The pampas agroecosystems record deep changes since some decades, resulted in an expansion to more dry areas, mainly due to economic, institutional and climatic conditions, and also to the incorporation of new technologies (Bert et al. 2021). The increase of international prices of raw matters, in special soybean (Zorzoli 2018), had motivated the production of certain crops. In Argentina, Uruguay and the southeastern of Brazil, had been recorded the highest climate changes worldwide in the last decades of 20th century. Especially, in the pampas area increased the spring- summer rainfalls, the minimum temperatures increased and the maximum decreased (Magrín et al. 2005 and Camillioni and Barros 2020).
The Azul municipality is located in Buenos Aires province, Argentina, to 300 km of the southwest of Buenos Aires city. Approximately, 50% of this municipality, center and north area has livestock talent, while the south area has better conditions for agriculture. In agreement with Requesens (2011), livestock had been moving to lands with higher edaphic restrictions and the agriculture has advancing in livestock places, with an important role of soybean among summer crops. Particularly, the corn (Zea mays) is one of the summer crops most important in Azul municipality. It is a crop that could play an important function from the environmental point of view, due to its contribution to atmospheric carbon capture and to the recovery of the soil organic matter, in accordance with its high productivity and stubble volume (Bocchio et al. 2019).
When consider the climate changes in the pampas area during spring and summer (Magrín et al. 2005), in addition of above mentioned, it is of interest to evaluate a summer crop as a corn with respect to the climate effects. This evaluation can provide valuable information that contributes to increase yields and mitigate the consequences of climate variability, when develop adaptation strategies to climatic changes (Zhou et al. 2017, Meira et al. 2019 and Bassu et al. 2021).
Materials and Methods
In order to determine the influence of agroclimatic variables on the increase of corn yield in the last decades in Azul municipality, the agroclimatic requirements of this crop with respect to the agroclimatic variables of Azul municipality were analyzed.
The daily agrometeorological data with their standard deviation were obtained, which include maximum and minimum temperatures, total rainfalls and global radiation of the center area from Azul municipality, recorded by Centro Regional de Agrometeorología de la Facultad de Agronomía (CRAGM) de la Universidad Nacional del Centro de la Provincia de Buenos Aires (UNCPBA). A climatic series of 30 years (1988-2018) was obtained. A total of three decades were considered because with relation to agrometeorological data in climatology, 30 years is the minimum data set that allow defining the climate of a place. From them, the monthly average values were analyzed, from November to March, because they are the most important month in the corn growth. The 2009-2018 decade and the date set of the two previous decades (the last 10 years with the previous 20) were compared, since a decade is a recommended period to study the climate variability in this type of study.
Later, the corn yield was obtained, according to the data of Estimaciones Agrícolas del Sistema Integral de Información Administrativa del Ministerio de Agricultura (MinAgri), named ‘MinAgri yield", for the period 1988-2018.
The MinAgri yield series at first was fitted to different linear and non linear models using the year as independent variable to separate the technology effect. Different authors suggest that when there is a positive tendency in the data (after linear and non linear fit), is associate to technology progress (new hybrids, improvements in fertilization levels, sowing systems, pests and diseases control, among others). In this way the relation with agroclimatic variables were analyzed. In addition, as do not superseded the non linear models of the linear model performance, it was chose this fit through the program Curve Expert v. 1.40 (Hyams 2009) to obtain the term used in equation 1, proved in Miranda del Fresno et al. (2017), which allow the series correction. "The corrected MinAgri" was obtained:
Rci = corrected yield in the yeari
Ri = original yield of the yeari
R(xi) = yield of the estimated year (i) by linear or non linear fit
R(x0) = yield of the initial year (0) estimated by linear or non linear fit
Later, two yield series were simulated with the program Decision Support System for Agrotechnology Transfer (DSSAT) v.4.6, which include 16 crops models of economic importance and allow simulating their growth, when provide true estimates of the crops performance, according to different management strategies and environmental conditions (Hoogemboom et al. 2019).
