Analysis of Relationship between Soybean Relative Maturity Group, Crop Heat Units and ≥10 °C Active Accumulated Temperature

Abstract

Crop heat units (CHU) and ≥10 °C active accumulated temperature (≥10 °C AAT) are important indexes to quantify the effects of temperature on soybean development. The relative maturity group (RMG) is widely used in the classification of different soybean varieties. However, CHU and ≥10 °C AAT (AAT) were applied in Canada and northeastern China, respectively, and the relationships among CHU, AAT and RMG are poorly documented. The objective of this study is to analyze the conversion function among CHU, AAT and RMG based on two datasets. The first dataset was obtained to analyze the relationship between RMG and AAT in 395 varieties in Northeast China. The second dataset was obtained to calculate the relationship between CHU and AAT at 95 weather stations based on 30-year climatic data (1990–2019). The results showed that both relationships were significantly and positively correlated, and the R-square of these relationships were 0.90 and 0.98, respectively. The distribution of CHU or AAT in the Northeast is proposed. These results can be extensively used for predicting the CHU or AAT of soybean cultivars given the known RMG, thus determining the adaptation zone as well as the growth stage of agricultural practices and responses to heat accumulation. The conclusion of the current study is expected to be widely adopted by soybean regionalization and germplasm exchanges throughout the world.

1. Introduction

Soybean is a warm-season crop, and heat has a great influence on the adaptation, yield, and seed quality of soybean. A temperature increase within a certain range is conducive to soybean yield increase [1]. Northeast China includes Heilongjiang, Jilin, Liaoning, and part of Northeast Inner Mongolia (38°42′~53°55′ N, 115°32′~135°09′ E) [2]. These regions are the most important soybean-producing areas in China, especially Heilongjiang and northeastern Inner Mongolia [3]. In 2019, the soybean planting area of Heilongjiang and Inner Mongolia amounted to about 5.33 million hectares, occupying 57% of the total soybean planting area in China (Ministry of Agriculture and Rural Affairs, Development of Seed Industry Management, 2020). Heilongjiang and northeastern Inner Mongolia have relatively low average temperatures and short frost-free periods compared with other soybean-producing regions, and therefore heat is the limiting factor for soybean adaptation, growth and yield in these regions [4]. Therefore, an index for evaluating the heat condition of the location and the heat requirement of the variety is necessary, and it facilitates the determination of the suitable adaptation area of each cultivar.

AAT is most widely used in the main producing regions in China, including Heilongjiang, northeastern Inner Mongolia and other places, and refers to the sum of daily active temperature (daily average temperature ≥10 °C) in a certain period of time or a certain growing season of crops [1,5,6]. Breeders and farmers used ≥10 °C AAT to describe the heat requirement and determine the adaptation zone of each cultivar in these regions. Spatiotemporal variation of AAT and its effect on crop yield was found in Heilongjiang Province, China; therefore, AAT is one of the criteria for releasing a variety in northeastern China, Heilongjiang province. The indicator provides an important theoretical basis for determining suitable soybean varieties in Northeast China [7,8].

Crop heat units (CHU) are recognized as one of the best indexes to quantify the effects of temperature on crop development [9]. Studies have shown that the average yields of soybeans were highly correlated (R² = 0.74) with the average available CHU, with yields increasing by about 0.0013 t ha⁻¹ CHU⁻¹ for soybeans [10]. CHU is an important index to evaluate the suitability of soybean varieties for production in various regions of Canada [11] and could be important in other regions. Determining the feasible planting area of soybean by CHU mapping is valuable and important in Canada [11]. As well, CHU can also provide information on the growth stage of agricultural practice, e.g., the timing for applying fertilizer and herbicide and the response to heat accumulation [9].

The maturity group (MG) system was used to group soybean varieties according to their photothermal response. Soybean has been classified into 13 MGs according to their maturity, which determines their geographical adaptation to different latitudes [12,13,14]. In the United States, studies have elucidated that the yield varies among different MGs of soybean cultivars in different planting environments [15]. The use of the MG system has been widely used in North America as well as worldwide [16]. The relative MG (RMG) is an gradient within the MG by being exact to a one-digit decimal of the MG value [13]. For the MG later than MG 0.0, the RMG is the MG value exact to a one-digit decimal. According to the regression model constructed between the maturity group and the growth period of the standard cultivar, we can calculate the maturity group of cultivars with its growth period.For the MG prior to MG 0, in order to precisely quantify them, negative-expressing RMG values were used for the rating system. An RMG of −1.1 to −2.0, −0.1 to −1.0 corresponds to an MG 000.9–MG 000.0, MG 00.9–MG 00.1, respectively (Supplementary Table S1). The maturity groups in Heilongjiang, Jilin, Liaoning, and northeastern Inner Mongolia are predominantly MG000-III from north to south [17], and the critical photoperiod is relatively long [18]. Temperature and accumulated heat become the most important factors that determine the adaptation zone of the soybean varieties [19,20].

