Influence of precipitation and temperature on maize production in the Czech Republic from 2002 to 2019

Maize is one of the important food crops in the Czech Republic, its growth and productivity are influenced by climate change. This study investigated the influence of precipitation under recent climate change on maize yield both for grain and silage in the whole Czech Republic during 2002–2019. Total maize yield and yield rate increased in the Czech Republic from 1961 to 2010, but they became to decrease after 2010. This is in line with the tendency of decreased precipitation and an increase in temperature after 2010, and changes are especially significant during the maize growing period, which indicates the importance of temperature and precipitation. In detail, there is a low to moderate negative correlation (−0.39 to −0.51) between grain maize yield and the average temperature in August for almost all the regions. While there is a low negative correlation between silage maize yield with the average temperature in July and August from some regions. The precipitation in July exhibited moderate to high positive correlation (0.54–0.79) to grain maize yield rate for almost all the regions, and it had low to moderate positive correlation (0.35–0.70) to silage maize yield rate for all the regions. Water deficit exhibited a negative correlation with both maize yield rate and its influence mainly in July for silage but both in July and August for grain. Farmer’s profit from grain maize is influence by yield rate, temperature, precipitation, and water deficit. A positive correlation was found between profit and grain yield rate and precipitation from July and August, while a negative correlation was detected between profit and water deficit and the average temperature in July and August. In conclusion, our results pointed out the factors influencing maize yield rate under changing climate conditions in the Czech Republic, and it warrants further studies on how to maintain maize production in a changing climate.


Material and methods
Study area. The Czech Republic belongs to central Europe, and it lies mostly between latitudes 48° and 51°N and longitudes 12° and 19°E. It has typical European continental influenced temperate climate with warm, dry summer and fairly cold winter. Temperatures vary depending on the elevation with temperatures decrease while precipitation increases at higher altitudes. The highest temperature normally in July, followed by August and June, and most precipitation during the summer period. More detailed information is shown in Fig. 1.
Data collection and analysis. The data of maize yields and cost for maize production (both for grain and silage) in the Czech Republic including 14 regions from 2002 to 2019 were obtained from the Institute of Agricultural Economics and Information (IAEI, http:// www. iaei. cz). The temperatures and rainfall data from 1961 to 2019 come from the Czech Hydrometeorological Institute (CHMI, https:// www. chmi. cz/ histo ricka-data/ pocasi/ zakla dni-infor mace?l= en). The climate maps for temperature and precipitation are obtained from the CHMI as well. The grain maize yield from 1920-2019 was collected from the Czech Statistical Office (https:// www. czso. cz).
Water deficit calculation. Water deficit is a very important parameter for crop production. We made the water deficit calculation for maize (both for grain and silage, 2002-2019) based on the Czech technical norm (ČSN 750,434, http:// www. techn icke-normy-csn. cz/ 750434-csn-75-0434_4_ 15406. html), which uses standardized temperatures (ST) according to the long term averages. Standardized temperatures are devoted to optimal rainfalls (OR) which are stated for the vegetation period (April-October) and they represent the sum of monthly rainfalls that ensure maximal yield for a given crop. In this study, we calculated the water deficit for grain from May to September, and silage from May to August. The obtained results can be negative values when there was a recorded water surplus, or positive if there was a lack of rainfall. The water deficit (DW) for maize is calculated according to the following equations. where WB = water balance, r = observed rainfalls for given period, AOR = Adjusted optimal rainfalls.
where AOR = Adjusted optimal rainfalls, aor = adjustment for optimal rainfalls, OR = optimal rainfalls where aor = adjustment for optimal rainfalls, td = temperature difference between optimal and observed values.
where td = temperature difference between optimal and observed values, ROT = Rounded observed temperatures (to the whole number), TS = Temperature standard (long term average). Pearson correlation was carried out to explain the effects of temperature, precipitation, and water deficit on maize production and profit. The profit (CZK/ha) was calculated by the difference between total income (yield rate (t/ha) X maize price (CZK/t)) and cost (CZK/t), without taking subsidy into account. The correlation coefficient r absolute value between 0.7 and 0.9 is considered as high correlation, 0.5-0.7 as moderate correlation, 0.3-0.5 as low correlation, and 0-0.3 as negligible correlation. All the statistical analyses and plots were realized by Origin 2019b. The significant difference was set as p < 0.05.

