TY - JOUR
AU - Tuinstra, Mitchell R
AB - Abstract Plant phenotypes are often descriptive, rather than predictive of crop performance. As a result, extensive testing is required in plant breeding programmes to develop varieties aimed at performance in the target environments. Crop models can improve this testing regime by providing a predictive framework to (i) augment field phenotyping data and derive hard-to-measure phenotypes and (ii) estimate performance across geographical regions using historical weather data. The goal of this study was to parameterize the Agricultural Production Systems sIMulator (APSIM) crop growth models with remote-sensing and ground-reference data to predict variation in phenology and yield-related traits in 18 commercial grain and biomass sorghum hybrids. Genotype parameters for each hybrid were estimated using remote-sensing measurements combined with manual phenotyping in West Lafayette, IN, in 2018. The models were validated in hybrid performance trials in two additional seasons at that site and against yield trials conducted in Bushland, TX, between 2001 and 2018. These trials demonstrated that (i) maximum plant height, final dry biomass and radiation use efficiency (RUE) of photoperiod-sensitive and -insensitive forage sorghum hybrids tended to be higher than observed in grain sorghum, (ii) photoperiod-sensitive sorghum hybrids exhibited greater biomass production in longer growing environments and (iii) the parameterized and validated models perform well in above-ground biomass simulations across years and locations. Crop growth models that integrate remote-sensing data offer an efficient approach to parameterize larger plant breeding populations. 1. INTRODUCTION 1.1 Importance of forage sorghum in rainfed environments Sorghum (Sorghum bicolor) is commercially important in semi-arid environments due to its substantial heat and drought tolerance. Grain sorghum is the fifth most important cereal in global production with over 57 million tonnes of grain produced on 40 million ha in 2017 (FAOSTAT). Sorghum is also an important forage and sugar crop and can be utilized to produce plant-based biofuels including starch from sorghum grain, sugar from sweet-stemmed sorghum and cellulose from plant leaves and stems. In the USA, almost one-third of the sorghum grain crop is processed through grain-based ethanol production systems. Limited quantities of sugar-based and cellulose-based biofuel are produced currently, but these are considered important feedstocks for the future, minimizing direct competition with food production (Tilman et al. 2006; Rubin 2008). Biomass sorghums can reach heights of 4–5 m with biomass yields maximized by high crop growth rates throughout the available growing season (Rocateli et al. 2012). When planted at high density, commercial sorghum hybrids exhibit a diversity of plant and canopy types to quickly reach maximum radiation interception (Rooney et al. 2007; Olson et al. 2012; Gill et al. 2014; Truong et al. 2017). Total leaf number was found to be highly correlated with length of vegetative period. Hence, early maturing sorghum has fewer leaves and lower biomass production (Sieglinger 1936). In contrast, at high latitudes, spring-sown photoperiod-sensitive sorghum hybrids exhibit extended vegetative periods resulting in high biomass yields. Moreover, since sorghum exhibits better drought tolerance during vegetative growth stages, the longer period of vegetative growth results in better drought tolerance or drought avoidance in rainfed environments (Rooney et al. 2007). 1.2 High-throughput phenotyping methods potentially facilitate the measurement of canopy and crop growth through the entire cropping season Marker-assisted selection (MAS), next-generation sequencing (NGS) technologies and data analytics pipelines have contributed to the implementation of genome-wide association studies (GWAS), genomic diversity studies, genetic linkage analyses, molecular marker discovery and genomic selection in large-scale plant breeding programmes (He et al. 2014). Although genomic technologies are developing quickly, understanding the biological determinants of quantitative phenotype variation remains the central challenge of modern genetic analysis. New, high-throughput phenotyping (HTP) technologies are expected to be the next step in developing association mapping, gene discovery and developing predictive genomic selection models in crop improvement (Cobb et al. 2013). High labour costs often constrain crop breeding programmes to single measurements of final yield in diverse testing environments over multiple seasons. This bottleneck in field phenotyping has driven intense interest in applying remote-sensing technologies to field crop monitoring (Furbank and Tester 2011). Remote sensing of crops includes passive and active sensing of plants to acquire and interpret data to extract information about features, objects and classes in the area of interest (Konare et al. 2003). Data are processed through an analysis pipeline to calibrate and convert digital data into interpretable information (Campbell 2006). For example, the dynamics of canopy cover influence the pattern of crop growth rate and eventual yield. Remote-sensing images acquired by unmanned aerial vehicles (UAVs) can be used directly for large-scale estimation of leaf coverage and are key components of high-throughput field phenotyping (Duan et al. 2014, 2017; Gouache et al. 2016; Stanton et al. 2017; Zhang et al. 2017; Masjedi et al. 2018; Ribera et al. 2018). Multiple remote-sensing approaches focused on quantifying variations in canopy cover and its dynamics have been investigated. An image-based workflow to monitor the growth and development of the wheat canopy dynamically using RGB cameras was developed by Duan et al. (2016). Similarly, Guo et al. (2017) evaluated the ground coverage ratio of rice from a large number of RGB images under variable field conditions. Light Detection and Ranging (LiDAR) has also been used to estimate canopy cover and above-ground biomass (Jimenez-Berni et al. 2018). Masjedi et al. (2019) introduced a strategy that incorporates multi-sensor time series data, environmental inputs and use Recurrent Neural Networks to predict sorghum biomass. Blancon et al. (2019) reported a high-throughput, model-assisted method for quantifying green leaf area (GLAI) dynamics in maize using multispectral imagery. Zhou et al. (2019) used an image-segmentation method based on machine learning to extract relatively accurate coverage information from RGB images. All of these approaches utilize remote-sensing technology to collect canopy-based phenotypes. However, interpreting dynamics of change is not easily done in empirical models. 1.3 Crop growth models Dynamic crop growth modelling and simulation have become accepted tools for agricultural research (e.g. WOFOST (Diepen et al. 1989), DSSAT (Jones et al. 2003), APSIM (Keating et al. 2003; Holzworth et al. 2018), CROPSYST (Stöckle et al. 2003), EPIC (Williams et al. 1983, 1989)) (Bouman et al. 1996; Jones et al. 2017). Unlike purely statistical approaches, these models have functions that respond to external drivers and how those responses affect other components in the system (Wallach et al. 2018). Well-developed crop growth models as well as HTP approaches have been developed in recent years (Demarez et al. 2008; Casa et al. 2010; Baret et al. 2018; Blancon et al. 2019; Jiang et al. 2019; Parent et al. 2019); however, strategies that accommodate crop growth models as part of HTP pipelines have not been thoroughly explored. Agricultural Production Systems sIMulator (APSIM) is a biophysical simulation model for cropping systems that was designed to predict the dynamics of crop growth, including biomass and grain yield, in response to climate and management conditions (Keating et al. 2003). Agricultural Production Systems sIMulator incorporates a generic crop model that utilizes a library of routines for simulating crop growth and development processes (Wang et al. 2002) and has been used to investigate diverse questions related to food security, climate change adaptation and mitigation, simulation of gene expression and multi-trial simulation (Holzworth 2014). In the investigation of biomass growth in crops like biomass sorghum, the key physiological processes are phenology, leaf area development and crop growth rate as affected by weather and soil conditions. Simulated phenology in APSIM is based on thermal time elapsed in growth stages. Thermal time is calculated from a piecewise linear function of the mean air temperature, depending on base, optimum and maximal temperatures, which are 11, 30 and 42 °C for sorghum, respectively (Hammer et al. 1993). Panicle initiation (conversion of the meristem from production of vegetative initials to reproductive initials) is triggered at a genotype-specific thermal time, which can be further influenced by a genotype-specific photoperiod response. The accumulated thermal time between emergence and simulated panicle initiation determines the value of the total leaf number when divided by the plastochron (°Cd per leaf), period between the appearance of two successive leaf primordia. Leaves are expanded at a rate determined by the phyllochron (°Cd per leaf), period between the appearance of two successive leaves, and thus the product of total leaf number and phyllochron determines the thermal time to reach flag leaf stage (°Cd) (Hammer et al. 2010). The duration of growth stages such as flag leaf to anthesis, anthesis to start of grain filling and start to end of grain filling are also simulated in the model by accumulation of thermal time to reach genotype-specific target values (Muchow and Carberry 1990; Hammer and Muchow 1994; Ravi Kumar et al. 2009). Canopy development is simulated based on the relationship between total plant leaf area (TPLA) and thermal time. Total plant leaf area accounts for the number of fully expanded leaves, size of each leaf and tiller number (Hammer et al. 1993, 2010). The model provides flexibility to simulate canopy development using other options such as leaf size distribution (Carberry et al. 1993; van Oosterom et al. 2001; Hammer et al. 2010) or the extension rate of each leaf (Hammer et al. 2010; Chenu et al. 2018). In the standard version of APSIM, the above-ground dry biomass accumulation is simulated as the minimum of light-limited or water-limited growth, then biomass is partitioned in different ratios to plant parts depending on the plant developmental stages through founded functions (Hammer et al. 2010). 1.4 Bioenergy sorghum The objectives of this study are to develop a crop model for biomass sorghum that can predict seasonal biomass production of diverse hybrids over multiple seasons at different locations by combining HTP and crop growth models. Canopy cover estimated from RGB images was used to estimate key parameters describing leaf cover dynamics, light interception and radiation use efficiency (RUE). Other canopy properties were derived as outputs of the APSIM model. This method provides a new approach for understanding the adaptation of biomass sorghum and its interaction with the environment to identify trait targets for plant breeding. 2. MATERIALS AND METHODS 2.1 Genotypes and field management A set of 18 sorghum hybrids (S. bicolor) (Table 1) were grown in 2015, 2017 and 2018 at the Agronomy Center for Research and Education (ACRE) of Purdue University in West Lafayette, IN, USA. Daily solar radiation, maximum and minimum temperatures and precipitation were recorded at the experimental site. Field trials were conducted each year using a randomized complete block design with four replicates. The hybrid entries were evaluated in 12-row plots with 76 cm spacing between rows measuring 3.81 m long. Seeds were sown at 30-mm depth on 19 May in 2015, 16 May in 2017 and 8 May in 2018 with emerged densities as shown (Table 1). Weeds and pests were controlled as required and there was negligible pest damage to the photosynthetic leaf surface throughout growth. Table 1. Details of the types and observed data for 18 hybrids in central-west Indiana from May to October. The value in a cell is mean plus minus standard deviation and the results of LSD test. **Significant at the 0.001 probability level. †PH, seeds from Pioneer Hi-Bred; RS, seeds from Richardson Seeds; SP, seeds from Sorghum Partners. Genotype† . Type . 