The DSSAT is a software application program consisting of various crop models that predict growth, development, and yield as defined by the soil-plant-atmosphere dynamics and several tools that manage database on crop, soil, experiment, and weather and include applications for graphical display, seasonal analysis, rotational analysis, and genotype coefficient estimation [13]. The DSSAT suite of crop models has been used for various applications such as precision crop management, agroecosystem sustainability, climate change impacts, and greenhouse gas emission studies [13]. Model simulations are carried out mainly on a daily basis using weather, soil, and crop management data. By integrating crop, soil, weather, and management options with crop models and application programs, the DSSAT suite simulates multi-year outcomes of crop management strategies.

Sequence Analysis, an important DSSAT tool, allows for rapid simulation, inspection, and analysis of results of long-term cropping sequences [14]. With this tool, a series of statistics can be calculated, and various graphics showing relationships between trends and variability can be created. Since simulation studies are conducted across multiple cropping seasons, this tool allows for the carryover of soil water and nutrients from the preceding crop to the following crop [15] [16].

The Texas A&M AgriLife Research & Extension Center at Amarillo (35.19˚N, 102.06˚W) is situated in the Llano Estacado region of the southern United States (https://familypedia.fandom.com/wiki/Llano_Estacado). This region, where agriculture is a major economic activity, comprises parts of eastern New Mexico and northwestern Texas. At the Amarillo Center, numerous experiments have been conducted to investigate, discover, develop, evaluate, and apply technology to sustain livestock and crop production in the Texas Panhandle region and beyond. In this study, Amarillo, Texas was used as a representative site for the Llano Estacado region [17].

To explore the interannual climate variability effects on the feasibility of growing cowpea-wheat double crops as well as on the yields of cowpea and wheat in the Llano Estacado region, a long-term weather dataset spanning 80 years (1942-2021) was used. The historical daily data on precipitation, temperature, and windspeed in the Amarillo region were obtained from the website of National Centers for Environmental Information (https://www.ncei.noaa.gov/access/search/data-search/daily-summaries); whereas those on solar radiation were generated using a reliable irradiation model described by [18].

Pullman clay loam (Torrertic paleustolls) was used as a representative soil for the study because it is one of the major soils used for agricultural purposes in the Llano Estacado region [19]. The soil data (Table 1) were obtained from the Gridded Soil Survey Geographic (gSSURGO) database of the USDA NRCS [20] [21]. The drainage coefficient of the soil was 0.60, whereas the run-off curve number was 81.

The DSSAT Sequence Analysis tool was used to simulate grain yields for cowpea and winter wheat in three cropping systems: cowpea-wheat, fallow-wheat, and

^aMH = master horizon, WP = wilting point, FC = field capacity, SA = saturation, WH = water holding capacity, HC = saturated hydraulic conductivity, TN = total N, OC = organic carbon.

fallow-cowpea. The wheat crop that followed cowpea or fallow, hereafter, will be referred to as ^cwheat and ^fwheat, respectively, and the cowpea crop that followed fallow as ^fcowpea. Simulations were made for a total of 94 scenarios comprising three N application rates to wheat, namely 0, 50, and 100 kg N ha⁻¹; four planting dates for cowpea, namely June 1, June 15, July 1, and July 15; and six planting dates for wheat, namely September 15, September 30, October 15, October 30, November 15, and November 30 (Table 2).

For simulations, the following management and environmental inputs were assumed. For soil, we used Pullman clay loam. For cultivars, we used “Newton” for winter wheat and “Cal #5 MG4” for cowpea. Because the genetic coefficients for this wheat cultivar were already estimated by [22] for the Texas Panhandle region, there was no need to further calibrate and evaluate the wheat model for this cultivar. For the Cal #5 MG4 cultivar, the default genetic coefficients for this cowpea cultivar are those in the standard DSSAT release [9] and correspond to the cultivar coefficients upon which the cowpea model was adapted (K. J. Boote, personal communication, 11 February 2022). For each scenario, simulation started on April 1, two months before the earliest cowpea planting date of June 1 in 1942, and terminated on the harvest date associated with the latest planting date of wheat in 2022. For simulations, the plant populations assumed were 30 plants m⁻² (about 40 kg∙ha⁻¹) for cowpea and 323 plants m⁻² (about 100 kg∙ha⁻¹) for wheat. Using the conventional tillage, dry seeds were planted on rows at 3 cm depth. Inorganic N fertilizer was applied only to wheat, not cowpea.