Results and discussion
Maize yield trends. In the Czech Republic, both grain maize yield and yield rate slightly increased from 1920 to 1960, and a faster increase since the 1960s, and a dramatic increase in the late 1990s (Fig. 2). The mean grain maize yield rate is 2.9 t/ha and 7.73 t/ha in the 1960s (1961)(1962)(1963)(1964)(1965)(1966)(1967)(1968)(1969)(1970) and 2010s (2001-2010), respectively. This means an average annual increase rate of 96.65 kg/ha over 50 years , which is comparable with the situation in the USA (100 kg/ha over the last 60 years of the last century from 1 to 7 t/ha). This is mostly driven by the improvement of cultivation technology and management levels 17 . However, both grain maize yield and yield rate exhibited a sharp decrease since 2010. The grain maize yield rate slightly decreased from 7.73 t/ ha (2001-2010) to 7.67 t/ha (2011-2019). This decrease can be significant when considering the improvement of technological and management on maize production, and it will have a significant influence on food self-  www.nature.com/scientificreports/ sufficiency. For instance, even a 5% yield decline of silage maize would lead to its shortage of animal feeding in the Czech Republic 18 . Therefore, a more detailed analysis of the change was evaluated for each region during 2002-2019. Since the 1990s, the interest in grows silage maize increased in the Czech Republic. Thus, we include both grain and silage maize for detailed changes from 2002 to 2019. The South Moravian region is the most productive region for grain maize which accounts for 42% (average production of 38,525 t) of the whole Czech Republic, and Central Bohemia region is the second with about 14% average production of 12,857 t) of the whole country. The top three regions for silage maize production are Vysocina, South Boheimia, and Central Boheimia with an average yield of 33,371 t (15.8%), 31044t (14.7%), and 30652t (14.5%), respectively. The change of yield rate is significant on silage maize in which all the regions achieved significantly (all p < 0.05) higher yield rate in 2011-2019 than in 2002-2010 (Table 1). On the contrary, there is an insignificant (all p > 0.05) difference in grain maize yield rate between 2011-2019 and 2002-2010 (Table 1).  However, there is a sharp decrease in the 2010s to 634 mm, which is 88 mm less than the previous decade 2000s (Fig. 3). What's more, the average precipitation during the maize growing season (May to August) was less in the 2010s than in the 2000s (Fig. 4). This phenomenon was more obvious in the Central Bohemia and South Moravian regions (Fig. 4). When we compare the annual precipitation of each year in the 2010s with   The average decade temperature in the Czech Republic increased over time, from 7.1 °C in the 1960s to 8.9 °C in the 2010s (Fig. 3), which is an increased rate of 0.036 °C per year. However, there is a sharp increase of 0.7 °C in the 2010s compared to 2000s. This increase is significate (1.9 times) compare to the normal increase of 0.036 °C per year. In the maize growing season (May to August), a higher mean temperature was found (June to August) in the 2010s than in the 2000s, and a lower mean temperature was found in May in the 2010s than in the 2000s (Fig. 4). In general, the annual temperature in the 2010s is significantly higher than in the previous three decades, (Fig. 6)   www.nature.com/scientificreports/ Effects of climate factors on maize production. Temperature is an important factor influencing the maize yield. This influence is more visible for grain maize than for silage maize (Tables 2 and 3). There is a low negative correlation (−0.34 to −0.5) between grain maize yield and the average temperature in July for 9 regions out of 14 (Table 2), and there is a low to moderate negative correlation (−0.39 to −0.51) between grain maize yield and the average temperature in August for almost all the regions (13 out 14). It's also observed that the silage maize yield has a low negative correlation with the average temperature in July and August (6 regions for both months). In general, climate change is affected by increasing concentration of greenhouse gas, for example, CO 2 . The increase of CO 2 has direct and indirect effects on maize production. The direct effect also called the CO 2 -fertilisation effect 19 , is because the increased CO 2 may enhance the photosynthesis and water use efficiency, thus increase the maize growth about 14% with doubled ambient CO 2 19,20 . The indirect effect also called the weather effect is through solar radiation, precipitation, and temperature to influence maize yield. For instance, the maize yields normally decrease with increasing temperature because of the shorter phenological phases 21 .