2015 plant density . 2017 plant density . 2018 plant density . 2017 flowering date . 2018 flowering date . 2015 final dry biomass . 2017 final dry biomass . 2018 final dry biomass . 2015 max height . 2017 max height . 2018 max height . . . stand count per m2 . . . DAS . . g m−2 . . . cm . . . PH 849F Forage sorghum 15.5 ± 1.3abcde 15.9 ± 1.3def 18.3 ± 0.8defg 72.8 ± 9.2ef 74.0 ± 4.0ef 1670 ± 194bc 2258 ± 382cde 2040 ± 111abc 244.6 ± 10.4def 242.5 ± 29.7de 273.8 ± 11.1bcd PH 877F Forage sorghum 16.1 ± 1.6abc 19.5 ± 1.7a 19.6 ± 0.5ab 66.3 ± 1.9gh 66.8 ± 1.7gh 1497 ± 72cde 2217 ± 448.7cde 2006 ± 187abcd 265.0 ± 7.9c 224.0 ± 52.8ef 294.7 ± 13.5ab RS 327x36 BMR Forage sorghum 15.6 ± 1.1abcd 18.5 ± 2.0abc 19.1 ± 0.8abcde 91.0 ± NAc 83.8 ± 12.5d 1343 ± 269def 2245 ± 401cde 1645 ± 118defg 255.8 ± 12.7cde 246.2 ± 19.7de 250.0 ± 9.6d RS 341x10 Food grade 15.5 ± 0.8abcde 16.7 ± 1.3cde 17.4 ± 1.1g 68.8 ± 0.5fgh 65.5 ± 1.3gh 973 ± 82g 1465 ± 191h 1138 ± 17h 77.5 ± 1.6j 82.7 ± 1.9i 126.0 ± 1.8h RS 366x58 Food grade 12.9 ± 1.8f 12.7 ± 0.3g 15.9 ± 0.7h 77.3 ± 3.1de 73.8 ± 2.1ef 1139 ± 85fg 1939 ± 160efg 1366 ± 173gh 123.5 ± 8.2i 139.3 ± 7.0h 153.3 ± 9.4g RS 374x66 Forage sorghum 14.2 ± 1.2cdef 13.9 ± 0.6fg 17.5 ± 0.9fg 75.0 ± 4.8def 69.8 ± 2.1efg 1634 ± 143bc 2049 ± 346defg 1988 ± 211abcd 263.1 ± 21.2cd 264.7 ± 8.9cd 263.9 ± 3.0cd RS 392x105 BMR Forage sorghum 16.4 ± 1.0ab 17.8 ± 1.9abcd 19.6 ± 0.7abc 91.0 ± 0.0c 90.0 ± 0.0c 1117 ± 99fg 2318 ± 358bcd 1563 ± 196efg 163.0 ± 5.8h 203.4 ± 31.2fg 190.5 ± 13.5f RS 400x38 BMR Sorghum-sudangrass 16.3 ± 0.9ab 17.7 ± 1.8abcd 18.6 ± 1.0bcdef 74.8 ± 1.9def 73.0 ± 1.4ef 1151 ± 148fg 1962 ± 102defg 1516 ± 55fgh 190.1 ± 11.0g 220.9 ± 7.3ef 216.5 ± 3.6e RS 400x82 BMR Sorghum-sudangrass 13.7 ± 1.3def 14.1 ± 0.8fg 15.0 ± 0.8h 76.0 ± NAdef 104.0 ± 5.7b 1248 ± 305efg 2128 ± 59cdefg 1569 ± 403efg 242.6 ± 6.0def 191.7 ± 27.7g 202.8 ± 6.4ef SP HIKANE II Forage sorghum 15.5 ± 0.6abcde 18.7 ± 1.6abc 20.1 ± 0.7a 74.0 ± 2.6def 70.3 ± 1.3efg 1607 ± 237bc 2230 ± 372cde 2080 ± 203abc 239.4 ± 14.8ef 244.9 ± 6.0de 254.5 ± 30.1d SP NK300 Forage sorghum 16.8 ± 2.1a 19.4 ± 2.0ab 19.4 ± 1.0abcd 79.3 ± 1.0d 75.3 ± 3.3e 1570 ± 232bcd 2140 ± 290cdef 1919 ± 188bcd 180.0 ± 10.9gh 189.4 ± 5.0g 189.6 ± 13.1f SP NK5418 Grain sorghum 15.4 ± 1.0abcde 18.7 ± 1.2abc 19.4 ± 0.5abcd 68.0 ± 1.0fgh 65.3 ± 1.9gh 1069 ± 94g 1754 ± 164gh 1210 ± 170h 66.8 ± 2.8j 73.8 ± 5.2i 116.0 ± 1.8h SP NK8416 Grain sorghum 13.5 ± 1.7ef 15.2 ± 2.2ef 15.4 ± 1.1h 79.3 ± 1.7d 70.3 ± 1.3efg 1183 ± 151fg 1929 ± 222efg 1364 ± 310gh 125.1 ± 9.6i 128.1 ± 5.9h 167.4 ± 8.9g SP Sordan 79 Forage sorghum 14.8 ± 0.7abcdef 18.2 ± 2.3abcd 17.7 ± 0.8fg 71.0 ± 2.3fg 69.3 ± 2.9fg 1795 ± 134ab 1942 ± 111efg 2209 ± 221ab 291.0 ± 17.4b 261.4 ± 15.1cd 307.5 ± 1.1a SP Sordan Headless Forage sorghum photoperiod-sensitive 15.1 ± 1.2abcde 18.0 ± 1.3abcd 18.5 ± 0.4cdefg 138.0 ± 0.0a NA 1525 ± 217cde 3117 ± 98a 1882 ± 218cde 240.9 ± 20.7ef 297.6 ± 7.4b 257.8 ± 13.5d SP SS405 Forage sorghum 14.2 ± 1.3cdef 17.4 ± 0.8abcde 18.1 ± 0.5efg NA 108.0 ± 0.0b 1976 ± 70a 2466 ± 299bc 2288 ± 339a 338.5 ± 28.7a 342.0 ± 19.5a 296.0 ± 13.8ab SP Trudan 8 Forage sorghum 12.9 ± 1.6f 17.1 ± 2.3bcde 15.2 ± 0.3h 63.7 ± 0.6h 62.8 ± 0.5h 1523 ± 319cde 1803 ± 104fgh 1834 ± 300cdef 227.1 ± 11.4f 200.8 ± 18.7fg 287.7 ± 1.7abc SP Trudan Headless Forage sorghum photoperiod-sensitive 14.5 ± 2.8bcdef 16.6 ± 2.8cde 15.2 ± 0.8h 131.5 ± 7.5b 120.0 ± NAa 1628 ± 165bc 2634 ± 191b 1766 ± 66cdef 251.1 ± 11.8cde 277.3 ± 15.8bc 262.3 ± 30.1d Degrees of freedom 53 54 54 43 44 53 53 47 53 52 30 Mean 15.0 17.0 17.78 82.6 75.1 1427 2143 1777 209.8 212.2 224.1 Coefficient of variation 9.6 9.9 4.4 4.5 5.2 13 12 13 6.5 9.0 5.8 P-value from ANOVA 3.3e-03 1.2e-06 <2e-16 <2e-16 <2e-16 1.0e-10 5.4e-09 1.3e-08 <2e-16 <2e-16 <2e-16 Significance ** ** ** ** ** ** ** ** ** ** ** Genotype† . Type . 2015 plant density . 2017 plant density . 2018 plant density . 2017 flowering date . 2018 flowering date . 2015 final dry biomass . 2017 final dry biomass . 2018 final dry biomass . 2015 max height . 2017 max height . 2018 max height . . . stand count per m2 . . . DAS . . g m−2 . . . cm . . . PH 849F Forage sorghum 15.5 ± 1.3abcde 15.9 ± 1.3def 18.3 ± 0.8defg 72.8 ± 9.2ef 74.0 ± 4.0ef 1670 ± 194bc 2258 ± 382cde 2040 ± 111abc 244.6 ± 10.4def 242.5 ± 29.7de 273.8 ± 11.1bcd PH 877F Forage sorghum 16.1 ± 1.6abc 19.5 ± 1.7a 19.6 ± 0.5ab 66.3 ± 1.9gh 66.8 ± 1.7gh 1497 ± 72cde 2217 ± 448.7cde 2006 ± 187abcd 265.0 ± 7.9c 224.0 ± 52.8ef 294.7 ± 13.5ab RS 327x36 BMR Forage sorghum 15.6 ± 1.1abcd 18.5 ± 2.0abc 19.1 ± 0.8abcde 91.0 ± NAc 83.8 ± 12.5d 1343 ± 269def 2245 ± 401cde 1645 ± 118defg 255.8 ± 12.7cde 246.2 ± 19.7de 250.0 ± 9.6d RS 341x10 Food grade 15.5 ± 0.8abcde 16.7 ± 1.3cde 17.4 ± 1.1g 68.8 ± 0.5fgh 65.5 ± 1.3gh 973 ± 82g 1465 ± 191h 1138 ± 17h 77.5 ± 1.6j 82.7 ± 1.9i 126.0 ± 1.8h RS 366x58 Food grade 12.9 ± 1.8f 12.7 ± 0.3g 15.9 ± 0.7h 77.3 ± 3.1de 73.8 ± 2.1ef 1139 ± 85fg 1939 ± 160efg 1366 ± 173gh 123.5 ± 8.2i 139.3 ± 7.0h 153.3 ± 9.4g RS 374x66 Forage sorghum 14.2 ± 1.2cdef 13.9 ± 0.6fg 17.5 ± 0.9fg 75.0 ± 4.8def 69.8 ± 2.1efg 1634 ± 143bc 2049 ± 346defg 1988 ± 211abcd 263.1 ± 21.2cd 264.7 ± 8.9cd 263.9 ± 3.0cd RS 392x105 BMR Forage sorghum 16.4 ± 1.0ab 17.8 ± 1.9abcd 19.6 ± 0.7abc 91.0 ± 0.0c 90.0 ± 0.0c 1117 ± 99fg 2318 ± 358bcd 1563 ± 196efg 163.0 ± 5.8h 203.4 ± 31.2fg 190.5 ± 13.5f RS 400x38 BMR Sorghum-sudangrass 16.3 ± 0.9ab 17.7 ± 1.8abcd 18.6 ± 1.0bcdef 74.8 ± 1.9def 73.0 ± 1.4ef 1151 ± 148fg 1962 ± 102defg 1516 ± 55fgh 190.1 ± 11.0g 220.9 ± 7.3ef 216.5 ± 3.6e RS 400x82 BMR Sorghum-sudangrass 13.7 ± 1.3def 14.1 ± 0.8fg 15.0 ± 0.8h 76.0 ± NAdef 104.0 ± 5.7b 1248 ± 305efg 2128 ± 59cdefg 1569 ± 403efg 242.6 ± 6.0def 191.7 ± 27.7g 202.8 ± 6.4ef SP HIKANE II Forage sorghum 15.5 ± 0.6abcde 18.7 ± 1.6abc 20.1 ± 0.7a 74.0 ± 2.6def 70.3 ± 1.3efg 1607 ± 237bc 2230 ± 372cde 2080 ± 203abc 239.4 ± 14.8ef 244.9 ± 6.0de 254.5 ± 30.1d SP NK300 Forage sorghum 16.8 ± 2.1a 19.4 ± 2.0ab 19.4 ± 1.0abcd 79.3 ± 1.0d 75.3 ± 3.3e 1570 ± 232bcd 2140 ± 290cdef 1919 ± 188bcd 180.0 ± 10.9gh 189.4 ± 5.0g 189.6 ± 13.1f SP NK5418 Grain sorghum 15.4 ± 1.0abcde 18.7 ± 1.2abc 19.4 ± 0.5abcd 68.0 ± 1.0fgh 65.3 ± 1.9gh 1069 ± 94g 1754 ± 164gh 1210 ± 170h 66.8 ± 2.8j 73.8 ± 5.2i 116.0 ± 1.8h SP NK8416 Grain sorghum 13.5 ± 1.7ef 15.2 ± 2.2ef 15.4 ± 1.1h 79.3 ± 1.7d 70.3 ± 1.3efg 1183 ± 151fg 1929 ± 222efg 1364 ± 310gh 125.1 ± 9.6i 128.1 ± 5.9h 167.4 ± 8.9g SP Sordan 79 Forage sorghum 14.8 ± 0.7abcdef 18.2 ± 2.3abcd 17.7 ± 0.8fg 71.0 ± 2.3fg 69.3 ± 2.9fg 1795 ± 134ab 1942 ± 111efg 2209 ± 221ab 291.0 ± 17.4b 261.4 ± 15.1cd 307.5 ± 1.1a SP Sordan Headless Forage sorghum photoperiod-sensitive 15.1 ± 1.2abcde 18.0 ± 1.3abcd 18.5 ± 0.4cdefg 138.0 ± 0.0a NA 1525 ± 217cde 3117 ± 98a 1882 ± 218cde 240.9 ± 20.7ef 297.6 ± 7.4b 257.8 ± 13.5d SP SS405 Forage sorghum 14.2 ± 1.3cdef 17.4 ± 0.8abcde 18.1 ± 0.5efg NA 108.0 ± 0.0b 1976 ± 70a 2466 ± 299bc 2288 ± 339a 338.5 ± 28.7a 342.0 ± 19.5a 296.0 ± 13.8ab SP Trudan 8 Forage sorghum 12.9 ± 1.6f 17.1 ± 2.3bcde 15.2 ± 0.3h 63.7 ± 0.6h 62.8 ± 0.5h 1523 ± 319cde 1803 ± 104fgh 1834 ± 300cdef 227.1 ± 11.4f 200.8 ± 18.7fg 287.7 ± 1.7abc SP Trudan Headless Forage sorghum photoperiod-sensitive 14.5 ± 2.8bcdef 16.6 ± 2.8cde 15.2 ± 0.8h 131.5 ± 7.5b 120.0 ± NAa 1628 ± 165bc 2634 ± 191b 1766 ± 66cdef 251.1 ± 11.8cde 277.3 ± 15.8bc 262.3 ± 30.1d Degrees of freedom 53 54 54 43 44 53 53 47 53 52 30 Mean 15.0 17.0 17.78 82.6 75.1 1427 2143 1777 209.8 212.2 224.1 Coefficient of variation 9.6 9.9 4.4 4.5 5.2 13 12 13 6.5 9.0 5.8 P-value from ANOVA 3.3e-03 1.2e-06 <2e-16 <2e-16 <2e-16 1.0e-10 5.4e-09 1.3e-08 <2e-16 <2e-16 <2e-16 Significance ** ** ** ** ** ** ** ** ** ** ** Open in new tab Table 1. Details of the types and observed data for 18 hybrids in central-west Indiana from May to October. The value in a cell is mean plus minus standard deviation and the results of LSD test. **Significant at the 0.001 probability level. †PH, seeds from Pioneer Hi-Bred; RS, seeds from Richardson Seeds; SP, seeds from Sorghum Partners. Genotype† . Type . 2015 plant density . 2017 plant density . 2018 plant density . 2017 flowering date . 2018 flowering date . 2015 final dry biomass . 2017 final dry biomass . 2018 final dry biomass . 2015 max height . 2017 max height . 2018 max height . . . stand count per m2 . . . DAS . . g m−2 . . . cm . . . PH 849F Forage sorghum 15.5 ± 1.3abcde 15.9 ± 1.3def 18.3 ± 0.8defg 72.8 ± 9.2ef 74.0 ± 4.0ef 1670 ± 194bc 2258 ± 382cde 2040 ± 111abc 244.6 ± 10.4def 242.5 ± 29.7de 273.8 ± 11.1bcd PH 877F Forage sorghum 16.1 ± 1.6abc 19.5 ± 1.7a 19.6 ± 0.5ab 66.3 ± 1.9gh 66.8 ± 1.7gh 1497 ± 72cde 2217 ± 448.7cde 2006 ± 187abcd 265.0 ± 7.9c 224.0 ± 52.8ef 294.7 ± 13.5ab RS 327x36 BMR Forage sorghum 15.6 ± 1.1abcd 18.5 ± 2.0abc 19.1 ± 0.8abcde 91.0 ± NAc 83.8 ± 12.5d 1343 ± 269def 2245 ± 401cde 1645 ± 118defg 255.8 ± 12.7cde 246.2 ± 19.7de 250.0 ± 9.6d RS 341x10 Food grade 15.5 ± 0.8abcde 16.7 ± 1.3cde 17.4 ± 1.1g 68.8 ± 0.5fgh 65.5 ± 1.3gh 973 ± 82g 1465 ± 191h 1138 ± 17h 77.5 ± 1.6j 82.7 ± 1.9i 126.0 ± 1.8h RS 366x58 Food grade 12.9 ± 1.8f 12.7 ± 0.3g 15.9 ± 0.7h 77.3 ± 3.1de 73.8 ± 2.1ef 1139 ± 85fg 1939 ± 160efg 1366 ± 173gh 123.5 ± 8.2i 139.3 ± 7.0h 153.3 ± 9.4g RS 374x66 Forage sorghum 14.2 ± 1.2cdef 13.9 ± 0.6fg 17.5 ± 0.9fg 75.0 ± 4.8def 69.8 ± 2.1efg 1634 ± 143bc 2049 ± 346defg 1988 ± 211abcd 263.1 ± 21.2cd 264.7 ± 8.9cd 263.9 ± 3.0cd RS 392x105 BMR Forage sorghum 16.4 ± 1.0ab 17.8 ± 1.9abcd 19.6 ± 0.7abc 91.0 ± 0.0c 90.0 ± 0.0c 1117 ± 99fg 2318 ± 358bcd 1563 ± 196efg 163.0 ± 5.8h 203.4 ± 31.2fg 190.5 ± 13.5f RS 400x38 BMR Sorghum-sudangrass 16.3 ± 0.9ab 17.7 ± 1.8abcd 18.6 ± 1.0bcdef 74.8 ± 1.9def 73.0 ± 1.4ef 1151 ± 148fg 1962 ± 102defg 1516 ± 55fgh 190.1 ± 11.0g 220.9 ± 7.3ef 216.5 ± 3.6e RS 400x82 BMR Sorghum-sudangrass 13.7 ± 1.3def 14.1 ± 0.8fg 15.0 ± 0.8h 76.0 ± NAdef 104.0 ± 5.7b 1248 ± 305efg 2128 ± 59cdefg 1569 ± 403efg 242.6 ± 6.0def 191.7 ± 27.7g 202.8 ± 6.4ef SP HIKANE II Forage sorghum 15.5 ± 0.6abcde 18.7 ± 1.6abc 20.1 ± 0.7a 74.0 ± 2.6def 70.3 ± 1.3efg 1607 ± 237bc 2230 ± 372cde 2080 ± 203abc 239.4 ± 14.8ef 244.9 ± 6.0de 254.5 ± 30.1d SP NK300 Forage sorghum 16.8 ± 2.1a 19.4 ± 2.0ab 19.4 ± 1.0abcd 79.3 ± 1.0d 75.3 ± 3.3e 1570 ± 232bcd 2140 ± 290cdef 1919 ± 188bcd 180.0 ± 10.9gh 189.4 ± 5.0g 189.6 ± 13.1f SP NK5418 Grain sorghum 15.4 ± 1.0abcde 18.7 ± 1.2abc 19.4 ± 0.5abcd 68.0 ± 1.0fgh 65.3 ± 1.9gh 1069 ± 94g 1754 ± 164gh 1210 ± 170h 66.8 ± 2.8j 73.8 ± 5.2i 116.0 ± 1.8h SP NK8416 Grain sorghum 13.5 ± 1.7ef 15.2 ± 2.2ef 15.4 ± 1.1h 79.3 ± 1.7d 70.3 ± 1.3efg 1183 ± 151fg 1929 ± 222efg 1364 ± 310gh 125.1 ± 9.6i 128.1 ± 5.9h 167.4 ± 8.9g SP Sordan 79 Forage sorghum 14.8 ± 0.7abcdef 18.2 ± 2.3abcd 17.7 ± 0.8fg 71.0 ± 2.3fg 69.3 ± 2.9fg 1795 ± 134ab 1942 ± 111efg 2209 ± 221ab 291.0 ± 17.4b 261.4 ± 15.1cd 307.5 ± 1.1a SP Sordan Headless Forage sorghum photoperiod-sensitive 15.1 ± 1.2abcde 18.0 ± 1.3abcd 18.5 ± 0.4cdefg 138.0 ± 0.0a NA 1525 ± 217cde 3117 ± 98a 1882 ± 218cde 240.9 ± 20.7ef 297.6 ± 7.4b 257.8 ± 13.5d SP SS405 Forage sorghum 14.2 ± 1.3cdef 17.4 ± 0.8abcde 18.1 ± 0.5efg NA 108.0 ± 0.0b 1976 ± 70a 2466 ± 299bc 2288 ± 339a 338.5 ± 28.7a 342.0 ± 19.5a 296.0 ± 13.8ab SP Trudan 8 Forage sorghum 12.9 ± 1.6f 17.1 ± 2.3bcde 15.2 ± 0.3h 63.7 ± 0.6h 62.8 ± 0.5h 1523 ± 319cde 1803 ± 104fgh 1834 ± 300cdef 227.1 ± 11.4f 200.8 ± 18.7fg 287.7 ± 1.7abc SP Trudan Headless Forage sorghum photoperiod-sensitive 14.5 ± 2.8bcdef 16.6 ± 2.8cde 15.2 ± 0.8h 131.5 ± 7.5b 120.0 ± NAa 1628 ± 165bc 2634 ± 191b 1766 ± 66cdef 251.1 ± 11.8cde 277.3 ± 15.8bc 262.3 ± 30.1d Degrees of freedom 53 54 54 43 44 53 53 47 53 52 30 Mean 15.0 17.0 17.78 82.6 75.1 1427 2143 1777 209.8 212.2 224.1 Coefficient of variation 9.6 9.9 4.4 4.5 5.2 13 12 13 6.5 9.0 5.8 P-value from ANOVA 3.3e-03 1.2e-06 <2e-16 <2e-16 <2e-16 1.0e-10 5.4e-09 1.3e-08 <2e-16 <2e-16 <2e-16 Significance ** ** ** ** ** ** ** ** ** ** ** Genotype† . Type . 2015 plant density . 2017 plant density . 2018 plant density . 2017 flowering date . 2018 flowering date . 2015 final dry biomass . 