Of the total amount of N set for application to wheat, 50% was applied at the time of planting and the remainder on February 15 of the following year. For organic amendments, the stover residues of each crop were assumed to be automatically incorporated into the soil on the harvest day of the crop so that the

^†Double crop.

nutrients in the residues would transfer from the preceding crop to the following crop in cycles. For soil organic matter estimation, Century was used as the method, with the five years’ field history of “Cultivated, good management, initial default SOM” [23].

For ^fwheat, regression analyses were carried out to explore relationships between grain yields and planting dates as influenced by N application rate. For ^fcowpea, however, since there was no N rate involved, the grain yields were regressed only against the planting dates in general.

In the case of ^cwheat double-cropping, however, our objective was to assess the potential success and yields of cowpea-wheat double cropping in the study region as affected by the planting dates of both crops and N application rate to wheat. For this assessment, we ran the DSSAT crop model with sequential analysis for 80 years, starting from the earliest cowpea planting date in 1942 through the latest wheat planting date in 2021, for each of the 72 double-cropping scenarios presented in Table 2.

For a given scenario—denoted as C i W j . k ; where C_i is cowpea with the i-th planting date of Jun 1, Jun 15, Jul 1, or Jul 15; and W j . k is wheat with the j-th planting date of Sep 15, Sep 30, Oct 15, Oct 30, Nov 15, or Nov 30 and the k-th N application rate of 0, 50, or 100 kg N ha⁻¹—we calculated the feasibility of a successful cowpea-wheat double-crop in the study region and the associated average yields of both cowpea and wheat crops.

The feasibility of a successful double-crop for the C i W j . k scenario was obtained by counting the number of successful cowpea-wheat double-cropped seasons, that is, the number of events with wheat following cowpea successively and then dividing that number by the total 80 seasons.

The average yield of each crop involved in double-cropping for the C i W j . k scenario was obtained by summing up yields in the successful double-crop years and then dividing the sum by 80. For unsuccessful (failed) double-crop years, the yields of each crop were assumed to be zero.

Each scenario, a cropping system, comprised two different crops: cowpea and wheat. To compare total crop yields (benefits) across systems, one would need to combine the yields of both crops. Since the yields of two different crops could not be added together in terms of absolute yields, we combined them in terms of the system benefit index (SBI), defined as follows. A greater value of SBI would be an indication of a more benefitted system in terms of cowpea-wheat double-cropping.

S c − w B I i , j , k = C i / C s + W j , k / W s (1)

where S c − w B I i , j , k is the system benefit index for the double-crop cowpea-wheat scenario associated with the i-th planting date of cowpea, the j-th planting date of wheat, and the k-th N rate to wheat; C_i is cowpea yield with the i-th planting date; C_s is the standard yield of cowpea in the study region; W j , k is wheat yield with the j-th planting date and the k-th N rate; and W_s is the standard yield of wheat in the study region. The crop yields are in kg∙ha⁻¹. The C_s and W_s, in turn, were estimated as follows.

C s = ∑ i = 1 n C f i n (2)

where C f i is the yield of cowpea following fallow for the i-th planting date, where n = 4 and i = 1 (Jun 1), …, n (Jul 15).

W s = ∑ j = 1 m W f j m (3)

where W f j is the yield of wheat following fallow for the zero N rate and the j-th planting date, where m = 6 andj = 1 (Sep 15), …, m (Nov 30).

Similarly, the SBIs for fallow-cowpea and fallow-wheat cropping systems were computed as:

S f − c B I i = C i / C s (4)

S f − w B I j , k = W j , k / W s (5)

where S f − c B I i and S f − w B I j , k are the system benefit indices for the fallow-cowpea and fallow-wheat scenarios, respectively. The other variables are already defined above.