This simulation was use to eliminate the effects of different technologies, sowing dates and soils, in order to determine the "potential yield"(without nutritional deficiency and irrigation) and the "dry potential yield" (without nutritional deficiency with the water contribution that comes from the rainfalls).
The Weather Man tool from the DSSAT program was used, adding the Ceres-corn model (Hoogemboom et al. 2019):
a) climate: maximum and minimum temperature, rainfalls and global radiation, with daily data from CRAGM-UNCPBA (1988-2018)
b) soil: chemical and physical properties of each horizon of the profile, with actual profiles (Pazos 2009)
c) management: waste, sowing dates (November 1st), fertilization, irrigation.
d) genetic coefficient of cultivars: species, ecotype and cultivar, obtained from the calibration performed by Confalone et al. (2016) for Azul municipality.
Finally, the five months of agrometereological data were averaged and related with the corrected MinAgri yield, estimating the average and the variation coefficient (VC = standard deviation/ average for the last decade and the previous years and the correlation between each climatic variable and the yield. The corrected yield series was contracted with the variability of the agroclimatic elements.
Results and Discussion
The corn is the third crop most important of Azul municipality, after the soybean and wheat, taking into account the sown surface. Between 1988/89 and 2007/08 were sown as average 23 745 ha, and 31 159 ha between 2008/09 and 20017/18 respectively.
In relation to the agroclimatic requirements, as being a C4 species, the corn is demanding regarding temperature. This is the main controller factor of growing and development of this crop, whenever there was an adequate humidity (Fassio et al. 2018). The mean minimum temperature of the soil for germination should be of 8-9 °C, with 23 °C as optimum and maximum 30 °C and sufficient humidity to germination takes place. The different phenological stages have different heat requirements, but generally are accepted that the basic temperature be 8 °C, the optimum from 30 to 34 °C, and the maximum from 40 to 44 °C (Andrade et al. 1996).
In Azul municipality, corn may growth between November and March, when global radiation is of wide availability. From halfway through October to halfway through March, receive 22 MJ/m2d or more, low limit for high cereal productions (Navarro Duymovich et al. 2011). The annual mean temperature is 13.61 °C. The maximum varies from 28 to 29 °C in January, and from 12 to 13 °C in June. The minimum range between 13° and 15 °C in January, and between 0 to 2 °C in June without reach as average less than 9 °C between November and March , which gave acceptable conditions for the growing and development of corn in the municipality.
Regarding rainfalls, from the sowing to maturity, the corn needs of 500 to 800 mm, with daily requirement of 5 mm, that could varies during the phenological stages. The sensitivity considering the lack of water increases from the differentiation of the male flower in the meristematic tissue, and have a maximum in the flowering, which is the critical period, in the maturity stage the required water decrease (Llano and Vargas 2015). For Azul region, the pattern of distribution of rainfalls is more or less constant in the year, with an average of 802 mm annual. Rainfalls are more abundant in summer months. In October - March period exceeded the 400 mm, this can be a limiting element in the crop development in the region.
The corn crop is very sensitive to frosts. Even, the later frosts can caused the plants loses and the senescence of emerged leaves. The earliest, at the end of the cycle, can shorten the grain filling (Martínez Álvarez 2015). In Azul region, the average date of the first frost is on May 30, with a free period of 202 days (Navarro Duymovich et al. 2011).It means that, if corn is sowing at the beginning of November, the later frosts dates had been passed and for the earliest the crop had been harvested.
Table 1 shows the monthly average of the four considered agroclimatic variables and their standard deviation for November to March period, among 1988-2008 and 2009-2018 decades.
As shown in the table, global radiation, as the climate main element, increase 4.12 % as average. December and January have high difference: 1.87 and 1.34 MJ/m2d, respectively.