2017 final dry biomass . 2018 final dry biomass . 2015 max height . 2017 max height . 2018 max height . . . stand count per m2 . . . DAS . . g m−2 . . . cm . . . PH 849F Forage sorghum 15.5 ± 1.3abcde 15.9 ± 1.3def 18.3 ± 0.8defg 72.8 ± 9.2ef 74.0 ± 4.0ef 1670 ± 194bc 2258 ± 382cde 2040 ± 111abc 244.6 ± 10.4def 242.5 ± 29.7de 273.8 ± 11.1bcd PH 877F Forage sorghum 16.1 ± 1.6abc 19.5 ± 1.7a 19.6 ± 0.5ab 66.3 ± 1.9gh 66.8 ± 1.7gh 1497 ± 72cde 2217 ± 448.7cde 2006 ± 187abcd 265.0 ± 7.9c 224.0 ± 52.8ef 294.7 ± 13.5ab RS 327x36 BMR Forage sorghum 15.6 ± 1.1abcd 18.5 ± 2.0abc 19.1 ± 0.8abcde 91.0 ± NAc 83.8 ± 12.5d 1343 ± 269def 2245 ± 401cde 1645 ± 118defg 255.8 ± 12.7cde 246.2 ± 19.7de 250.0 ± 9.6d RS 341x10 Food grade 15.5 ± 0.8abcde 16.7 ± 1.3cde 17.4 ± 1.1g 68.8 ± 0.5fgh 65.5 ± 1.3gh 973 ± 82g 1465 ± 191h 1138 ± 17h 77.5 ± 1.6j 82.7 ± 1.9i 126.0 ± 1.8h RS 366x58 Food grade 12.9 ± 1.8f 12.7 ± 0.3g 15.9 ± 0.7h 77.3 ± 3.1de 73.8 ± 2.1ef 1139 ± 85fg 1939 ± 160efg 1366 ± 173gh 123.5 ± 8.2i 139.3 ± 7.0h 153.3 ± 9.4g RS 374x66 Forage sorghum 14.2 ± 1.2cdef 13.9 ± 0.6fg 17.5 ± 0.9fg 75.0 ± 4.8def 69.8 ± 2.1efg 1634 ± 143bc 2049 ± 346defg 1988 ± 211abcd 263.1 ± 21.2cd 264.7 ± 8.9cd 263.9 ± 3.0cd RS 392x105 BMR Forage sorghum 16.4 ± 1.0ab 17.8 ± 1.9abcd 19.6 ± 0.7abc 91.0 ± 0.0c 90.0 ± 0.0c 1117 ± 99fg 2318 ± 358bcd 1563 ± 196efg 163.0 ± 5.8h 203.4 ± 31.2fg 190.5 ± 13.5f RS 400x38 BMR Sorghum-sudangrass 16.3 ± 0.9ab 17.7 ± 1.8abcd 18.6 ± 1.0bcdef 74.8 ± 1.9def 73.0 ± 1.4ef 1151 ± 148fg 1962 ± 102defg 1516 ± 55fgh 190.1 ± 11.0g 220.9 ± 7.3ef 216.5 ± 3.6e RS 400x82 BMR Sorghum-sudangrass 13.7 ± 1.3def 14.1 ± 0.8fg 15.0 ± 0.8h 76.0 ± NAdef 104.0 ± 5.7b 1248 ± 305efg 2128 ± 59cdefg 1569 ± 403efg 242.6 ± 6.0def 191.7 ± 27.7g 202.8 ± 6.4ef SP HIKANE II Forage sorghum 15.5 ± 0.6abcde 18.7 ± 1.6abc 20.1 ± 0.7a 74.0 ± 2.6def 70.3 ± 1.3efg 1607 ± 237bc 2230 ± 372cde 2080 ± 203abc 239.4 ± 14.8ef 244.9 ± 6.0de 254.5 ± 30.1d SP NK300 Forage sorghum 16.8 ± 2.1a 19.4 ± 2.0ab 19.4 ± 1.0abcd 79.3 ± 1.0d 75.3 ± 3.3e 1570 ± 232bcd 2140 ± 290cdef 1919 ± 188bcd 180.0 ± 10.9gh 189.4 ± 5.0g 189.6 ± 13.1f SP NK5418 Grain sorghum 15.4 ± 1.0abcde 18.7 ± 1.2abc 19.4 ± 0.5abcd 68.0 ± 1.0fgh 65.3 ± 1.9gh 1069 ± 94g 1754 ± 164gh 1210 ± 170h 66.8 ± 2.8j 73.8 ± 5.2i 116.0 ± 1.8h SP NK8416 Grain sorghum 13.5 ± 1.7ef 15.2 ± 2.2ef 15.4 ± 1.1h 79.3 ± 1.7d 70.3 ± 1.3efg 1183 ± 151fg 1929 ± 222efg 1364 ± 310gh 125.1 ± 9.6i 128.1 ± 5.9h 167.4 ± 8.9g SP Sordan 79 Forage sorghum 14.8 ± 0.7abcdef 18.2 ± 2.3abcd 17.7 ± 0.8fg 71.0 ± 2.3fg 69.3 ± 2.9fg 1795 ± 134ab 1942 ± 111efg 2209 ± 221ab 291.0 ± 17.4b 261.4 ± 15.1cd 307.5 ± 1.1a SP Sordan Headless Forage sorghum photoperiod-sensitive 15.1 ± 1.2abcde 18.0 ± 1.3abcd 18.5 ± 0.4cdefg 138.0 ± 0.0a NA 1525 ± 217cde 3117 ± 98a 1882 ± 218cde 240.9 ± 20.7ef 297.6 ± 7.4b 257.8 ± 13.5d SP SS405 Forage sorghum 14.2 ± 1.3cdef 17.4 ± 0.8abcde 18.1 ± 0.5efg NA 108.0 ± 0.0b 1976 ± 70a 2466 ± 299bc 2288 ± 339a 338.5 ± 28.7a 342.0 ± 19.5a 296.0 ± 13.8ab SP Trudan 8 Forage sorghum 12.9 ± 1.6f 17.1 ± 2.3bcde 15.2 ± 0.3h 63.7 ± 0.6h 62.8 ± 0.5h 1523 ± 319cde 1803 ± 104fgh 1834 ± 300cdef 227.1 ± 11.4f 200.8 ± 18.7fg 287.7 ± 1.7abc SP Trudan Headless Forage sorghum photoperiod-sensitive 14.5 ± 2.8bcdef 16.6 ± 2.8cde 15.2 ± 0.8h 131.5 ± 7.5b 120.0 ± NAa 1628 ± 165bc 2634 ± 191b 1766 ± 66cdef 251.1 ± 11.8cde 277.3 ± 15.8bc 262.3 ± 30.1d Degrees of freedom 53 54 54 43 44 53 53 47 53 52 30 Mean 15.0 17.0 17.78 82.6 75.1 1427 2143 1777 209.8 212.2 224.1 Coefficient of variation 9.6 9.9 4.4 4.5 5.2 13 12 13 6.5 9.0 5.8 P-value from ANOVA 3.3e-03 1.2e-06 <2e-16 <2e-16 <2e-16 1.0e-10 5.4e-09 1.3e-08 <2e-16 <2e-16 <2e-16 Significance ** ** ** ** ** ** ** ** ** ** ** Open in new tab 2.2 Ground validation studies Ground-reference data from trials conducted in 2018 were used to parameterize the APSIM model. Plant population density was determined from row 2 and row 3 of each 12-row plot at 31 days after sowing (DAS). Days to flowering were measured as the number of days from sowing to when 50 % of the panicles in the plot were at 50 % anthesis. Plant height was measured after flowering. Destructive harvests at four different stages of development were used to determine biomass yields; leaf, stem, tiller and panicle weights of individual plants; and leaf size distribution. Four plants were harvested manually from plot row 11 on 7 June (31 DAS), two plants from plot rows 8 and 9 on 25 June (49 DAS), two plants from plot rows 5 and 6 on 12 July (66 DAS) and two plants from plot rows 2 and 3 on 9 August (94 DAS). After harvesting, each plant was dissected to determine the weight of the collared leaves, leaves that had not fully emerged, stems, tillers and panicle fractions. Leaves were removed from each plant in order and scanned individually to determine leaf size distribution using a LI-3100C Leaf Area Meter (LI-COR, Lincoln, NE, USA). The final tiller number was estimated from the tiller dry weight and total plant dry weight. Percent moisture of each plant was determined from the combined fresh weights and, later, dry weights of all fractions from each plant. Repeated non-destructive measurements of plant development were also made during the vegetative period including the number of fully expanded leaves (collared leaves) of four tagged plants in rows 2 and 3 of each plot. Collection dates were 7 June (31 DAS), 19 June (43 DAS), 28 June (52 DAS), 5 July (59 DAS), 11 July (65 DAS) and 26 July (80 DAS). The final leaf numbers were the maximum value of leaf collar counts of each plot across dates. Final tiller number per plant was determined on 12 July (66 DAS). The average leaf biomass fraction and specific leaf weight (SLW) were used to compute leaf area and leaf area index (LAI) for each plot and sampling date. Total biomass yields were measured in each plot on 7 June (31 DAS), plot rows 8 and 9 on 25 June (49 DAS), plot rows 5 and 6 on 12 July (66 DAS) and plot rows 2 and 3 on 9 August (94 DAS); one replicate was not harvested at 49 DAS due to inclement weather. On 7 June (31 DAS), a 2-m section of row segment 11 was hand-harvested, weighed and dried to compare fresh weights and dry weights. For the next three harvest dates, the entire 2-row segment of each plot was harvested with a Wintersteiger Cibus 2-row Biomass Harvester (Wintersteiger Inc., Salt Lake City, UT, USA). After harvesting a plot, ~500 g of the shredded plant material from each plot was taken to determine fresh weight, dry weight and moisture content. For the mechanically harvested plots, a 0.614-kg fresh weight correction factor was added back to the biomass estimate of each plot to account for the short stem segments that were left behind after machine harvesting. At the last sampling date (94 DAS), several plots were lodged and could not be harvested. 2.3 Ground validation data from 2015 and 2017 Replicated trials conducted in 2015 and 2017 were used to validate the parameterized APSIM models for each hybrid. Total above-ground biomass was measured by manual sampling and by machine harvesting. Manual sampling was conducted at 65 and 93 DAS in 2015, and 42, 63, 84 DAS in 2017 by harvesting plants from three 1-m sections of row in rows 5–8 of the 12-row plot. Plant count and biomass fresh weight and dry weight were measured for each sample. An individual plant from each sample was dissected to measure leaf, stem, tiller and panicle weights. The leaf sizes were determined using ImageJ, an open source software package developed by NIH for the analysis of scientific images (Schneider et al. 2012). The leaves were laid on a white board in leaf order from top to bottom. RGB images of the leaves were acquired using a Cannon EOS 6D camera with a Canon 35-mm lens under a white light source and ~1.5 m height. Leaves were segmented by thresholding in HSB (Hue, Saturation, Brightness) colour space with four thresholds. Total leaf area per plant and plant stand information were used to calculate the LAI. A Wintersteiger Cibus 2-row Biomass Harvester (Wintersteiger Inc., Salt Lake City, UT, USA) was also used to mechanically harvest plants from plot rows 10 and 11 on 25 August 2015 (99 DAS) and 31 July 2017 (77 DAS) and from plot rows 2 and 3 on 27 September 2017 (135 DAS) as described above. 2.4 Remote-sensing data collection Remote-sensing data were used to measure canopy cover for each plot. RGB images were collected in 2017 and 2018 using a DJI Matrice M600 Pro UAV as a platform, equipped with an APX-15 V2 as the GNSS (Global Navigation Satellite System)/INS (Inertial Navigation System) unit for direct geo-referencing. Images were collected using a Sony Alpha 7R (ILCE-7R) camera with a Sony 35-mm lens at a height of 50 m, resulting in a ground sampling distance of 0.7 cm. Spatial and temporal calibration of the imaging systems in this study were done by methods described in Ravi et al. (2018). The RGB images were collected in 2017 on 6 June (22 DAS), 21 June (37 DAS), 28 June (44 DAS), 5 July (51 DAS), 11 July (57 DAS), 17 July (63 DAS), 25 July (71 DAS), 2 August (79 DAS), 8 August (85 DAS), 16 August (93 DAS) and 30 August (107 DAS). RGB images were taken in 2018 on 16 May (9 DAS), 22 May (15 DAS), 29 May (22 DAS), 4 June (28 DAS), 11 June (35 DAS), 20 June (44 DAS), 27 June (51 DAS), 2 July (56 DAS), 11 July (65 DAS), 18 July (72 DAS), 23 July (77 DAS), 1 August (86 DAS) and 6 August (91 DAS). The RGB images were collected in 2015 using a DJI Phantom 2 platform, and a GoPro Hero3+ camera at a height of 15 m, with ground sampling distance of 0.7 cm. The images were acquired on 15 June (28 DAS), 26 June (39 DAS), 6 July (49 DAS), 15 July (58 DAS) and 25 July (68 DAS). Orthomosaics were obtained using modified Structure from Motion (SfM) strategies introduced in He et al. (2018) with ground control targets, and then used to identify the coordinates of the plots and row segments. While multiple photos may have overlapping plot coverage, the image coordinates for the same row segment vary from photo to photo. Row segments at the image border suffer more lens and perspective distortion than the row segments at the photo centre, which will have a big impact on canopy cover calculation. Therefore, the photo where the plot is closest to the centre of the image was used for canopy cover estimation. Each row segment was defined by a rectangle whose dimensions were 0.76 m × 3.81 m on average, and then 0.4 m was trimmed from each end of the row to minimize effects of the alley between plots. The canopy cover was estimated for rows 2 and 3 as the ratio of vegetative to non-vegetative pixels within the box, using segmentation methods described previously (Ribera et al. 2018) and canopy cover for each plot taken as the average of the two rows. 2.5 Agricultural Production Systems sIMulator Weather data, soil data, field management and sorghum physiological parameters were used to parameterize the APSIM model for West Lafayette. Weather data included daily solar radiation (MJ), maximum and minimum temperatures (°C) and precipitation (mm). Field management parameters included sowing date, sowing depth and plant density (Table 1). The sorghum physiological parameters included observed parameters (final leaf number, final tiller number, maximum leaf area (m2) and maximum leaf multiplier) and derived parameters (extinction coefficient of canopy (k) and RUE (g MJ−1)) determined from the 2018 data set (Table 2, see explanation of computation of k and RUE below). Table 2. 2018 parameters from the pipeline calculating derived parameters based on observed parameters. K for extinction coefficient and RUE for radiation use efficiency. Genotype . K . RUE (g MJ−1) . PH 849F 0.57 1.47 PH 877F 0.98 1.53 RS 327x36 BMR 0.35 1.36 RS 341x10 0.46 1.15 RS 366x58 0.58 1.15 RS 374x66 0.48 1.60 RS 392x05 BMR 0.44 1.29 RS 400x38 BMR 0.46 1.27 RS 400x82 BMR 0.65 1.28 SP HIKANE II 0.54 1.61 SP NK300 0.84 1.40 SP NK5418 0.79 0.97 SP NK8416 0.55 1.10 SP Sordan 79 0.69 1.64 SP Sordan Headless 0.38 1.40 SP SS405 0.35 1.70 SP Trudan 8 1.43 1.44 SP Trudan Headless 0.38 1.40 Genotype . K . RUE (g MJ−1) . PH 849F 0.57 1.47 PH 877F 0.98 1.53 RS 327x36 BMR 0.35 1.36 RS 341x10 0.46 1.15 RS 366x58 0.58 1.15 RS 374x66 0.48 1.60 RS 392x05 BMR 0.44 1.29 RS 400x38 BMR 0.46 1.27 RS 400x82 BMR 0.65 1.28 SP HIKANE II 0.54 1.61 SP NK300 0.84 1.40 SP NK5418 0.79 0.97 SP NK8416 0.55 1.10 SP Sordan 79 0.69 1.64 SP Sordan Headless 0.38 1.40 SP SS405 0.35 1.70 SP Trudan 8 1.43 1.44 SP Trudan Headless 0.38 1.40 Open in new tab Table 2. 