Precipitation is another important factor influencing the maize yield. Both grain and silage maize yield rate showed a positive correlation with precipitation in July (Tables 2 and 3), and the correlation is stronger for grain than for silage. The grain maize yield rate had a moderate to high positive correlation (0.54-0.79) with precipitation in July for almost all the regions (expect Karlovy Vary Region). The silage maize yield rate had a low to    (Tables 2 and 3). Our results are being consistent with the previous results that relatively high correlations between precipitation in July with silage yield, which explained 64% of the observed variability in the average silage yields of maize 8 . In principle, the effect of precipitation may be either positive or negative. Positive correlation may happen if precipitation reduces the existing water stress, and negative correlation may because of the intensified nitrogen leaching by the excessive precipitation 21 . However, the relationships between maize yields and climate change characteristics can be different due to individual sites depending on the present climate conditions. For example, the relationship between precipitation (total in the growing season) and maize yield was stronger in the southeastern than in the northeastern U.S., but the relationship between critical month precipitation and maize yield was stronger in the northeastern than in the southeastern U.S. 22 . Nevertheless, the Czech Republic is supposed to face a decrease in spring and summer precipitation in most of the regions 23 . Water deficit exhibited a negative correlation with both maize yield rate (Tables 2 and 3), and its influence mainly in July for silage but both in July and August for grain. There is a low to high negative correlation (−0.38 to −0.71) between grain maize yield rate and average water deficit in July for almost all regions (expect Karlovy Vary Region), and also a low to high negative correlation (−0.30 to −0.71) between grain maize yield rate and average water deficit in August for 12 regions out of 14. There is a low to moderate negative correlation (−0.37 to −0.69) between silage maize yield rate and average water deficit in July for all the regions. Therefore, the critical month for water deficit of both grain and silage is July. What's more, the relationship between maize yield and precipitation showed that moisture shortage rather than excess determined maize yield in the Eastern United States 22 .
Water deficit is one of the most significant stress factors in crop production globally 24 , and it leads to significant yield reduction. The Czech Republic is not generally considered as a drought-prone region in Europe, but there are more and more droughts were recorded, for example, 2000, 2001, 2003, 2014, 2015, and 2018. It was reported that the highest water deficit was recorded for maize growing areas from the field block scale in the Czech Republic due to low precipitation and high water requirements, and about 39.6% (9745.5 km 2 ) of silage maize growing area was effected 25 . The droughts had a profound effect on national and regional agricultural production in the Czech Republic, with yields being consistently lower than in normal years 26 . There is a statistically significant correlation (p < 0.05) between the sum of Palmer's Z-index for maize, and the yield departure is 48% 27 . Unfortunately, the tendency to more intensive dry episodes may happen more frequently in the Czech Republic because it's driven by temperature increase and precipitation decrease, and it already very prominent since 2010 (Fig. 3).
However, except the effect of climate change on maize production, some other technological boundaries also demonstrate their influence on maize production 28 . Lower estimations of maize productivity and technological dependability of current maize cultivars can identify with production techniques predominately pointed toward expanding yield. It is all around recorded in France and the United Kingdom, where a reformist expansion in maize yields by 0.12 t/ha each year was identified with a decline in maize production during the most recent 50 years. In the Czech Republic, maize yield is mostly influenced by location, nitrogen fertilization and year with 120 kg N/ha being considered a sufficient application 29 . However, yield was also related to the percentage of perennial legumes used, particularly when low rates of nutrients were applied (lower than 100 kg NPK ha −1 of arable land). The influence of varieties was relatively low in comparison to the influences of location, year, nitrogen application, use of growth regulators and fungicides.
