mature plant char- acteristics and grain yield in some tropical maize cultivars. Maydica, 28, 327-338. Fakorede, M. A. B and Opeke, B. O. (1985). Weather factors affecting the response of maize to planting dates in a tropi- cal rainforest location. Experimental Agriculture, 21, 31–40. https://doi.org/10.1017/S0014479700012229 FAO, (2016). FAOSTAT. Food and Agricultural Organization of the United Nations, Rome, Italy. http://www.fao.org/fa- ostat/en/#rankings/countries_by_commodity Fayose, C. A. and Fakorede M. A. B. (2014). Growth analysis of maize varieties at the seedling and vegetative develop- mental stages. Proceedings of the 38th Annual Conference of Acta agriculturae Slovenica, 117/2 – 2021 12 C. A. FAYOSE, M. A. B. FAKOREDE the Genetics Society of Nigeria, Benin, Edo State, Nigeria. 19–23rd October. Pp 453-459. Fayose, C. A. and Fakorede M. A. B. (2021). Planting date effects on maize growth and development in a rainforest agroeco- logical location of Southwestern Nigeria. Acta Agricultu- rae Slovenica, 117(1), 101–109. https://doi.org/10.14720/ aas.2021.117.1.1297 Fayose, C. A., Fakorede M. A. B., Badu-Apraku, B. and Menkir, A. (2021). Canopy architecture and grain yield of maize in the rainforest of Southwestern Nigeria. Manuscript submit- ted to Experimental Agriculture. Jong, S. K., Brewbaker, J. L. and Lee, C. H. (1982). Effects of solar radiation on the performance of maize in 41 succes- sive monthly plantings in Hawaii. Crop Science, 22, 13–18. https://doi.org/10.2135/cropsci1982.0011183X0022000100 04x Li, C., Li, Y., Shi, Y., Song, Y., Zhang, D., Buckler, E. S., Zhang, Z., W ang, T . & Li, Y . (2015). Genetic control of the leaf angle and leaf orientation value as revealed by ultra-high density maps in three connected maize populations. PLOS One Journal, 10(3), e0121624. https://doi.org/10.1371/journal. pone.0121624 Mohammadi, S.A., Prasanna, B.M., and Singh, N.N. (2003). Sequential path model for determining interrelationship among grain yield and related characters in maize. Crop Science, 42, 1690-1697. https://doi.org/10.2135/crop- sci2003.1690 Oke, O. F. (2016). Effects of agro-climatic variables on yield of Zea mays L. in a humid tropical rainforest agroecosystem. Journal of Environment and Earth Science, 6(1), 22-30. SAS Institute Inc. (2000). “Base SAS ® 9.0 Procedures Guide.” Cary, NC: SAS Institute Inc. Song, L., Jin, J. & He, J. (2019). Effect of severe water stress on maize growth process in the field. Sustainability, 11, 1-18 https://doi.org/10.3390/su11010001 SPSS Inc. (2007). SPSS Base 17.0 for windows user’s guide. SPSS Inc., Chicago, USA. Talabi, A. O., Badu-Apraku, B. and Fakorede, M. A. B. (2017). Genetic variances and relationship among traits of an early maturing maize population under drought-stress and low nitrogen environments. Crop Science, 57, 681-692. https:// doi.org/10.2135/cropsci2016.03.0177 Tokoro, T. Ishikawa, T., Shirai, S. and Nakamura, T. (2016). Thermal conductivity of sandy soil with electrical charac- teristics. Soils and Foundation, 56(5), 927–936. https://doi. org/10.1016/j.sandf.2016.08.016 Tunde, A. M., Usman, B. A. and Olawepo, V . O. (2011). Effects of climatic variables on crop production in Patigi L. G. A., Kwara State, Nigeria. Journal of Geography and Regional Planning, 4(14), 695-700.