2018 parameters from the pipeline calculating derived parameters based on observed parameters. K for extinction coefficient and RUE for radiation use efficiency. Genotype . K . RUE (g MJ−1) . PH 849F 0.57 1.47 PH 877F 0.98 1.53 RS 327x36 BMR 0.35 1.36 RS 341x10 0.46 1.15 RS 366x58 0.58 1.15 RS 374x66 0.48 1.60 RS 392x05 BMR 0.44 1.29 RS 400x38 BMR 0.46 1.27 RS 400x82 BMR 0.65 1.28 SP HIKANE II 0.54 1.61 SP NK300 0.84 1.40 SP NK5418 0.79 0.97 SP NK8416 0.55 1.10 SP Sordan 79 0.69 1.64 SP Sordan Headless 0.38 1.40 SP SS405 0.35 1.70 SP Trudan 8 1.43 1.44 SP Trudan Headless 0.38 1.40 Genotype . K . RUE (g MJ−1) . PH 849F 0.57 1.47 PH 877F 0.98 1.53 RS 327x36 BMR 0.35 1.36 RS 341x10 0.46 1.15 RS 366x58 0.58 1.15 RS 374x66 0.48 1.60 RS 392x05 BMR 0.44 1.29 RS 400x38 BMR 0.46 1.27 RS 400x82 BMR 0.65 1.28 SP HIKANE II 0.54 1.61 SP NK300 0.84 1.40 SP NK5418 0.79 0.97 SP NK8416 0.55 1.10 SP Sordan 79 0.69 1.64 SP Sordan Headless 0.38 1.40 SP SS405 0.35 1.70 SP Trudan 8 1.43 1.44 SP Trudan Headless 0.38 1.40 Open in new tab 2.6 Model calibration An R pipeline for APSIM parameters calculation was developed to process the 2018 data set. The input data included of weather data and sorghum physiological parameters by plot. Weather data were comprised of maximum daily temperature, minimum daily temperature, precipitation and solar radiation. The sorghum physiological parameters for APSIM: observed leaf number, final tiller number, two leaf size distribution parameters, observed canopy cover and observed biomass, were extracted after spatial analysis of the variable values using spline fits (Rodríguez-Álvarez et al. 2018). Two leaf size distribution parameters, maximum leaf area (aMaxI) and maximum leaf multiplier (aX0) were determined for each hybrid. The leaf size functions were computed as follows (Carberry et al. 1993; Chenu et al. 2008): aMax=aMaxI Individual Leaf Size (cm2)=aMax×exp(a×(Leaf number−Largest leaf position)2+b×(Leaf number−Largest leaf position)3)×100; The factor 100 is a percentage of the maximum leaf size; Largest leaf position=aX0×Final leaf number (FLN); FLN is counted along the stem upwards; a=a0−exp⁡(a1∗FLN); b=b0−exp⁡(b1∗FLN); a0=−0.009a1=−0.2b0=0.0006b1=−0.43 Using the leaf size function, the largest leaf area and the position multiplier of this leaf within the whole plant of each hybrid were determined. Leaf appearance rate was calculated using an assistant function created with global optimization through DEoptim from Package ‘RcppDE’ and read in the R pipeline. The leaf appearance rate was determined by plotting number of fully expanded leaves from the weekly measurements plotted against accumulated thermal time. The leaf appearance rate during the early vegetative stage is typically different from the late vegetative stage, so the regression was split into two parts, with the last four leaves set apart. Leaf appearance rates were determined from the estimated slope of a linear regression, leaf appearance rate 1 (early vegetative) and leaf appearance rate 2 (late vegetative). The fraction of incident radiation intercepted (RI) was computed as described previously (Charles-Edwards 1982; Lafarge and Hammer 2002): RI=1−e−k∗LAI RI is a function of the LAI and the canopy extinction coefficient (k), which is related to canopy structure. Each day the value of LAI was computed from a sigmoidal curve as a function of leaf number, leaf appearance rate, final tiller number and leaf size distribution through accumulated thermal time, and observed canopy cover was then used to derive k based on the RI equation. To avoid any effects of senescent leaves, the canopy cover data collected after anthesis were not used for k calculation. The RUE is defined as the quantity of dry biomass produced under non-stressed conditions based on the amount of intercepted radiation (IR). The maximum RUE for each variety was determined using the slope of the estimated linear relationship between above-ground biomass and cumulative IR, which was derived from the calculated k, calculated LAI and daily radiation. 2.7 Model validation The APSIM models were validated using the performance trials conducted in West Lafayette, IN, in 2015 and 2017. Agricultural Production Systems sIMulator models were also validated for nine of the hybrids evaluated in multi-year trials in Bushland, TX as part of the Texas A&M Forage Sorghum Test (https://amarillo.tamu.edu/amarillo-center-programs/agronomy/forage-sorghum/). For each hybrid, there were different sowing and harvesting dates. When plant stand count was not collected, we applied 90 % germination rate to the seeding rate as the assumed plant density (Table 3). Regression was used to compare predicted and observed values and slope and intercept parameters against the 1:1 line (Piñeiro et al. 2008). Table 3. The genotypes and management details in Bushland trials. Genotype . Year . Sowing date . Harvest date . Stand count (plants per m2) . 849F 2017 6/13 10/4 16.7 849F 2016 6/8 9/15 17.8 849F 2014 6/13 9/8 22.2 849F 2011 5/19 9/2 22.2 849F 2010 5/28 9/7 22.2 849F 2009 5/28 9/9 22.2 849F 2008 5/27 9/22 22.2 849F 2007 5/30 9/25 20.0 877F 2006 5/25 10/6 28.5 HIKANE II 2016 6/8 8/27 17.8 HIKANE II 2011 5/19 9/2 22.2 HIKANE II 2009 5/28 9/16 22.2 HIKANE II 2008 5/27 9/17 22.2 HIKANE II 2007 5/30 9/25 20.0 HIKANE II 2006 5/25 9/11 26.1 HIKANE II 2005 5/25 9/8 26.7 HIKANE II 2004 5/24 9/9 26.7 HIKANE II 2003 5/21 9/5 26.7 HIKANE II 2002 5/23 8/28 26.7 NK 300 2016 6/8 9/26 17.8 NK 300 2011 5/19 9/22 22.2 NK 300 2009 5/28 10/14 22.2 NK 300 2006 5/25 9/14 25.7 NK 300 2004 5/24 9/9 26.7 NK 300 2003 5/21 9/22 26.7 NK 300 2002 5/23 9/27 26.7 Sordan 79 2006 5/25 9/14 25.4 Sordan 79 2005 5/25 9/29 26.7 Sordan 79 2004 5/24 10/13 26.7 Sordan Headless 2016 6/8 10/25 17.8 Sordan Headless 2014 6/13 10/6 22.2 Sordan Headless 2008 5/27 10/26 22.2 Sordan Headless 2006 5/25 10/6 22.5 Sordan Headless 2005 5/25 9/29 26.7 Sordan Headless 2004 5/24 10/13 26.7 Sordan Headless 2003 5/21 10/15 26.7 Sordan Headless 2002 5/23 10/11 26.7 SS405 2017 6/13 10/26 16.7 SS405 2016 6/8 10/15 17.8 SS405 2014 6/13 9/17 22.2 SS405 2011 5/19 10/6 22.2 SS405 2009 5/28 10/14 22.2 SS405 2008 5/27 10/26 22.2 SS405 2007 5/30 9/25 20.0 SS405 2006 5/25 9/28 29.1 SS405 2005 5/25 9/29 26.7 SS405 2004 5/24 9/30 26.7 SS405 2002 5/23 9/27 26.7 SS405 2000 5/24 9/27 26.7 Trudan 8 2006 5/25 8/31 23.8 Trudan 8 2005 5/25 9/1 26.7 Trudan 8 2004 5/24 9/9 26.7 Trudan Headless 2014 6/13 10/6 22.2 Trudan Headless 2008 5/27 10/26 22.2 Trudan Headless 2006 5/25 10/6 24.5 Trudan Headless 2005 5/25 9/29 26.7 Trudan Headless 2004 5/24 10/13 26.7 Trudan Headless 2003 5/21 10/15 26.7 Trudan Headless 2002 5/23 10/11 26.7 Genotype . Year . Sowing date . Harvest date . Stand count (plants per m2) . 849F 2017 6/13 10/4 16.7 849F 2016 6/8 9/15 17.8 849F 2014 6/13 9/8 22.2 849F 2011 5/19 9/2 22.2 849F 2010 5/28 9/7 22.2 849F 2009 5/28 9/9 22.2 849F 2008 5/27 9/22 22.2 849F 2007 5/30 9/25 20.0 877F 2006 5/25 10/6 28.5 HIKANE II 2016 6/8 8/27 17.8 HIKANE II 2011 5/19 9/2 22.2 HIKANE II 2009 5/28 9/16 22.2 HIKANE II 2008 5/27 9/17 22.2 HIKANE II 2007 5/30 9/25 20.0 HIKANE II 2006 5/25 9/11 26.1 HIKANE II 2005 5/25 9/8 26.7 HIKANE II 2004 5/24 9/9 26.7 HIKANE II 2003 5/21 9/5 26.7 HIKANE II 2002 5/23 8/28 26.7 NK 300 2016 6/8 9/26 17.8 NK 300 2011 5/19 9/22 22.2 NK 300 2009 5/28 10/14 22.2 NK 300 2006 5/25 9/14 25.7 NK 300 2004 5/24 9/9 26.7 NK 300 2003 5/21 9/22 26.7 NK 300 2002 5/23 9/27 26.7 Sordan 79 2006 5/25 9/14 25.4 Sordan 79 2005 5/25 9/29 26.7 Sordan 79 2004 5/24 10/13 26.7 Sordan Headless 2016 6/8 10/25 17.8 Sordan Headless 2014 6/13 10/6 22.2 Sordan Headless 2008 5/27 10/26 22.2 Sordan Headless 2006 5/25 10/6 22.5 Sordan Headless 2005 5/25 9/29 26.7 Sordan Headless 2004 5/24 10/13 26.7 Sordan Headless 2003 5/21 10/15 26.7 Sordan Headless 2002 5/23 10/11 26.7 SS405 2017 6/13 10/26 16.7 SS405 2016 6/8 10/15 17.8 SS405 2014 6/13 9/17 22.2 SS405 2011 5/19 10/6 22.2 SS405 2009 5/28 10/14 22.2 SS405 2008 5/27 10/26 22.2 SS405 2007 5/30 9/25 20.0 SS405 2006 5/25 9/28 29.1 SS405 2005 5/25 9/29 26.7 SS405 2004 5/24 9/30 26.7 SS405 2002 5/23 9/27 26.7 SS405 2000 5/24 9/27 26.7 Trudan 8 2006 5/25 8/31 23.8 Trudan 8 2005 5/25 9/1 26.7 Trudan 8 2004 5/24 9/9 26.7 Trudan Headless 2014 6/13 10/6 22.2 Trudan Headless 2008 5/27 10/26 22.2 Trudan Headless 2006 5/25 10/6 24.5 Trudan Headless 2005 5/25 9/29 26.7 Trudan Headless 2004 5/24 10/13 26.7 Trudan Headless 2003 5/21 10/15 26.7 Trudan Headless 2002 5/23 10/11 26.7 Open in new tab Table 3. The genotypes and management details in Bushland trials. Genotype . Year . Sowing date . Harvest date . Stand count (plants per m2) . 849F 2017 6/13 10/4 16.7 849F 2016 6/8 9/15 17.8 849F 2014 6/13 9/8 22.2 849F 2011 5/19 9/2 22.2 849F 2010 5/28 9/7 22.2 849F 2009 5/28 9/9 22.2 849F 2008 5/27 9/22 22.2 849F 2007 5/30 9/25 20.0 877F 2006 5/25 10/6 28.5 HIKANE II 2016 6/8 8/27 17.8 HIKANE II 2011 5/19 9/2 22.2 HIKANE II 2009 5/28 9/16 22.2 HIKANE II 2008 5/27 9/17 22.2 HIKANE II 2007 5/30 9/25 20.0 HIKANE II 2006 5/25 9/11 26.1 HIKANE II 2005 5/25 9/8 26.7 HIKANE II 2004 5/24 9/9 26.7 HIKANE II 2003 5/21 9/5 26.7 HIKANE II 2002 5/23 8/28 26.7 NK 300 2016 6/8 9/26 17.8 NK 300 2011 5/19 9/22 22.2 NK 300 2009 5/28 10/14 22.2 NK 300 2006 5/25 9/14 25.7 NK 300 2004 5/24 9/9 26.7 NK 300 2003 5/21 9/22 26.7 NK 300 2002 5/23 9/27 26.7 Sordan 79 2006 5/25 9/14 25.4 Sordan 79 2005 5/25 9/29 26.7 Sordan 79 2004 5/24 10/13 26.7 Sordan Headless 2016 6/8 10/25 17.8 Sordan Headless 2014 6/13 10/6 22.2 Sordan Headless 2008 5/27 10/26 22.2 Sordan Headless 2006 5/25 10/6 22.5 Sordan Headless 2005 5/25 9/29 26.7 Sordan Headless 2004 5/24 10/13 26.7 Sordan Headless 2003 5/21 10/15 26.7 Sordan Headless 2002 5/23 10/11 26.7 SS405 2017 6/13 10/26 16.7 SS405 2016 6/8 10/15 17.8 SS405 2014 6/13 9/17 22.2 SS405 2011 5/19 10/6 22.2 SS405 2009 5/28 10/14 22.2 SS405 2008 5/27 10/26 22.2 SS405 2007 5/30 9/25 20.0 SS405 2006 5/25 9/28 29.1 SS405 2005 5/25 9/29 26.7 SS405 2004 5/24 9/30 26.7 SS405 2002 5/23 9/27 26.7 SS405 2000 5/24 9/27 26.7 Trudan 8 2006 5/25 8/31 23.8 Trudan 8 2005 5/25 9/1 26.7 Trudan 8 2004 5/24 9/9 26.7 Trudan Headless 2014 6/13 10/6 22.2 Trudan Headless 2008 5/27 10/26 22.2 Trudan Headless 2006 5/25 10/6 24.5 Trudan Headless 2005 5/25 9/29 26.7 Trudan Headless 2004 5/24 10/13 26.7 Trudan Headless 2003 5/21 10/15 26.7 Trudan Headless 2002 5/23 10/11 26.7 Genotype . Year . Sowing date . Harvest date . Stand count (plants per m2) . 849F 2017 6/13 10/4 16.7 849F 2016 6/8 9/15 17.8 849F 2014 6/13 9/8 22.2 849F 2011 5/19 9/2 22.2 849F 2010 5/28 9/7 22.2 849F 2009 5/28 9/9 22.2 849F 2008 5/27 9/22 22.2 849F 2007 5/30 9/25 20.0 877F 2006 5/25 10/6 28.5 HIKANE II 2016 6/8 8/27 17.8 HIKANE II 2011 5/19 9/2 22.2 HIKANE II 2009 5/28 9/16 22.2 HIKANE II 2008 5/27 9/17 22.2 HIKANE II 2007 5/30 9/25 20.0 HIKANE II 2006 5/25 9/11 26.1 HIKANE II 2005 5/25 9/8 26.7 HIKANE II 2004 5/24 9/9 26.7 HIKANE II 2003 5/21 9/5 26.7 HIKANE II 2002 5/23 8/28 26.7 NK 300 2016 6/8 9/26 17.8 NK 300 2011 5/19 9/22 22.2 NK 300 2009 5/28 10/14 22.2 NK 300 2006 5/25 9/14 25.7 NK 300 2004 5/24 9/9 26.7 NK 300 2003 5/21 9/22 26.7 NK 300 2002 5/23 9/27 26.7 Sordan 79 2006 5/25 9/14 25.4 Sordan 79 2005 5/25 9/29 26.7 Sordan 79 2004 5/24 10/13 26.7 Sordan Headless 2016 6/8 10/25 17.8 Sordan Headless 2014 6/13 10/6 22.2 Sordan Headless 2008 5/27 10/26 22.2 Sordan Headless 2006 5/25 10/6 22.5 Sordan Headless 2005 5/25 9/29 26.7 Sordan Headless 2004 5/24 10/13 26.7 Sordan Headless 2003 5/21 10/15 26.7 Sordan Headless 2002 5/23 10/11 26.7 SS405 2017 6/13 10/26 16.7 SS405 2016 6/8 10/15 17.8 SS405 2014 6/13 9/17 22.2 SS405 2011 5/19 10/6 22.2 SS405 2009 5/28 10/14 22.2 SS405 2008 5/27 10/26 22.2 SS405 2007 5/30 9/25 20.0 SS405 2006 5/25 9/28 29.1 SS405 2005 5/25 9/29 26.7 SS405 2004 5/24 9/30 26.7 SS405 2002 5/23 9/27 26.7 SS405 2000 5/24 9/27 26.7 Trudan 8 2006 5/25 8/31 23.8 Trudan 8 2005 5/25 9/1 26.7 Trudan 8 2004 5/24 9/9 26.7 Trudan Headless 2014 6/13 10/6 22.2 Trudan Headless 2008 5/27 10/26 22.2 Trudan Headless 2006 5/25 10/6 24.5 Trudan Headless 2005 5/25 9/29 26.7 Trudan Headless 2004 5/24 10/13 26.7 Trudan Headless 2003 5/21 10/15 26.7 Trudan Headless 2002 5/23 10/11 26.7 Open in new tab The validated models were used to run a long-term simulation for these hybrids from 1980 to 2017 in both locations. In the simulation, we assumed the sowing date for all years in both locations was 1 June and the plant density was 20 plants per m2 with no irrigation in the West Lafayette simulation and with irrigation in the Bushland simulation. The simulation harvest dates were 80, 100 and 120 DAS. 3. RESULTS 3.1 Field conditions The average maximum temperature from sowing to the end of October in 2015, 2017 and 2018 were 26.1, 26.6 and 27.1 °C, respectively. The average minimum temperatures were 13.2, 13.8 and 14.6 °C, respectively. Total precipitation from sowing date to the end of October in 2015, 2017 and 2018 was 471.9, 628.4 and 722.2 mm, respectively, and the crops did not experience water stress. 