Furthermore, some multi-temporal-scale index that quantifies persistent anomalies in the local soil moisture balance may better identify the effects of climate factors on maize production. For example, the combined stress index (CSI) which integrates the standardised precipitation evapotranspiration index (SPEI) and the heat Table 3. Pearson correlation between silage maize yield rate with temperature, precipitation, and water deficit in all regions. www.nature.com/scientificreports/ magnitude day index (HMD) was used to predict maize production 30 . In which, the SPEI is a multi-temporalscale index and it is able to capture the impact of drought on agricultural production 31 . This should be taken into account for future research, for example, combined with water deficit.
Profit and its relationship with temperature, precipitation, and water deficit. Farmer's profit from maize was calculated only for grain because silage was most used by the farmers. The heatmap shows that profit has a moderate to high positive correlation to grain yield rate ( Table 4). The average temperature in July and August have a low to moderate negative influence on profit. Precipitation of the whole growing season (May to August) has a low to moderate positive influence on profit, with the negative effect from May and June but the positive effect from July and August. In general, water deficit hurts the farm's profit, and the water deficit in July and August have a moderate to high negative influence on profit, while the effects from May and June are neglective. Water deficit can be express as drought and it is an issue for all European Union (EU) countries. It was reported that 15% of the EU territory and 17% of the EU population from 2006 to 2010 have been affected with economic losses of over € 100 billion 32 . From the economic aspect, the regional maize prices are highly correlated with the world markets (Fig. 7). There is a high correlation in the long-run (0.8 for Czech to EU prices, and 0.823 for Czech to US CME prices, and 0.883 for EU to USD CME prices). At the beginning of the observed period, the maize prices followed the increasing trend of other commodities which was started by increasing crude oil prices in 2002 33 . Since 2006 maize price started to peak to its historical maximums. The demand was particularly led by the increasing production of bio-ethanol in the United States 34 . The relationship 35 and asymmetric volatility transmission between Table 4. Pearson correlation between profit with temperature, precipitation, and water deficit. www.nature.com/scientificreports/ maize and ethanol prices has been also found as one of the key variables 36 . The decrease in 2009 was common for most of the commodity markets and it is related to the financial crisis in 2008 37 . The maize markets peaked to its all-time highs in 2012 and 2013. The maize prices followed again the other commodities including the crude oil and the decision of Saudi Arabia to remove the extraction limits on crude oil production which brought the overproduction and started the price decrease 38 . There are many factors which affect maize markets. The widely discussed bio-ethanol production and policies supporting the bio-fuel usage can contribute to the higher effectivity in corn production and yield, as there comes more investments in technologies from farmers 39 . From the environmental and climate aspects, agriculture is by a long shot the greatest worldwide user of freshwater resources and subsequently exceptionally helpless against environmental and climate change 40 . Climatic conditions straightforwardly influence global agriculture trade and the water resources expected to keep a steady production in numerous territories of Europe 41 . The pressure forced by environmental and climate change on agriculture and water will upgrade existing provincial incongruities in the Czech Republic and other European countries 42,43 .

Conclusion
Even there are other factors (e.g. fertility level, plant management, and insect, disease, and weed pressures, etc.) that can affect the production of maize, there is no doubt that precipitation is one of the most important factors controlling maize yield both for grain and silage. Furthermore, water deficit with the combination of precipitation and temperature (which is one of the most important parameters for climate change) is even more significant and critical for maize (both for grain and silage) production. Meanwhile, water deficit rather than water surplus determined maize yield in the Czech Republic. As a result, farmer's profit from maize was significantly associated with water deficit, precipitation, and temperature, and it's decreasing under the changing climate (warmer and dryer) in the Czech Republic. Therefore, more technology and management approaches (e.g. appropriate tillage, and agricultural water management) are needed to cope with climate change.