RMG, AAT, and CHU are three indexes that are very important to soybean production, but they are applied in different regions. AAT and CHU are widely used in northeastern China and Canada, respectively, whereas RMG is used worldwide. Given that the relationship between RMG, AAT, and CHU is unclear, the exchange of soybean germplasm resources between China and other countries is impossible. The aim of this study was to (1) establish a regression function between RMGs and AAT based on 395 soybean varieties from Northeast China; (2) construct a regression function between CHU and AAT based on the 30-year (1990–2019) average climatic data of 95 National Meteorological Science Data Centers in the provinces of Heilongjiang, Jilin, Liaoning, and northeastern Inner Mongolia. The conclusion is expected to be adopted widely in soybean-producing regions throughout the world, and it will assist China and other countries in exchanging soybean germplasm resources.

2. Data and Methods

2.1. Calculation

The average daily CHU was then computed from the daily temperature normals in each of the 95 sites using the following Formulas (1) and (2):

Y_max = 3.33 (T_max − 10.0) − 0.084 (T_max − 10.0)² (if T_max < 10.0, Y_max = 0.0)

(1)

Y_min = 1.8 (T_min − 4.44) (if T_min < 4.44, Y_min = 0.0)

(2)

where Y_max and Y_min are the contributions to CHU from average daily maximum (T_max) and minimum (T_min) air temperatures, respectively.

Then, the average daily CHU = (Y_max + Y_min)/2.0.

CHU value of each year was calculated as the sum of the average daily CHU throughout the year, and the CHU value of each site in Northeast China was set to a 30-year average.

The AAT was calculated from the daily average temperature in each of the 95 sites using the following Formula (3):

$$\mathrm{AAT}={\sum }_{i=1}^n(T_i) T_i\ge B (\mathrm{if} T_i<B, \mathrm{AAT}=0.0)$$

(3)

where n is the number of days for growth, T_i is the average temperature for the ith day, B is the base temperature for soybean growth which is usually 10 °C.

The AAT value of each year was calculated as the sum of the average daily AAT throughout the year, and the AAT value of each site in Northeast China was set to a 30-year average.

2.2. Data Collection

2.2.1. Data 1: Identification of the Association between RMGs and AAT in 395 Soybean Varieties

The RMGs of 395 soybean varieties were identified in our previous study, which included varieties from Heilongjiang, Jilin, Liaoning, and Northeast Inner Mongolia [13,17], and 181, 120, 60, and 34 varieties originated from Heilongjiang, Jilin, Liaoning, and northeastern Inner Mongolia, respectively. The AAT required for 395 soybean varieties during the growth period in the location of the cultivar released were downloaded from the database of the Ministry of Agriculture and Rural Affairs, China (last accessed on 25 November 2021, http://202.127.42.145/bigdataNew/home/index), Soybean Variety Records (1993–2014) and relevant breeding institutes (Supplementary Table S2).

2.2.2. Data 2: Identification of the Association between AAT and CHU Based on 30-Year Average Climatic Data of 95 Weather Stations in Northeast China

The daily climate data (average daily maximum, average and minimum air temperature) for the full period of records (30-year averages) for the period from 1990 to 2019 were obtained from the China National Meteorological Science Data Center for 95 available stations in the provinces of Heilongjiang, Jilin, Liaoning, and northeastern Inner Mongolia (Supplemental Table S3). The distribution map of weather stations was drawn using ArcGIS10.1. Among them, 35, 32, 10, and 18 sites were distributed in Heilongjiang, Jilin, Liaoning, and northeastern Inner Mongolia, respectively. The AAT and CHU of each site were calculated based on the climatic data in each site, and the average values across the 30 years of AAT and CHU were calculated. Given that the length of the growing season will vary with region/maturity group/planting date, the required AAT and CHU of the varieties is generally 85% of the AAT and CHU throughout the local year. Therefore, in this paper, we regarded 85% of the year-round AAT and CHU as AAT and CHU that were effectively utilized in the soybean reproductive period [5]. We considered the 30-year average value for each weather site as the annual ≥10 °C activity accumulated temperature at that site (Figure 1). The longitude, latitude, CHU, and AAT of each site are listed in Supplementary Table S3.

