|Zhang, Bangbang - China Agricultural University|
|Kong, Xiangbin - China Agricultural University|
|Ouyang, Ying - US Department Of Agriculture (USDA)|
Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/22/2016
Publication Date: 9/1/2016
Publication URL: https://handle.nal.usda.gov/10113/5496421
Citation: Zhang, B., Feng, G.G., Read, J.J., Kong, X., Ouyang, Y., Adeli, A., Jenkins, J.N. 2016. Simulating soybean productivity under rainfed conditions for major soil types using APEX model in East Central Mississippi. Agricultural Water Management. 177:379-391.
Interpretive Summary: Under rainfed conditions from 2002 to 2014, simulated soybean grain yield ranged broadly from 2.24 to 6.14 Mg ha-1 across nine soil types, and the highest yield difference reached to 3.90 Mg ha-1. The average grain yield for wet, normal and dry years was 4.88, 4.51 and 3.74 Mg ha-1 respectively. The difference of average yield between wet and dry years was as much as 1.14 Mg ha-1. High yield in wet years was associated with less yield variation, as standard deviation (STDEV) was 0.63, 0.7 and 0.82 for wet, normal and dry category years, respectively. Apparently, with more rainfall during the growing season, soybean was capable of increased productivity and less yield variation on the nine soil types tested. A simulated grain yield range between 4.0 and 6.2 Mg ha-1 was found mainly in wet and normal categories years, accounted for 92% and 83% of yield observations, respectively; whereas, this yield range accounted for 44% of observations in dry category years. For the relatively low yield range of =2.2~4 Mg ha-1, approximately 55% of observations were distributed in dry years, 17% were distributed in normal years, and 8% were distributed in wet years. At the same time, the wet years had no distribution in the lowest yield range =2.2~3 Mg ha-1 and the dry years had no distribution in higher yield range =5.0~6.2 Mg ha-1. In general, “average yield of nine types of soils in each category year (AYNS)” for a wet or a normal category years ranged from 4.41 to 5.42 Mg ha-1 except for normal year 2010 with 3.68 Mg ha-1. While the dry years had the lowest three AYNS ranging from 3.55 to 4.06 Mg ha-1. When results were averaged across 13 years, Vaiden, Leeper and Catalpa had the highest grain yield among, with values of 4.97, 4.87 and 4.77 Mg ha-1, respectively; whereas, Demopolis, Sumter and Griffith had the lowest yield, with values of 3.93, 3.86 and 3.71 Mg ha-1, respectively. Moderate yield levels that ranged from 5.69 to 5.83 Mg ha-1 were simulated for Brooksville, Okolona and Kipling soil types. Hence, the average yield of 13years had great difference among nine soil types, ranging from 3.71 to 4.97 Mg ha-1, the yield gap reach to 1.26 Mg ha-1. Soil types of Vaiden, Leeper and Catalpa had the highest average yield for wet years from 5.17 to 5.34 Mg ha-1, normal years from 4.88 to 4.89 Mg ha-1 and dry years from 4.02 to 4.64 Mg ha-1. Demopolis, Sumter and Griffith had the lowest yield for wet years from 4.22 to 4.41 Mg ha-1, normal years from 3.89 to 4.05 and dry years from 2.69 to 3.16 Mg ha-1. Okolona, Kipling and Demopolis had moderate yield for wet year from 5.01 to 5.09 Mg ha-1, normal years from 4.63 to 4.70 Mg ha-1 and dry years from 3.88 to 4.15 Mg ha-1 . The grain yield difference between wet and normal years (DIFw-n) was very similar, ranging from 0.29 to 0.44 Mg ha-1, while the difference between wet and dry years (DIFw-d) was varied widely from 0.69 to 1.57 Mg ha-1. Results also suggested that soil types of Vaiden, Leeper, Catalpa, Brooksville, Okolona and Kipling had a lower DIFw-d, with values from 0.69 to 1.15 Mg ha-1, whereas soil types of Sumter, Griffith and Demopolis had the bigger DIFw-d with values of 1.18, 1.57 and 1.53 Mg ha-1, respectively. Hence, in the dry years, there was a great potential for yield increase if some irrigation measures were taken to afford soybean enough water during critical growing stage, such as R4, R5 and R6; while, in the normal years, the potential yield was relatively lower. Additionally, sustainable soybean production on Griffith, Sumter and Demopolis soil types may rely on specific management practices that consider improvements in soil-plant-water relations and their influence on final grain yield.
Technical Abstract: Because soybean is commonly grown under rainfed conditions in Mississippi, and farmers recognize benefits of installing irrigation, knowledge of rainfed soybean productivity and yield difference of different soil types is needed for deciding where irrigation may be most effective. This research employed APEX model to simulate rainfed soybean grain yield in each of 13 years (2002 to 2014) and with nine soil types. Objectives were to: (1) calibrate and validate APEX model using soil properties provided by NRCS SSURGO database, and experimental data obtained from three soil types and three irrigation levels; (2) estimate the rainfed soybean productivity and grain yield difference within and among soil types for years with different seasonal rainfall; and (3) recommend irrigation and soil management measures that may reduce the yield difference. Simulated grain yield ranged broadly from 2.24 to 6.14 Mg ha-1 across the nine soil types among the 13 years, yield difference reached to 3.90 Mg ha-1. Wet, normal and dry years had average grain yields of 4.88, 4.51 and 3.74 Mg ha-1 , respectively. For the average grain yield of nine soil types in each year, the range was from 3.55 to 5.42 Mg ha-1 over 13 years, a difference of 1.87 Mg ha-1. Meanwhile, the average yield of 13 years also existed great difference among nine soil types, ranging from 3.71 to 4.97 Mg ha-1, the yield gap reached to 1.26 Mg ha-1. Hence, in the dry years, there is a great potential for yield increase if some irrigation is provided during critical water stress period (R4, R5 and R6 growing stage); while, in the normal years, the potential yield difference is accordingly less. Meanwhile, Griffith, Sumter and Demopolis soils should be taken measures to increase their productivity and promote yield stability.