2015 and 2017 were slightly cooler and dryer years than 2018, but there were no extreme differences between the 3 years. 3.2 Calibration of APSIM models The commercial sorghum hybrids were compared for variations in plant density, flowering date, final dry biomass and max height (Table 1). Significant variations in plant density were detected among hybrids within and between trials. These results demonstrate that plant stand count is an important parameter and should not be replaced by seeding rate. Most of the 18 hybrids flowered at ~75 DAS, except Sordan Headless, Trudan Headless, SP SS405 and RS 400x82 BMR, which exhibited substantially later flowering dates. Analyses of variation in plant height among hybrids revealed that forage sorghum hybrids were taller (average height ~200 cm) while the grain sorghum hybrids were shorter (average height ~100 cm). These differences in morphology between the two types of sorghum represent alternate ideotypes that optimize biomass production versus grain. Final dry biomass was collected on 25 August 2015 (98 DAS), 27 September 2017 (134 DAS) and 9 August 2018 (93 DAS). In all 3 years, SP SS405 exhibited the highest final dry biomass and RS 341x10 exhibited the lowest final dry biomass. In addition to variation in plant development and productivity, the 18 sorghum hybrids also exhibited surprising variations in leaf size distribution (Fig. 1). Maximum leaf area of these hybrids ranged from 300 to 600 cm2. SP SS405 was late-flowering and exhibited the largest maximum leaf area while SP Trudan 8 was an early-flowering type and exhibited the smallest maximum leaf area (Fig. 1; Table 1). For most hybrids, the maximum leaf size occurred close to the middle leaf of the plant (Fig. 1). However, SP Sordan Headless and SP Trudan Headless are photoperiod-sensitive and flower very late in temperate environments (120 to 138 DAS in West Lafayette, respectively; Table 1). During the data collection from 49 to 94 DAS, these two hybrids were in vegetative growth stage and produced more full-size leaves than other hybrids. While the photoperiod-insensitive hybrids exhibit a clear, bell-shaped leaf size distribution with the largest leaf in the middle of the plant, the leaf size distribution for the photoperiod-sensitive hybrids show that each hybrid achieves a near-maximum leaf size at leaf 11 or 12, then continues to produce similar-sized leaves while the plant maintains vegetative growth (Fig. 1). This pattern of development is similar to what has been observed and parameterized for the APSIM sugarcane model (Keating et al. 1999, 2003). Leaf size distributions show that each hybrid has a unique canopy structure. Figure 1. Open in new tabDownload slide Leaf size distributions collected from 25 June (48 DAS), 12 July (65 DAS) and 9 August (93 DAS) at West Lafayette, IN, in 2018. Vertical bars indicate ± 1 SEM for measured values. Figure 1. Open in new tabDownload slide Leaf size distributions collected from 25 June (48 DAS), 12 July (65 DAS) and 9 August (93 DAS) at West Lafayette, IN, in 2018. Vertical bars indicate ± 1 SEM for measured values. The management practices and biophysiological characteristics of each hybrid, including sowing date, sowing depth, plant density, observed final leaf number, final tiller number, two leaf size distribution parameters, leaf number, observed canopy cover and observed biomass were input to the pipeline for the APSIM simulation. The extinction coefficients (k) of the hybrids (Fig. 2) and estimates of RUE (Table 2) indicated that, whether photoperiod-sensitive or -insensitive, forage sorghum hybrids exhibited higher RUE. For k of all 18 hybrids, please see Supporting Information—Fig. S1. Thus, given the same amount of solar radiation, forage sorghum can fix more CO2 and produce more biomass per unit of land compared to dwarf or semi-dwarf grain sorghum hybrids or to sorghum-sudan hybrids used for hay production. Figure 2. Open in new tabDownload slide The canopy cover (CC) versus leaf area index (LAI) for different types of sorghum. The fitted curve (CC = 1 − e−k·LAI) indicates the extinction coefficient (k) of different types of sorghum and the values shown in Table 2. Figure 2. Open in new tabDownload slide The canopy cover (CC) versus leaf area index (LAI) for different types of sorghum. The fitted curve (CC = 1 − e−k·LAI) indicates the extinction coefficient (k) of different types of sorghum and the values shown in Table 2. To evaluate the accuracy of the parameterized and calibrated models, simulated and observed traits were evaluated over years and environments. For LAI, the six hybrids shown in Fig. 3 are representative of hybrids of different types of sorghum that farmers produce. The LAI for all 18 hybrids is in the Supporting Information—Fig. S2. Most of the simulation lines fall within 1 SEM, except under late-season conditions, when LAI is underestimated. Figure 3. Open in new tabDownload slide Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for six represented hybrids of each sorghum type. The experiments were sown on 8 May 2018 at West Lafayette. Vertical bars indicate ± 1 SEM for measured values. Figure 3. Open in new tabDownload slide Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for six represented hybrids of each sorghum type. The experiments were sown on 8 May 2018 at West Lafayette. Vertical bars indicate ± 1 SEM for measured values. Simulations of total plant biomass production and biomass partitioning into leaves, stems and panicles are shown in Fig. 4. The APSIM simulations report green stem and leaf weights; however, senesced and non-senesced leaves and stems were not differentiated in the observed data. Therefore, some leaf and stem simulation results are underestimated in the late-season data points. The simulations of senesced leaves show that the observed leaf dry biomass is close to the simulated green leaf dry biomass plus dead leaf dry biomass. The parameterized APSIM models performed well for most of the different types of sorghum; however, there are some differences between forage sorghum and grain sorghum hybrids. When we consider the stem and leaf dry biomass simulations, the simulations of grain sorghum (Fig. 4, O–R) exhibit a better fit than in the forage sorghum hybrids (Fig. 4, A–N). For the panicle dry biomass simulations, the models perform better for forage sorghum. Figure 4. Open in new tabDownload slide Simulated crop attributes throughout the crop life cycle (lines) compared to measured values (symbols) for a range of treatments for the experiments sown on 8 May 2018 at West Lafayette. Vertical bars indicate ± 1 SEM for measured values. For each forage (A–N) and grain (O–R) type hybrid, the panel shows the time course of total and organ (stem, leaf, grain) biomass. The simulated lines are in the same colour as their measured types except the simulated total dry biomass (black line) and the simulated dead leaf dry weight (brown line). Figure 4. Open in new tabDownload slide Simulated crop attributes throughout the crop life cycle (lines) compared to measured values (symbols) for a range of treatments for the experiments sown on 8 May 2018 at West Lafayette. Vertical bars indicate ± 1 SEM for measured values. For each forage (A–N) and grain (O–R) type hybrid, the panel shows the time course of total and organ (stem, leaf, grain) biomass. The simulated lines are in the same colour as their measured types except the simulated total dry biomass (black line) and the simulated dead leaf dry weight (brown line). 3.3 Validation of APSIM models To validate the parameterized APSIM models over environments, LAI was simulated in West Lafayette using 2015 and 2017 weather data. The models performed well in both years with simulations for six of the hybrids shown in Fig. 5. Model performance of 2015 and 2017 LAI for all 18 hybrids are shown in the Supporting Information—Fig. S3. Leaf area index was overestimated in hybrids with later flowering dates such as SP SS405 and SP Sordan Headless. Figure 5. Open in new tabDownload slide Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for six represented hybrids of each sorghum type sown on 19 May 2015 and 16 May 2017 at West Lafayette. The simulated lines are in the same colour as their measured types. Vertical bars indicate ± 1 SEM for measured values. Figure 5. Open in new tabDownload slide Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for six represented hybrids of each sorghum type sown on 19 May 2015 and 16 May 2017 at West Lafayette. The simulated lines are in the same colour as their measured types. Vertical bars indicate ± 1 SEM for measured values. Given these results, above-ground dry biomass production was simulated for West Lafayette, IN and Bushland, TX representing two very different production environments (Fig. 6). The P-values in the plot test the null hypothesis that the fitted line slope is not different from 1. Only SP SS405 exhibited a slope significantly lower than 1. Figure 6. Open in new tabDownload slide Model validation through comparing observed and predicted biomass of West Lafayette 2015, West Lafayette 2017 and Bushland data from 2000 to 2017. The P-value is to test the null hypothesis that the fitted line slope is not different from 1. Figure 6. Open in new tabDownload slide Model validation through comparing observed and predicted biomass of West Lafayette 2015, West Lafayette 2017 and Bushland data from 2000 to 2017. The P-value is to test the null hypothesis that the fitted line slope is not different from 1. Given that APSIM models can simulate above-ground biomass in multiple years and different regions, the above-ground biomass for nine hybrids was simulated in West Lafayette, IN and Bushland, TX using historical weather data from 1980 to 2017. Results are shown in a biomass probability exceedance plot across years (Fig. 7). Overall, the simulated biomass in West Lafayette, IN was larger than in Bushland, TX for each of three different harvest dates. The patterns of hybrid biomass performance in the two locations differed. Considering the rank performance of hybrids, the ranks over the three harvest dates do not change much in Bushland, TX but show considerable variation from year-to-year in West Lafayette, IN. SP SS405 and the SP Sordan 79 hybrids had the highest simulated biomass, and SP Trudan Headless had the lowest biomass. Under early harvesting conditions in Bushland, TX, PH 849F, PH 877F, SP HIKANE II and SP Sordan Headless had similar simulated biomass production but indicated more variation when harvested later in the season. Plots of simulated biomass production in West Lafayette, IN showed that SP SS405 and SP Sordan 79 had highest simulated biomass yields and the SP Trudan Headless had the lowest simulated biomass at 80 DAS and 100 DAS. However, the hybrids with the highest biomass also have a large range of potential biomass. For example, SP SS405 has potential biomass between 2200 (g m−2) and 3950 (g m−2) at 120 DAS simulation, which has larger range than other hybrids (Fig. 7, F). SP SS405, SP Sordan 79 and SP Sordan Headless had the highest simulated biomass in West Lafayette at 120 DAS. Other hybrids