2.3. Data Analysis

Linear regression analysis for RMGs and AAT was computed in 395 soybean varieties using the SPSS 26. To determine the relationships, the RMGs and AAT were considered to be independent and dependent variables to estimate the parameter of the linear model. The linear regression analysis of CHU and AAT was conducted with the same approach.

2.4. Geographical Distribution Mapping

The geographical distribution maps of the annual AAT and CHU in Northeast China were constructed with ArcGIS 10.1 using ordinary kriging interpolation [21]. To process the map in the study, we subjected the AAT and CHU at 95 sites, as well as geographical factors (longitude and latitude), to kriging interpolation. The original map base was downloaded from the Ministry of Natural Resources (last accessed on 12 October 2021, http://bzdt.ch.mnr.gov.cn), which is free for public use, and the approval number is GS(2019)1822.

3. Results

3.1. Relationship between RMGs and AAT in the Growth Period of Northern Spring Soybean

The linear regression model was estimated with AAT and RMGs. The material in the current study is 395 cultivars for different RMGs, which include 8, 33, 86, 114, 97, 54, and 3 in MG 000, MG00, MG0, MG I, MG II, MG III, and MG IV, respectively. The models reached an extremely significant level (p < 0.01) with a linear regression line fitted to the data of an R² of 0.8988. The required AAT showed an increasing trend from early to late maturity groups (Figure 2). This result suggests that for each increase of 1 MG, AAT could potentially increase by 254.2 °C. The AAT for MG 000-00, MG 0, MG I, MG II, and MG III are listed in

Table 1. ≥10 °C active accumulated temperature and crop heat units in different maturity groups.

Maturity Groups	MG 000–00	MG 0	MG I	MG II	MG III
≥10 °C active accumulated temperature (°C)	1597–2080	2081–2335	2336–2589	2590–2843	2844–3097
Crop heat units	2158–2740	2741–3047	3048–3353	3354–3659	3660–3965

. Based on the above relationship, we can predict the AAT with RMG and further determine the adaptation zone.

3.2. Relationship between Annual CHU and AAT of Soybean Growth Period in Main Soybean Producing Areas of Northeastern China

The annual AAT ranged from about 1646 °C to 3736 °C, and that for CHU ranged from 2293 to 4763 (Supplementary Table S3). Negative relationships between CHU and latitude, between AAT and latitude, were found, and R² were 0.62 and 0.53, respectively. This indicates that both CHU and AAT increase with the decrease in the latitude of the location.

The relationship between the annual CHU and AAT of 95 sites was estimated through the linear regression model, and the value of the fitting degree (R²) was 0.98 with an extremely significant level (p < 0.01). The significant and positive relationship between annual CHU and AAT indicated that every 1 unit of increase in CHU would cause 0.83 units of increase in AAT (Figure 3).

Based on the data of 95 sites in Northeast China, we predicted the annual AAT and CHU of unknown places using the kriging function. The predicted CHU and AAT of unknown places are listed in Supplementary Table S4. We mapped their distributions in Northeast China using ArcGIS 10.1 (Figure 4 and Figure 5). Furthermore, the decrease in AAT and CHU was observed with the increase in latitude in northeastern China, suggesting that the higher latitude region has lower available heat; cultivars with lower heat requirements should be utilized within these regions. It provides an important reference for the selection of soybean varieties in these regions.

According to the results of the conversion function of AAT to CHU in our study, we can exchange soybean germplasm resources worldwide. For example, the CHU is approximately 3350 in Brockville, Ontario, Canada [22], which corresponds to Beian, Heilongjiang, China. Considering this from a temperature perspective, Kenjiandou 27 (MG 0) in Northeast China fits Brockville, Ontario, Canada. Foreign varieties can also be introduced into Northeast China based on the above results.