Months | November | December | January | February | March | |
---|---|---|---|---|---|---|
1988-2008 | Maximum temperature, °C | 23.57 | 27.08 | 28.74 | 27.29 | 25.16 |
Standard deviation | 2.14 | 1.60 | 1.37 | 1.43 | 1.33 | |
Minimum temperature, °C | 9.71 | 12.51 | 14.10 | 14.03 | 12.29 | |
Standard deviation | 1.32 | 1.23 | 0.94 | 1.46 | 1.50 | |
Rainfalls , mm | 74.70 | 96.44 | 109.22 | 90.18 | 111.98 | |
Standard deviation | 53.59 | 56.87 | 62.26 | 62.59 | 70.26 | |
Global radiation, MJ/m2.d | 24.23 | 25.59 | 25.82 | 23.41 | 17.91 | |
Standard deviation | 2.14 | 2.31 | 1.56 | 2.36 | 2.08 | |
2009-2018 | Maximum temperature, °C | 24.26 | 28.81 | 29.82 | 27.49 | 24.88 |
Standard deviation | 1.54 | 1.52 | 1.30 | 1.60 | 1.32 | |
Minimum temperature, °C | 10.69 | 13.53 | 14.98 | 14.73 | 11.79 | |
Standard deviation | 1.86 | 0.94 | 0.38 | 1.36 | 1.60 | |
Rainfalls, mm | 72.25 | 39.42 | 72.44 | 80.38 | 72.36 | |
Standard deviation | 45.56 | 49.73 | 68.44 | 61.16 | 37.03 | |
Global radiation , MJ/m2d | 24.89 | 27.46 | 27.16 | 23.38 | 18.89 | |
Standard deviation | 1.27 | 1.78 | 1.77 | 2.32 | 2.64 |
The maximum temperature increase 2.59 % as average. It was positive from November to February with high difference in December (1.73 °C) and negative in March (-0.28 °C). Minimum temperature increase 4.89 %, as same as maxium temperature which was positive from November to February, with higher difference in November and December (1 °C in both months).It was negative in March (-0.50 °C).
Regarding the rainfalls, they decreased 30.19 % as average, with negative differences in all months, but more marked in January (-36.78 mm) and March (-36.62 mm).
Table 2 show the average and variation coefficient of the four yield series.
The outlets of the Ceres- corn potential model and dry potential model allow verifying the yields differences, whose values exceeded those of MinAgri and corrected MinAgri (table2), as being a synthetic series without the technology interferences that are in MinAgri series. The difference between the dry potential yield and the MinAgri series is 6569.30 kg/ha for all the period. It means, in Azul municipality, the true yield is 56.63 % less than the reached limit. In the last decade, due to the technology progress and the decrease of the dry potential yields, the difference was of 33.42 % (2913.50 kg/ha).
Yield, kg/ha | |||||
---|---|---|---|---|---|
Period | MinAgri | Corrected MinAgri | Ceres-Corn model potential | Ceres-Maíz model dry potential | |
Average 1988-2018 | 5032.10 | 3523.67 | 16673.73 | 11601.40 | |
Variation coefficient, % | 27.25 | 34.51 | 10.97 | 37.67 | |
1988/89 to 2007/08 |
Average | 4645.85 | 3657.57 | 16336.15 | 13043.05 |
Variation coefficient , % | 22.98 | 23.47 | 10.93 | 29.19 | |
2008/09 to2017/18 |
Average | 5804.60 | 3255.87 | 1815.15 | 8718.10 |
Variation coefficient, % | 28.08 | 53.98 | 10.46 | 47.43 |
As figure 1 show, the technology progress (MinAgri series) increases the yield 25 % in the last decade. In turn, in the corrected MinAgri series, in the last decade, the yield decrease 11 % and the variability increase from 23.47 % to 53.98 %. When comparing the two series, there was a general difference of 30 %, 43.91 % in the last decade and 21.27 % in the previous two, which showed that the linear fits between the two are expanded when years passes. The original data of the yield series of corn turned off the correction with the fit:
(r2 = 0.80; standard error of the model = 870.5)
The outlet of the Ceres-corn model (table 2 and figure 2) show that the potential yield increase 6.20 % in the last decade. However, under dry potential conditions, decrease 33.16 % and the variability increase. It also increases the differences between both series (49.75 % in the last decade).