exhibited a similar range of simulated biomass yields. Figure 7. Open in new tabDownload slide Biomass probability exceedance of nine hybrids from 1980 to 2017. The plots from (A) to (C) are harvested on 80, 100 and 120 DAS in Bushland, TX; the plots from (D) to (F) are harvested on 80, 100 and 120 DAS in West Lafayette, IN. Figure 7. Open in new tabDownload slide Biomass probability exceedance of nine hybrids from 1980 to 2017. The plots from (A) to (C) are harvested on 80, 100 and 120 DAS in Bushland, TX; the plots from (D) to (F) are harvested on 80, 100 and 120 DAS in West Lafayette, IN. 4. DISCUSSION 4.1 Plant height and final dry biomass of photoperiod-sensitive and -insensitive forage sorghum hybrids are similar and greater than grain sorghum in medium- and short-season environments Renewable fuels produced from plants could help to ensure future energy sustainability. Different feedstocks are used in starch-based, sugar-based and cellulose-based ethanol production. Whereas starch- and sugar-based ethanol compete with food production (Tilman et al. 2006), lignocellulosic biofuels do not have a potential negative influence on food production (Rubin 2008). Not surprisingly, the yield trials and simulation studies of biomass sorghum hybrids reported in this study showed that photoperiod-sensitive and photoperiod-insensitive forage sorghum hybrids have larger max height and final dry biomass than grain sorghum. This indicates that these types of sorghum can produce more lignocellulosic biomass for ethanol and are better choices as feedstocks compared to grain sorghum. Based on our final harvest data in 2018 (Fig. 4), the proportion of stem to total biomass for forage sorghum and grain sorghum are 0.70 and 0.37, respectively. Variation in maximum height and final dry biomass of these hybrid cultivars depends on the length of the growing season. The final dry biomass of the photoperiod-insensitive forage hybrids was higher than the photoperiod-sensitive sorghum in 2018 at 94 DAS, while the photoperiod-sensitive hybrids outperformed the photoperiod-insensitive hybrids at 99 DAS in 2015 and 135 DAS in 2017 (Table 1). This is consistent with observations that the photoperiod-sensitive sorghum extends pre-floral development up to 8 months, resulting in taller plants with more leaves (Rooney 2004; Rooney et al. 2007; Clerget et al. 2008; Olson et al. 2012). Photoperiod-sensitive sorghum hybrids maximize the yield of lignocellulosic material not only directly through delay of reproductive growth stage but also indirectly through enhancement of drought tolerance or drought avoidance in rainfed environments (Rooney et al. 2007). Our results suggest that photoperiod-sensitive sorghum improves biomass production in longer growing periods by inhibiting the transition from vegetative to reproductive growth, which can add value to bioenergy production in locations that have longer growing periods with sufficiently warm temperatures. 4.2 Sorghum hybrids exhibit diverse canopy structures In conditions of sufficient water supply, the crop biomass is determined by the accumulated radiation interception and the efficiency with which radiant energy is converted to dry matter (Monteith et al. 1977; Muchow 1989). The amount of RI is a function of the pattern of leaf area development. Therefore, leaf size distribution is an important determinant of crop growth. In maize, Hammer et al. (2009) found that the change in canopy architecture may also have indirect effects via leaf area retention and partitioning of carbohydrate to the ear. The leaf size distributions vary considerably among the hybrids reported in this study (Fig. 1). Some hybrids with larger leaf areas may produce more biomass in stress-free environments while hybrids with smaller leaf areas may perform better under drought stress. Hammer et al. (2009) found that crops with smaller leaf area have a yield advantage because they can reduce water use before flowering and conserve subsoil moisture that can then be accessed during the critical grain-filling period under drought stress (He et al. 2017). Borrell et al. (2014a, b) also found that the size of the crop canopy has important consequences for water use in sorghum, where the stay-green trait contributes to drought tolerance by conferring reduced tillering and smaller plant leaf areas before flowering. Photoperiod-sensitive sorghum can achieve higher biomass when there is a longer vegetative growth supporting its potential value as a feedstock for lignocellulosic biofuel. Photoperiod-sensitive sorghum hybrids exhibit a unique pattern of leaf size distribution (Fig. 1). These hybrids remain vegetative throughout the growing season and do not produce a flag leaf or have a clear maximum leaf in the leaf size distributions. These hybrids continued growing and producing more leaves until the last harvest date in 2018 at 94 DAS. This pattern may explain why the photoperiod-sensitive sorghum had larger final dry biomass when harvested at later dates (99 DAS in 2015 and 135 DAS in 2017). 4.3 Photoperiod-sensitive and photoperiod-insensitive forage sorghum hybrids exhibit similar RUE Radiation use efficiency is a robust and theoretically appropriate parameter for describing crop growth. The total production of dry matter is strongly correlated with intercepted solar radiation in many different species (Monteith et al. 1977). DeWit (1965) and Goudriaan (1982) found that RUE values are essentially stable throughout the growing season and over a wide range of production conditions for most crop species. Further analyses suggested that RUE is not particularly sensitive to leaf angle even with extreme leaf angles (Duncan 1971). Consistent with these findings, some of the hybrids in this study have relatively high extinction coefficients (k) and still have reasonable RUE (Table 2). In general, RUE is higher for C4 plants than C3 plants; Kiniry et al. (1989) reported the RUE for both C4 plants and C3 plants showing that C4 plants exhibited the highest RUE, with maize at 1.75 g MJ−1 and sorghum at 1.4 g MJ−1 of intercepted short-wave solar radiation. Other studies have shown maximum RUE of maize in the range 1.6–1.7 g MJ−1 during vegetative growth and 1.2–1.4 g MJ−1 for sorghum during vegetative growth, suggesting the range of potential RUE for sorghum is less than that of maize (Muchow and Davis 1988; Muchow 1989; Muchow and Sinclair 1994; Sinclair and Muchow 1999; Lindquist et al. 2005). Most of the RUE studies in sorghum are for grain cultivars; however, our studies in photoperiod-sensitive and photoperiod-insensitive forage hybrids showed that these hybrids have similar RUE to one another and higher RUEs than reported for grain sorghums. Within commercial forage sorghum hybrids, the observed RUE ranged from 1.29 to 1.70 g MJ−1 with the highest RUE similar to reports in maize (Sinclair and Muchow 1999). The sorghum hybrid with highest RUE of the 18 commercial grain and biomass sorghum hybrids in our studies was SP SS405 (Table 2). This hybrid also exhibits a larger max height and greater final dry biomass. Other studies have reported similar findings of tall sorghum hybrids exhibiting 1.65 g MJ−1 RUE (Hammer et al. 2010). Narayanan et al. (2013) also reported that two taller sorghum hybrids had the highest biomass and RUE in their study. Conversely, to test the hypothesis that height affects RUE in sorghum, George-Jaeggli et al. (2011) used dwarf sorghum to examine the effects of plant height on RUE. They found that sorghum dwarfing genes negatively affect radiation capture and in some cases RUE. 4.4 Forage sorghum models perform well in above-ground biomass simulations across years and locations The forage and grain sorghum biomass models described in this study performed well in simulations in both West Lafayette, IN and Bushland, TX. These studies showed that the simulated above-ground biomass was higher in West Lafayette than in Bushland over multiple years. Within the set of nine hybrids evaluated at both locations, SP SS405 and SP Sordan 79 exhibited the highest RUE and simulated biomass in both locations. The photoperiod-sensitive sorghum hybrids exhibited the highest predicted biomass yields over time. Interestingly, the photoperiod-sensitive sorghum hybrids did not perform as well in Bushland as in West Lafayette. This may be because West Lafayette has comparatively higher rainfall, and the photoperiod-sensitive sorghum hybrids had more vegetative growing time to produce biomass in West Lafayette than in Bushland. The APSIM models reported in this study can be used to explore differences in productivity among sorghum hybrids through long-term simulation. Hammer et al. (2014) have used APSIM to study locally optimal G × M combinations and demonstrated that significant improvements in yield and or reduction in failure risk are possible. Hammer et al. (2009) used the past 50 years of climate data to simulate canopy and root system architecture effects for maize that was planted at a range of densities at three representative locations throughout the US Corn Belt. Their results indicated that change in canopy architecture had little direct effect on biomass accumulation and historical yield trends, but likely had important, indirect effects via leaf area retention and partitioning of carbohydrate to the ear (Hammer et al. 2009). Applying the APSIM model to sorghum can have similar benefits. White et al. (2015) simulated a rainfed sorghum–winter wheat rotation at Bushland, TX, from 1958 to 1999 comparing no-till versus tillage. The simulated grain sorghum biomass was lower than the one observed. Agricultural Production Systems sIMulator should also be able to improve mid-season predictions of yield. Soler et al. (2007) used CERES-Maize to simulate the impacts of different planting dates on four different maize hybrids under rainfed and irrigated conditions in a subtropical region of Brazil. These studies showed that an accurate yield forecast could be provided at ~45 days prior to the harvest date for all four maize hybrids (Soler et al. 2007). These kinds of studies are promising for farmers, decision makers and researchers, as they could provide longer-term information for strategic management decisions, without extensive yield trials. In the future, our adapted biomass sorghum models can be applied to diverse areas and provide credible simulations for sorghum crop growth and development across a range of environments and management practices. SUPPORTING INFORMATION The following additional information is available in the online version of this article—Figure S1. The canopy cover (CC) versus leaf area index (LAI) for 18 sorghum hybrids. The fitted curve (CC = 1 − e−k·LAI) indicates the extinction coefficient (k) of different types of sorghum and the values shown in Table 3. Figure S2. Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for all sorghum hybrids of each sorghum type. The experiments were sown on 8 May 2018 at West Lafayette. Vertical bars indicate ± 1 SEM for measured values. Figure S3. Simulated crop leaf area index (LAI) throughout the crop life cycle (lines) compared to measured values (symbols) for all sorghum hybrids of each sorghum type sown on 19 May 2015 and 16 May 2017 at West Lafayette. The simulated lines are in the same colour as their measured types. Vertical bars indicate ± 1 SEM for measured values. DATA AVAILABILITY The ‘R Pipeline for Calculation of APSIM Parameters and Generating the XML File’ is stored at the Purdue University Research Repository and includes the data processing pipeline, data for model input parameters and outputs comparisons, and R-codes for generating or processing central data sets (Yang et al. 2020a). The APSIM files used in the model calibration procedures are stored at the Purdue University Research Repository in ‘2018 West Lafayette Simulation of 18 Sorghum Hybrids’ (Yang et al. 2020b). The APSIM files used for model validations are stored at the Purdue University Research Repository in the ‘2015 West Lafayette Simulation of 18 Sorghum Hybrids’ (Yang et al. 2020c) and ‘2017 West Lafayette Simulation of 18 Sorghum Hybrids’ (Yang et al. 2020d). The APSIM files used for the scenario simulations are stored at the Purdue University Research Repository in the ‘Texas Simulation of Sorghum Hybrids Using Historical Weather Data’ (Yang et al. 2020e) and ‘West Lafayette Scenario Simulation of Sorghum Hybrids Using Historical Weather Data’ (Yang et al. 2020f) using multi-year historical weather data of Bushland, TX, and West Lafayette, IN. ACKNOWLEDGEMENTS We would like to thank the plant phenomics and crop modelling teams at Purdue University and The University of Queensland for support in agronomic trials and remote-sensing studies. We also thank Al Doherty for technical support of analyses in R and APSIM. SOURCES OF FUNDING This project was funded by U.S. Department of Energy ARPA-E Program Award Number DE-AR0001135. CONTRIBUTIONS BY THE AUTHORS K.