4. Discussion

Soybean is native to our country and forms different types of soybean varieties due to different climatic conditions and cultivation systems around the country [23]. Indicators are required to reflect the adaptive characteristics of different soybean varieties that would meet the thermal condition of different planting areas. The geographic distribution of CHU for the warm season crops such as corn, soybean, and tomato was made based on the day-by-day accumulations from the earliest planting to a season-ending date between 1961 and 1990 in Ontario [24]. Essex County has the highest CHU, as it is the farthest south. In the current study, we also found a gradual linear increase in annual AAT and CHU from north to south in Northeast China, which verified the negative relationship between CHU and latitude as well as AAT and latitude. The CHU and AAT are higher in low latitude sites compared with high latitude sites, which is consistent with another previous study in South China [25]. This is attributed to the temperature having a negative relationship with the latitude of the sites. In addition, the elevation and distance to a lake or ocean also have an influence on CHU [24].

According to the distribution of ≥10 °C AAT and MG, we can determine the cultivated region of each variety. The regions with annual ≥10 °C AAT below 2447 °C mainly include the Da Xing An Ling region of Heilongjiang Province and the city of Erguna, Genhe, Oroqen Autonomous Banne, Yakeshi and Hulun Buir in Inner Mongolia, and MG 00 or earlier soybean varieties adapt to this region. The representative varieties of this region are Heihe 35 (MG 000, the required AAT of 1780 °C) and Dongnong 44 (MG 00, the required AAT of 1900 °C). The region with annual AAT 2448–2747 °C is mainly in the central Heilongjiang Province (including Heihe, Yichun, and Hegang northern), Hinggan League north in Inner Mongolia, and Yanji in Jilin. Heihe 43 (MG 0) and Kenjiandou 27 (MG 0) are representative varieties of the region, and the required ≥10 °C AATs are 2150 °C and 2200 °C, respectively. The range of annual ≥ 10 °C AATs was from 2748 °C to 3046 °C in Jiamusi, Shuangyashan, Qitaihe, Mudanjang, Suihua, Harbin eastern, Daqing northeastern, Qiqihar northern, Heilongjiang Province, southeastern Jilin City and eastern Tonghua City, and Kangxian 8 (MG I) and Kenfeng 16 (MG I) are the representative varieties with the required ≥10 °C AATs of 2500 °C and 2450 °C, respectively. The fourth regional AAT ranged from 3047 °C to 3345 °C and mainly included southwest Daqing, Heilongjiang Province, western Harbin, Baicheng, Songyuan, east-central Changchun, western Jilin, western Baishan, Tonghua, eastern Liaoyuan, and northwest Chifeng, Inner Mongolia. Changnong 16 (MG II) of the required AAT of 2600 °C is a representative variety. The annual AAT exceeded 3346 °C in Tongliao and southeastern Chifeng, Inner Mongolia, west of Dandong, Liaoning, and Siping, Jilin Province. Kaiyu 12 and Tiedou 44 adapt to the region and require ≥10 °C AATs of 2900 °C and 3000 °C, respectively. In addition, the distribution of annual AAT and CHU are similar to the above profiles.

The relationship between CHU and MG identified in the current study is consistent with that in Ontario, Canada, for MG 000-0; however, for cultivars of MG I-III, the CHU identified in the current study is higher than that of the corresponding MG in Ontario, Canada (last accessed on 1 October 2021, https://www.gosoy.ca/mat_groups.php). It is because, in northeastern China, the average temperature is relatively low, and the short frost-free period is short; temperature is the limiting factor influencing soybean yield formation. In order to adequately take advantage of active accumulated temperature, soybean is sown early in April when the temperature is low and harvested after frost. Since the sowing time is early, the emergence and growth are slow, which takes a longer time to complete the growth cycle. This is the reason that the results in the current study may have a little deviation from that in Ontario, and therefore accumulated temperature is more widely-used in Heilongjiang and Inner Mongolia (mainly MG 000-0) than in Jilin and Liaoning Province (mainly MG I-III) in China.

This study elucidates the relationship between RMG, annual AAT, and the CHU of the growth period of different cultivars in Northeast China. When we know ≥10 °C AAT is required for the growing period of soybean in Northeast China, we can predict the RMG or CHU of this variety. Similarly, with the CHU or RMG, we can obtain the AAT required by the cultivar through linear function, and further determine the suitable planting area of this cultivar in Northeast China as well as select proper cultivars, which can lower the risk of failure of maturation for a certain region.

The conclusion of this study establishes a bridge between soybeans from Northeast China and other countries in the world. It provides an important theoretical basis for germplasm exchange between China and other countries.