According with the results of table 1, rainfalls decrease in the last decade. While, global radiation and temperatures increase, which make possible beneficial conditions for a C4 tropical grass, like corn. There is an increase of potential yield (6.20 %) by the radiation increase, because the amount of fixed grains is positively associated with the incident energy (Otegui and López Pereira 2003).
In accordance with these results, Zhou et al. (2017) showed that the variations of temperature and radiation were the main factors that influence on the grain weight of the corn and on its filled indicators. So, it is concluded that the variation in these climatic variables greatly affected the growth rate, the time of the grain filled, its weight and corn yield. The increase in the yield also depends on the temperature. That is why it is beneficial the temperature increase when there is not water stress, according to the phenological stage (Confalone et al. 2017).
In the last decade, the dry potential cultivation differ in 49.75 % of potential yields, because of rainfalls decrease. While, in the previous decades the water effect only explained 20.16 %. If the data of corrected MinAgri yields and dry potential have wide differences under dry conditions, it was showed in both series decrease of the last decade values. Test show that rainfalls explain the high incidence of this environmental factor on yield, under dry conditions. This is because when the water in not a limitation, have low incidence on the results. (Fassio et al. 2018).
The inclusion of corn in an agronomic project that includes strategies adapted to each environment could has more available water in the water balance of the soil, which allow higher production of grains, biomass and carbon, and favors the systems sustainability (Bocchio et al. 2019).
As table 3 show, the average of MinAgri yield decreased and the variability increased. While, the agroclimatic data refers to energy, global radiation, maximum temperature and minimum temperature, increased in the last decade. The rainfalls, decreased in its cumulative values with respect to 1988-2008 period. Related to variability, there was increase of the maximum temperature in the last decade, and decrease of other variables. In study by Rizzo et al. (2022), it was found that 48 % of the yield gain was associated with a decennial climatic tendency: 39 % with agronomic improvements and only 13 % with improvements in the genetic yield potential. It means, the future production gains will depend more on the yield gains of the improvements agronomic experiences.
Yield, kg/ha | Agroclimatic variables | |||||
---|---|---|---|---|---|---|
Periods | Corrected MinAgri | Maximum temperature, °C | Minimum temperature, C | Rainfalls, m | Global radiation, MJ/m2d | |
1988/89 to 2007/08 |
Average | 3657.57 | 26.38 | 12.53 | 482.52 | 23.39 |
Variation coefficient, % | 23.47 | 2.72 | 6.06 | 33.09 | 5.19 | |
2008/09 to 2017/18 |
Average | 3255.87 | 27.07 | 13.14 | 336.84 | 24.37 |
Variation coefficient, % | 53.98 | 3.06 | 5.49 | 28.57 | 3.16 |
The comparison of the correlations between each climate element (climatic variables) and the yield of MinAgri series show that there are significant (P < 0.001), only for the cumulative rainfall from November to March (Pearson coefficient = 0.249).
The performed analysis between the corn yield and the agroclimatic variables showed the relations between climate and corn yield, whose study allows understanding those relations and take decisions in consequence. In studies where the corn growth under future climatic conditions was formed, the relation with the climatic variables had been proved. In those studies it is concluded that looses only could partially compensate, by changes in the phenology and sowing dates (Bassu et al. 2021). It has been state that to modify the sowing date has effect on yield (Zhou et al. 2017, Fassio et al. 2018 and Zhai et al. 2022).
Conclusions
It is concluded that in Azul municipality the climate tendencies in the last years exerts a negative effect on the corn yield, although technology hide it, since when the effects is get off (corrected MinAgri) the yield decrease.
The climate influence increase in the last years of the series. This can be observing from the rainfalls decrease of the last decade, when the yield values decrease in the corrected series. The fall of rainfalls in the spring - summer months is the most influence element in the yield, which shows the marked difference in the series under dry conditions and water contribution.