-W. Yang contributed to data collection, data analysis and writing (original draft). S.C. contributed to conceptualization, software, writing (review & editing), supervision and methodology. N.C. contributed to data collection and data analysis. G.H. contributed to conceptualization, software and resources. G.M. and B.Z. contributed to technical support of analyses in R and APSIM. Y.C. and E.D. contributed to extraction of canopy cover from RGB images. A.M., M.C., D.E. and A.H. contributed to remote-sensing data collection and processing as well as writing (review & editing). A.T. contributed to establishment of the preliminary data collection and analysis protocol. C.W. contributed to generation of the plant materials used in this study and edited the manuscript. M.R.T. contributed to conceptualization, writing (review & editing), supervision, project administration, funding acquisition, methodology and resources. CONFLICT OF INTEREST None declared. LITERATURE CITED Baret F , Madec S, Irfan K, Lopez J, Comar A, Hemmerlé M, Dutartre D, Praud S, Tixier MH. 2018 . Leaf-rolling in maize crops: from leaf scoring to canopy-level measurements for phenotyping . Journal of Experimental Botany 69 : 2705 – 2716 . Google Scholar OpenURL Placeholder Text WorldCat Blancon J , Dutartre D, Tixier MH, Weiss M, Comar A, Praud S, Baret F. 2019 . A high-throughput model-assisted method for phenotyping maize green leaf area index dynamics using unmanned aerial vehicle imagery . Frontiers in Plant Science 10 : 685 . Google Scholar OpenURL Placeholder Text WorldCat Borrell AK , Mullet JE, George-Jaeggli B, van Oosterom EJ, Hammer GL, Klein PE, Jordan DR. 2014a . Drought adaptation of stay-green sorghum is associated with canopy development, leaf anatomy, root growth, and water uptake . Journal of Experimental Botany 65 : 6251 – 6263 . Google Scholar OpenURL Placeholder Text WorldCat Borrell AK , van Oosterom EJ, Mullet JE, George-Jaeggli B, Jordan DR, Klein PE, Hammer GL. 2014b . Stay-green alleles individually enhance grain yield in sorghum under drought by modifying canopy development and water uptake patterns . The New Phytologist 203 : 817 – 830 . Google Scholar OpenURL Placeholder Text WorldCat Bouman BAM , van Keulen H, van Laar HH, Rabbinge R. 1996 . The ‘School of de Wit’ crop growth simulation models: a pedigree and historical overview . Agricultural Systems 52 : 171 – 198 . Google Scholar OpenURL Placeholder Text WorldCat Campbell , K . 2006 . Remote sensing . In: El‐Shaarawi AH, Piegorsch WW, Guttorp P, eds. Encyclopedia of environmetrics . John Wiley & Sons, Ltd . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Carberry PS , Muchow RC, Hammer GL. 1993 . Modelling genotypic and environmental control of leaf area dynamics in grain sorghum. II. Individual leaf level - ScienceDirect . Field Crops Research 33 : 311 – 328 . Google Scholar OpenURL Placeholder Text WorldCat Casa R , Baret F, Buis S, Lopez-Lozano R, Pascucci S, Palombo A, Jones HG. 2010 . Estimation of maize canopy properties from remote sensing by inversion of 1-D and 4-D models . Precision Agriculture 11 : 319 – 334 . Google Scholar OpenURL Placeholder Text WorldCat Charles-Edwards DA . 1982 . Physiological determinants of crop growth , vol. 1 . London : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chenu K , Chapman SC, Hammer GL, McLean G, Salah HB, Tardieu F. 2008 . Short-term responses of leaf growth rate to water deficit scale up to whole-plant and crop levels: an integrated modelling approach in maize . Plant, Cell & Environment 31 : 378 – 391 . Google Scholar OpenURL Placeholder Text WorldCat Chenu K , van Oosterom EJ, McLean G, Deifel KS, Fletcher A, Geetika G, Tirfessa A, Mace ES, Jordan DR, Sulman R, Hammer GL. 2018 . Integrating modelling and phenotyping approaches to identify and screen complex traits: transpiration efficiency in cereals . Journal of Experimental Botany 69 : 3181 – 3194 . Google Scholar OpenURL Placeholder Text WorldCat Clerget B , Dingkuhn M, Gozé E, Rattunde HF, Ney B. 2008 . Variability of phyllochron, plastochron and rate of increase in height in photoperiod-sensitive sorghum varieties . Annals of Botany 101 : 579 – 594 . Google Scholar OpenURL Placeholder Text WorldCat Cobb JN , DeClerck G, Greenberg A, Clark R, McCouch S. 2013 . Next-generation phenotyping: requirements and strategies for enhancing our understanding of genotype–phenotype relationships and its relevance to crop improvement . Theoretical and Applied Genetics 126 : 867 – 887 . Google Scholar OpenURL Placeholder Text WorldCat Demarez V , Duthoit S, Baret F, Weiss M, Dedieu G. 2008 . Estimation of leaf area and clumping indexes of crops with hemispherical photographs . Agricultural and Forest Meteorology 148 : 644 – 655 . Google Scholar OpenURL Placeholder Text WorldCat de DeWit CT . 1965 . Photosynthesis of leaf canopies . Wageningen, The Netherlands : Pudoc . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC van Diepen CA , Wolf J, van Keulen H, Rappoldt C. 1989 . WOFOST: a simulation model of crop production . Soil Use and Management 5 : 16 – 24 . Google Scholar OpenURL Placeholder Text WorldCat Duan T , Chapman SC, Guo Y, Zheng B. 2017 . Dynamic monitoring of NDVI in wheat agronomy and breeding trials using an unmanned aerial vehicle . Field Crops Research 210 : 71 – 80 . Google Scholar OpenURL Placeholder Text WorldCat Duan T , Chapman SC, Holland E, Rebetzke GJ, Guo Y, Zheng B. 2016 . Dynamic quantification of canopy structure to characterize early plant vigour in wheat genotypes . Journal of Experimental Botany 67 : 4523 – 4534 . Google Scholar OpenURL Placeholder Text WorldCat Duan S-B , Li Z-L, Wu H, Tang B-H, Ma L, Zhao E, Li C. 2014 . Inversion of the PROSAIL model to estimate leaf area index of maize, potato, and sunflower fields from unmanned aerial vehicle hyperspectral data . International Journal of Applied Earth Observation and Geoinformation 26 : 12 – 20 . Google Scholar OpenURL Placeholder Text WorldCat Duncan WG . 1971 . Leaf angles, leaf area, and canopy photosynthesis 1 . Crop Science 11 : 482 – 485 . Google Scholar OpenURL Placeholder Text WorldCat Furbank RT , Tester M. 2011 . Phenomics–technologies to relieve the phenotyping bottleneck . Trends in Plant Science 16 : 635 – 644 . Google Scholar OpenURL Placeholder Text WorldCat George-Jaeggli B , Jordan DR, van Oosterom EJ, Hammer GL. 2011 . Decrease in sorghum grain yield due to the dw3 dwarfing gene is caused by reduction in shoot biomass . Field Crops Research 124 : 231 – 239 . Google Scholar OpenURL Placeholder Text WorldCat Gill JR , Burks PS, Staggenborg SA, Odvody GN, Heiniger RW, Macoon B, Moore KJ, Barrett M, Rooney WL. 2014 . Yield results and stability analysis from the sorghum regional biomass feedstock trial . BioEnergy Research 7 : 1026 – 1034 . Google Scholar OpenURL Placeholder Text WorldCat Gouache D , Beauchêne K, Mini A, Fournier A, de Solan B, Baret F, Comar A. 2016 . Applying remote sensing expertise to crop improvement: progress and challenges to scale up high throughput field phenotyping from research to industry . In: Autonomous air and ground sensing systems for agricultural optimization and phenotyping , vol. 9866 . Baltimore, USA : International Society for Optics and Photonics , 986604 . doi:10.1117/12.2229389. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Goudriaan J . 1982 . Potential production processes . In: Penning de Vries FWT, van Laar HH, eds. Simulation of plant growth and crop production . Wageningen : Pudoc , 98 – 113 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Guo W , Zheng B, Duan T, Fukatsu T, Chapman S, Ninomiya S. 2017 . EasyPCC: benchmark datasets and tools for high-throughput measurement of the plant canopy coverage ratio under field conditions . Sensors 17 : 798 . Google Scholar OpenURL Placeholder Text WorldCat Hammer GL , Carberry PS, Muchow RC. 1993 . Modelling genotypic and environmental control of leaf area dynamics in grain sorghum. I. Whole plant level - ScienceDirect . Field Crops Research 33 : 293 – 310 . Google Scholar OpenURL Placeholder Text WorldCat Hammer GL , Dong Z, McLean G, Doherty A, Messina C, Schussler J, Zinselmeier C, Paszkiewicz S, Cooper M. 2009 . Can changes in canopy and/or root system architecture explain historical maize yield trends in the U.S. Corn Belt? Crop Science 49 : 299 – 312 . Google Scholar OpenURL Placeholder Text WorldCat Hammer GL , McLean G, Chapman S, Zheng B, Doherty A, Harrison MT, van Oosterom E, Jordan D. 2014 . Crop design for specific adaptation in variable dryland production environments . Crop and Pasture Science 65 : 614 – 626 . Google Scholar OpenURL Placeholder Text WorldCat Hammer GL , Muchow RC. 1994 . Assessing climatic risk to sorghum production in water-limited subtropical environments I. Development and testing of a simulation model - ScienceDirect . Field Crops Research 36 : 221 – 234 . Google Scholar OpenURL Placeholder Text WorldCat Hammer GL , van Oosterom E, McLean G, Chapman SC, Broad I, Harland P, Muchow RC. 2010 . Adapting APSIM to model the physiology and genetics of complex adaptive traits in field crops . Journal of Experimental Botany 61 : 2185 – 2202 . Google Scholar OpenURL Placeholder Text WorldCat He J , Du Y-L, Wang T, Turner NC, Yang R-P, Jin Y, Xi Y, Zhang C, Cui T, Fang X-W, Li F.-M. 2017 . Conserved water use improves the yield performance of soybean (Glycine max (L.) Merr.) under drought . Agricultural Water Management 179 : 236 – 245 . Google Scholar OpenURL Placeholder Text WorldCat He J , Zhao X, Laroche A, Lu Z-X, Liu H, Li Z. 2014 . Genotyping-by-sequencing (GBS), an ultimate marker-assisted selection (MAS) tool to accelerate plant breeding . Frontiers in Plant Science 5 : 484 . Google Scholar OpenURL Placeholder Text WorldCat He F , Zhou T, Xiong W, Hasheminnasab SM, Habib A. 2018 . Automated aerial triangulation for UAV-based mapping . Remote Sensing 10 : 1952 . Google Scholar OpenURL Placeholder Text WorldCat Holzworth DP , Huth NI, deVoil PG, Zurcher EJ, Herrmann NI, McLean G, Chenu K, Oosterom EJ van, Snow V, Murphy C, Moore AD, Brown H, Whish JPM, Verrall S, Fainges J, Bell LW, Peake AS, Poulton PL, Hochman Z, Thorburn PJ, Gaydon DS, Dalgliesh NP, Rodriguez D, Cox H, Chapman S, Doherty A, Teixeira E, Sharp J, Cichota R, Vogeler I, Li FY, Wang E, Hammer GL, Robertson MJ, Dimes JP, Whitbread AM, Hunt J, Rees H van, McClelland T, Carberry PS, Hargreaves JNG, MacLeod N, McDonald C, Harsdorf J, Wedgwood S, Keating BA. 2014 . APSIM – Evolution towards a new generation of agricultural systems simulation . Environmental Modelling & Software 62 : 327 – 350 . Google Scholar OpenURL Placeholder Text WorldCat Holzworth D , Huth NI, Fainges J, Brown H, Zurcher E, Cichota R, Verrall S, Herrmann NI, Zheng B, Snow V. 2018 . APSIM Next Generation: overcoming challenges in modernising a farming systems model . Environmental Modelling & Software 103 : 43 – 51 . Google Scholar OpenURL Placeholder Text WorldCat Jiang R , He W, Zhou W, Hou Y, Yang JY, He P. 2019 . Exploring management strategies to improve maize yield and nitrogen use efficiency in northeast China using the DNDC and DSSAT models . Computers and Electronics in Agriculture 166 : 104988 . Google Scholar OpenURL Placeholder Text WorldCat Jimenez-Berni JA , Deery DM, Rozas-Larraondo P, Condon ATG, Rebetzke GJ, James RA, Bovill WD, Furbank RT, Sirault XRR. 2018 . High throughput determination of plant height, ground cover, and above-ground biomass in wheat with