Once the 72 values of S c − w B I i , j , k , 18 values of S f − w B I j , k , and 4 values of S f − c B I i were computed, the S c − w B I i , j , k value of a system in question was then compared with the corresponding values of S f − c B I i and S f − w B I j , k (associated with the 0 N rate) to find out if the double-crop system being compared would be more beneficial than the corresponding fallow-cowpea or fallow-wheat systems. The double-crop system would be more beneficial than either of the fallow-cowpea or fallow-wheat system if:

S c − w B I i , j , k > S f − c B I i     and     S f − w B I j , k (6)

The possibility of planting a wheat crop might depend on the length of the growing season of the preceding crop cowpea. The cowpea seasonal length, in turn, could be affected by drought, if any, during the season [24] [25]. Thus, to explore the effect of drought on cowpea growing season, an agricultural drought index—Agricultural Reference Index for Drought (ARID)—was used as it is computationally simple and physically and physiologically sound [26] ; is applicable to a wide range of crops, soils, topographies, and management with fairly small uncertainties [27] ; has potential to predict the yield loss from drought for several field crops that are more sensitive to water stress [28] ; and, as a drought indicator, may be used in drought forecasting [29]. Using the weather data and soil parameters, the daily values of ARID during the cowpea growing seasons were computed, and these values were used to associate seasonal length with drought.

As the SBI values in Table 6 showed, except for certain systems with zero N rate (10 in total), the fallow-wheat monocropping systems were more beneficial than the cowpea-wheat double-cropping systems. Particularly, in all systems with N rate greater than zero, the fallow-wheat systems were the most beneficial.

The response of fallow-wheat grain yields to wheat planting date as affected by N rate is presented in Figure 7. Regardless of planting date, the effect of N rate on wheat yields was significant. With an increase in N rate from 0 through 100, wheat yields increased almost linearly. This indicated that the suggested N rate of 34 kg∙ha⁻¹ [38] [39], especially for the years with above average seasonal

precipitation, might not be adequate to meet the yield potential of fallow-wheat systems in the study region under rainfed conditions.

For each N rate studied, the relationship between wheat grain yield and planting date was quadratic (Figure 8). When no fertilizer was applied, wheat yields increased, although at a diminishing rate, until the planting date of November 30 (day of year, DOA = 333). With an increase in N rate, however, wheat yields reached the peak and decreased thereafter with an earlier planting date. That is, the maximum yields of wheat at the N rates of 50 and 100 kg∙ha⁻¹ were associated with the planting dates of November 15 (DOY = 318) and October 15 (DOY = 288), respectively. This could be likely due to the weather conditions of cooler temperatures, optimal rainfall, and reduced drought.

Irrespective of planting dates, the grain yields of fallow-wheat at the N rates of 0, 50, and 100 kg N ha⁻¹ were about 1200, 3200, and 5200 kg∙ha⁻¹. Reference [38] found that, at the N rate of 34 kg∙ha⁻¹ in Bushland, Texas (the study region), the 77-year (1940-2016) average value of observed wheat grain yields associated with high yielding cultivars under dryland conditions was 2300 kg∙ha⁻¹. In our study, the 80-year (1942-2021) average value of simulated wheat yields associated with the N rate of 50 kg∙ha⁻¹ was about 3200 kg∙ha⁻¹. Because the yield of 2300 kg∙ha⁻¹ at 34 kg N was approximately equal to the yield of 3200 kg∙ha⁻¹ at 50 kg N, results indicated that the simulated yield values were reliable.

The SBI values in Table 6 showed that within the fallow-cowpea monocropping systems considered the monocrop of cowpea planted on July 1 was the most beneficial of all systems, with the SBI of 1.06. However, no fallow-cowpea monocropping system was the best among the entire 94 systems studied or among the 34 systems associated with zero N. Nevertheless, certain fallow-cowpea monocropping systems were more beneficial than certain fallow-wheat or cowpea-wheat systems. The monocrop of cowpea planted on June 1 was better than the systems associated with the earlier planting dates of wheat (September 15 and 30). With a delay in cowpea planting, the cowpea monocropping system was better than an additional system associated with an additional planting date of wheat. Thus, when the monocrop of cowpea was planted on July 15, the number of better monocropping systems was the greatest.

The detailed response of fallow-cowpea grain yields to cowpea planting date is presented in Figure 9. The relationship between cowpea grain yield and planting date was quadratic (Figure 10). Of all the planting dates considered, the planting

date of July 1 (DOY = 181) was associated with the highest yields of cowpea. The cowpea crops planted before and after this date had lower yields. The reduction in yields with planting dates after July 1 was likely because there was not enough time for cowpea to flower and set seeds.