LiDAR . Frontiers in Plant Science 9 : 237 . Google Scholar OpenURL Placeholder Text WorldCat Jones JW , Antle JM, Basso B, Boote KJ, Conant RT, Foster I, Godfray HCJ, Herrero M, Howitt RE, Janssen S, Keating BA, Munoz-Carpena R, Porter CH, Rosenzweig C, Wheeler TR. 2017 . Brief history of agricultural systems modeling . Agricultural Systems 155 : 240 – 254 . Google Scholar OpenURL Placeholder Text WorldCat Jones JW , Hoogenboom G, Porter CH, Boote KJ, Batchelor WD, Hunt LA, Wilkens PW, Singh U, Gijsman AJ, Ritchie JT. 2003 . The DSSAT cropping system model . European Journal of Agronomy 18 : 235 – 265 . Google Scholar OpenURL Placeholder Text WorldCat Keating BA , Carberry PS, Hammer GL, Probert ME, Robertson MJ, Holzworth D, Huth NI, Hargreaves JNG, Meinke H, Hochman Z, McLean G, Verburg K, Snow V, Dimes JP, Silburn M, Wang E, Brown S, Bristow KL, Asseng S, Chapman S, McCown RL, Freebairn DM, Smith CJ. 2003 . An overview of APSIM, a model designed for farming systems simulation . European Journal of Agronomy 18 : 267 – 288 . Google Scholar OpenURL Placeholder Text WorldCat Keating BA , Robertson MJ, Muchow RC, Huth NI. 1999 . Modelling sugarcane production systems I. Development and performance of the sugarcane module . Field Crops Research 61 : 253 – 271 . Google Scholar OpenURL Placeholder Text WorldCat Kiniry JR , Jones CA, O’Toole JC, Blanchet R, Cabelguenne M, Spanel DA. 1989 . Radiation-use efficiency in biomass accumulation prior to grain-filling for five grain-crop species . Field Crops Research 20 : 51 – 64 . Google Scholar OpenURL Placeholder Text WorldCat Konare D , Pierre S, Weng JY, Morand E. 2003 . Real-time image processing for remote sensing . In: CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No. 03CH37436) , vol. 2 . Montreal, Quebec, Canada , 699 – 702 . doi:10.1109/CCECE.2003.1225991. Lafarge TA , Hammer GL. 2002 . Predicting plant leaf area production: shoot assimilate accumulation and partitioning, and leaf area ratio, are stable for a wide range of sorghum population densities . Field Crops Research 77 : 137 – 151 . Google Scholar OpenURL Placeholder Text WorldCat Lindquist JL , Arkebauer TJ, Walters DT, Cassman KG, Dobermann A. 2005 . Maize radiation use efficiency under optimal growth conditions . Agronomy Journal 97 : 72 – 78 . Google Scholar OpenURL Placeholder Text WorldCat Masjedi A , Carpenter NR, Crawford MM, Tuinstra MR. 2019 . Prediction of sorghum biomass using UAV time series data and recurrent neural networks . In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops , Long Beach, CA, USA , 16 – 20 . Masjedi A , Zhao J, Thompson AM, Yang K-W, Flatt JE, Crawford MM, Ebert DS, Tuinstra MR, Hammer G, Chapman S. 2018 . Sorghum biomass prediction using UAV-based remote sensing data and crop model simulation . In: IGARSS 2018 – 2018 IEEE International Geoscience and Remote Sensing Symposium , Valencia, Spain , 22 – 27 July, 7719 – 7722 . Monteith JL , Moss CJ, William CG, Wingate PN, Hutton BGD. 1977 . Climate and the efficiency of crop production in Britain . Philosophical Transactions of the Royal Society of London. B, Biological Sciences 281 : 277 – 294 . Google Scholar OpenURL Placeholder Text WorldCat Muchow RC . 1989 . Comparative productivity of maize, sorghum and pearl millet in a semi-arid tropical environment I. Yield potential . Field Crops Research 20 : 191 – 205 . Google Scholar OpenURL Placeholder Text WorldCat Muchow RC , Carberry PS. 1990 . Phenology and leaf-area development in a tropical grain sorghum - ScienceDirect . Field Crops Research 23 : 221 – 237 . Google Scholar OpenURL Placeholder Text WorldCat Muchow RC , Davis R. 1988 . Effect of nitrogen supply on the comparative productivity of maize and sorghum in a semi-arid tropical environment II. Radiation interception and biomass accumulation . Field Crops Research 18 : 17 – 30 . Google Scholar OpenURL Placeholder Text WorldCat Muchow RC , Sinclair TR. 1994 . Nitrogen response of leaf photosynthesis and canopy radiation use efficiency in field-grown maize and sorghum . Crop Science 34 : 721 – 727 . Google Scholar OpenURL Placeholder Text WorldCat Narayanan S , Aiken RM, Vara Prasad PV, Xin Z, Yu J. 2013 . Water and radiation use efficiencies in sorghum . Agronomy Journal 105 : 649 . Google Scholar OpenURL Placeholder Text WorldCat Olson SN , Ritter K, Rooney W, Kemanian A, McCarl BA, Zhang Y, Hall S, Packer D, Mullet J. 2012 . High biomass yield energy sorghum: developing a genetic model for C4 grass bioenergy crops . Biofuels, Bioproducts and Biorefining 6 : 640 – 655 . Google Scholar OpenURL Placeholder Text WorldCat van Oosterom EJ , Carberry PS, O’Leary GJ. 2001 . Simulating growth, development, and yield of tillering pearl millet: I. Leaf area profiles on main shoots and tillers - ScienceDirect . Field Crops Research 72 : 51 – 66 . Google Scholar OpenURL Placeholder Text WorldCat Parent B , Millet EJ, Tardieu F. 2019 . The use of thermal time in plant studies has a sound theoretical basis provided that confounding effects are avoided . Journal of Experimental Botany 70 : 2359 – 2370 . Google Scholar OpenURL Placeholder Text WorldCat Piñeiro G , Perelman S, Guerschman J, Paruelo J. 2008 . How to evaluate models: observed vs. predicted or predicted vs. observed? Ecological Modelling 216 : 316 – 322 . Google Scholar OpenURL Placeholder Text WorldCat Ravi R , Lin Y-J, Elbahnasawy M, Shamseldin T, Habib A. 2018 . Bias impact analysis and calibration of terrestrial mobile LiDAR system with several spinning multibeam laser scanners . IEEE Transactions on Geoscience and Remote Sensing 56 : 5261 – 5275 . Google Scholar OpenURL Placeholder Text WorldCat Ravi Kumar S , Hammer G, Broad I, Harland P, McLean G. 2009 . Modelling environmental effects on phenology and canopy development of diverse sorghum genotypes - ScienceDirect . Field Crops Research 111 : 157 – 165 . Google Scholar OpenURL Placeholder Text WorldCat Ribera J , He F, Chen Y, Habib AF, Delp EJ. 2018 . Estimating phenotypic traits from UAV based RGB imagery . In: Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Workshop on Data Science for Food, Energy, and Water , August 2016, San Francisco, CA . Rocateli AC , Raper RL, Balkcom KS, Arriaga FJ, Bransby DI. 2012 . Biomass sorghum production and components under different irrigation/tillage systems for the southeastern U.S . Industrial Crops and Products 36 : 589 – 598 . Google Scholar OpenURL Placeholder Text WorldCat Rodríguez-Álvarez MX , Boer MP, van Eeuwijk FA, Eilers PHC. 2018 . Correcting for spatial heterogeneity in plant breeding experiments with P-splines . Spatial Statistics 23 : 52 – 71 . Google Scholar OpenURL Placeholder Text WorldCat Rooney WL . 2004 . Sorghum improvement-integrating traditional and new technology to produce improved genotypes . Advances in agronomy 83 : S0065 – 2113 . Google Scholar OpenURL Placeholder Text WorldCat Rooney WL , Blumenthal J, Bean B, Mullet JE. 2007 . Designing sorghum as a dedicated bioenergy feedstock . Biofuels, Bioproducts and Biorefining 1 : 147 – 157 . Google Scholar OpenURL Placeholder Text WorldCat Rubin EM . 2008 . Genomics of cellulosic biofuels . Nature 454 : 841 – 845 . Google Scholar OpenURL Placeholder Text WorldCat Schneider CA , Rasband WS, Eliceiri KW. 2012 . NIH Image to ImageJ: 25 years of image analysis . Nature Methods 9 : 671 – 675 . Google Scholar OpenURL Placeholder Text WorldCat Sieglinger JB . 1936 . Leaf number of sorghum stalks1 . Agronomy Journal 28 : 636 – 642 . Google Scholar OpenURL Placeholder Text WorldCat Sinclair TR , Muchow RC. 1999 . Radiation use efficiency . Advances in Agronomy 65 : 215 – 265 . Google Scholar OpenURL Placeholder Text WorldCat Soler CMT , Sentelhas PC, Hoogenboom G. 2007 . Application of the CSM-CERES-Maize model for planting date evaluation and yield forecasting for maize grown off-season in a subtropical environment . European Journal of Agronomy 27 : 165 – 177 . Google Scholar OpenURL Placeholder Text WorldCat Stanton C , Starek MJ, Elliott N, Brewer M, Maeda MM, Chu T. 2017 . Unmanned aircraft system-derived crop height and normalized difference vegetation index metrics for sorghum yield and aphid stress assessment . Journal of Applied Remote Sensing 11 : 026035 . Google Scholar OpenURL Placeholder Text WorldCat Stöckle CO , Donatelli M, Nelson R. 2003 . CropSyst, a cropping systems simulation model . European Journal of Agronomy 18 : 289 – 307 . Google Scholar OpenURL Placeholder Text WorldCat Tilman D , Hill J, Lehman C. 2006 . Carbon-negative biofuels from low-input high-diversity grassland biomass . Science 314 : 1598 – 1600 . Google Scholar OpenURL Placeholder Text WorldCat Truong SK , McCormick RF, Mullet JE. 2017 . Bioenergy sorghum crop model predicts VPD-limited transpiration traits enhance biomass yield in water-limited environments . Frontiers in Plant Science 8 : 335 . Google Scholar OpenURL Placeholder Text WorldCat Wallach D , Makowski D, Jones JW, Brun F. 2018 . Working with dynamic crop models: methods, tools and examples for agriculture and environment , 3rd edn. Academic Press , 613 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wang E , Robertson MJ, Hammer GL, Carberry PS, Holzworth D, Meinke H, Chapman SC, Hargreaves JNG., Huth NI, McLean G. 2002 . Development of a generic crop model template in the cropping system model APSIM . European Journal of Agronomy 18 : 121 – 140 . Google Scholar OpenURL Placeholder Text WorldCat White JW , Alagarswamy G, Ottman MJ, Porter CH, Singh U, Hoogenboom G. 2015 . An overview of CERES–sorghum as implemented in the cropping system model version 4.5 . Agronomy Journal 107 : 1987 – 2002 . Google Scholar OpenURL Placeholder Text WorldCat Williams JR , Jones CA, Kiniry JR, Spanel DA. 1989 . The EPIC crop growth model . Transactions of the ASAE 32 : 0497 – 0511 . Google Scholar OpenURL Placeholder Text WorldCat Williams JR , Renard KG, Dyke PT. 1983 . EPIC: a new method for assessing erosion’s effect on soil productivity . Journal of Soil and Water Conservation 38 : 381 – 383 . Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020a . R pipeline for calculation of APSIM parameters and generating the XML file . Purdue University Research Repository . doi:10.4231/69H7-CV75 Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020b . 2018 West Lafayette simulation of 18 sorghum hybrids . Purdue University Research Repository . doi:10.4231/KMK0-J993 Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020c . 2015 West Lafayette simulation of 18 sorghum hybrids . Purdue University Research Repository . doi:10.4231/0NX5-RT34 Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020d . 2017 West Lafayette simulation of 18 sorghum hybrids . Purdue University Research Repository . doi:10.4231/6NW4-TB31 Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020e . Texas scenario simulation of sorghum hybrids using historical weather data . Purdue University Research Repository . doi:10.4231/PRS2-AC22 Google Scholar OpenURL Placeholder Text WorldCat Yang K , Chapman S, Tuinstra M. 2020f . West Lafayette scenario simulation of sorghum hybrids using historical weather data . Purdue University Research Repository . doi:10.4231/63GJ-CJ23 Google Scholar OpenURL Placeholder Text WorldCat Zhang Z , Masjedi A, Zhao J, Crawford MM. 2017 . Prediction of sorghum biomass based on image based features derived from time series of UAV images . In: 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) , Fort Worth, TX, USA , 23 – 28 July, 6154–6157. Zhou C , Ye H, Xu Z, Hu J, Shi X, Hua S, Yue J, Yang G. 2019 . Estimating maize-leaf coverage in field conditions by applying a machine learning algorithm to UAV remote sensing images . Applied Sciences 9 : 2389 . Google Scholar OpenURL Placeholder Text WorldCat © The Author(s) 2021. Published by Oxford University Press on behalf of the Annals of Botany Company. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
TI - Integrating crop growth models with remote sensing for predicting biomass yield of sorghum
JF - in silico Plants
DO - 10.1093/insilicoplants/diab001
DA - 2021-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/integrating-crop-growth-models-with-remote-sensing-for-predicting-N0AeSlYsio
VL - 3
IS - 1
DP - DeepDyve
ER -