*3.2. Impacts of Planting Dates on Water Conservation and Irrigated and Dryland Cotton Yield*

Simulated irrigation amount increased by 7.3% with late planting of cotton during wet years relative to the baseline planting date (Table 4). However, early planting resulted in a reduction in irrigation amount by 16.4% during the wet years. This reduction was mainly caused by sufficient early season rainfall during the wet years in the Texas Panhandle. The percentage changes in ETc were within ±4% under different hydroclimatic years and planting dates (Table 4). A large variation was found in soil water content according to the hydroclimatic years and planting dates. The absolute changes in surface runoff and water yield were relatively small in the dry and normal years irrespective of the planting dates (Table 4). However, the surface runoff increased by 18.7% and 46.6% in the case of the early and late planting dates in the wet years, respectively. Simulated cotton yields were decreased by 14.4%, 25.4%, and 26.4% for the delayed planting date in the dry, normal, and wet years, respectively (Table 4). However, 1.3%, 7.6%, and 5.1% increase in irrigated cotton yield was found with early planting date in the dry, normal, and wet years, respectively (Table 4). Planting and harvesting dates impacted cotton growth and yield [36,37]. Early planting dates could extend the growing season and help producers avoid inclement weather in the late season [38]. Mauget et al. [10] also found that early planting could increase cotton yield by maximizing growing season degree days and total cool hours in the Texas Panhandle. Cotton requires accumulations of larger amounts of heat units to maturity compared to corn and sorghum (*Sorghum bicolor* L.) [39]. Therefore, the early planting of cotton might be feasible in improving yield and water conservation.

In the study basin, cotton is usually planted in the middle of May according to local field studies. Planting of cotton a half month ahead caused a clear increase in irrigation compared to the baseline planting date in May (54.5% and 24.3%) and July (10.7% and 15.6%) in the dry and normal years, and the late planting led to an apparent increase in irrigation in September of 151.8%, 77.6%, and 81.5% during the dry, normal, and wet years, respectively (Figure 4a,d and Figures S3a,d and S4a,b). A similar trend to irrigation was detected for monthly cotton ETc from May to September using the alternative planting dates under three hydroclimatic years (Figure 4b,e and Figures S3b,e and S4c,d). There were clearly high soil water contents in September (146.4%, 37.6%, and 39.7%) with the delayed planting of cotton for dry, normal, and wet years, which was associated with the increased irrigation amounts with the late planting date (Figure 4c,f and Figures S3c,f and S4e,f).

**Figure 3.** Comparison of average monthly irrigation (**a**,**d**), crop evapotranspiration (ETc) (**b**,**e**), and soil water content (**c**,**f**) during dry and wet years using different irrigation application depths in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin.

The assessment of results from the dryland cotton HRUs provided a better comparison because rainfall was the sole source of water input. Thus, the hydroclimatic years were the dominant factors for water balances. Results indicated a negligible change in ETc with the alternate planting dates under three hydroclimatic years (Table 5). An increase in soil water content was found in the case of the early planting of dryland cotton under different hydroclimatic regimes (Table 5). However, a decrease in soil water content was identified for the late planting date. Generally, the late planting date resulted in reductions in surface runoff and water yield, particularly in the wet years. An evident increase in surface runoff (42.4%) and water yield (28.5%) were also found for the early planting date in the wet years (Table 5). The delayed planting of dryland cotton led to an evident reduction in cotton yield of 9.8%, 21.1%, and 20.5% during the dry, normal, and wet years, respectively (Table 5). Nevertheless, dryland cotton yields increased by 0.7%, 9.4%, and 5.4% during dry, normal, and wet years, respectively, for the early planting date. Therefore, to increase both irrigated and dryland cotton yields, early sowing may be warranted in the Texas Panhandle.

Under the dryland cotton land use, changes in ETc were noticed in June (6.0% increase) and July (6.4% decrease) with the early planting date in the dry years relative to the baseline planting date (Figure 5a and Figure S5a). The opposite trends were found for the late planting date in the dry years. In the normal years, considerable variation in ETc was found in June (9.8% increase) and October (25.5% decrease) with early planting of dryland cotton (Figure S6a,b). Relatively small absolute changes were noticed for the delayed planting in the normal years. Notable changes in ETc were found in July (4.3% increase) and August (4.5% decrease) for the early planting in the wet years (Figure 5c and Figure S3c). By contrast, a 7.4% decrease and an 8.2% increase in ETc were detected in July and August, respectively, for the late planting date in the wet years. The marked increases in soil water content were only found in June (14.1%) and July (14.9% and 19.6%) during the dry, normal, and wet years, respectively, for the late planting date. However, distinct increases in soil water content were found from October to December and from October to May with the early planting of dryland cotton during the dry years and the normal and wet years, respectively (Figure 5b,d and Figures S5b,d and S6c,d).

**Table 5.** Comparison of the average annual water balance parameters and cotton yield under three hydroclimatic regimes using various planting dates and maturity cultivars in the dryland cotton HRUs in the Double Mountain Fork Brazos basin.


**#** The number in the parentheses is the percent change using an alternative scenario relative to the respective baseline scenario.

**Figure 4.** Comparison of average monthly irrigation (**a**,**d**), crop evapotranspiration (ETc) (**b**,**e**), and soil water content (**c**,**f**) during dry and wet years using different planting dates in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin.

**Figure 5.** Comparison of average monthly crop evapotranspiration (ETc) (**a**,**c**) and soil water content (**b**,**d**) during dry and wet years using different planting dates in the dryland cotton HRUs in the Double Mountain Fork Brazos basin.

#### *3.3. Effects of Different Cotton Maturity Cultivars under Both Irrigation and Dryland Conditions on Water Balances and Cotton Production*

The absolute differences in the studied hydrologic parameters were small when using different maturity cotton cultivars under both irrigation and dryland management regardless of hydroclimatic years (Tables 4 and 5). However, the short-season cultivar produced 11.6%, 15.7%, and 19.2% higher irrigated cotton yield during the dry, normal, and wet years, respectively, compared to their respective baseline scenarios. Those increases in dryland cotton yields were 4.0%, 10.1%, and 10.0% in the dry, normal, and wet years. Similar percentage reductions were found with the long-season cultivar in various hydroclimatic regimes under the irrigation and dryland conditions (Tables 4 and 5). Like alternate planting dates, different maturity cotton cultivars highlighted the importance of heat unit accumulation for crop development. The concept of heat units emerged from observations that plants do not grow below a threshold temperature. This temperature for cotton is 15.6 ◦C. Cotton growth and development are directly related to accumulated heat units when there are no other environmental limiting factors [40]. Recently, Masasi et al. [41] also reported that under adequate irrigation supply, cotton yield responds positively and strongly to the increase of heat units using the AquaCrop model in the U.S. Southern Great Plains. As for the monthly analysis, small recognizable changes in irrigation and ETc were found in June and July in the dry years among diverse maturity cultivars (Figure 6a,b and Figure S7a,b,d,e). A clear decrease in irrigation and ETc was found in July with the long-season cultivar in the normal and wet years (Figure 6d,e and Figures S7a,b,d,e and S8a,b,c,d). In general, no considerable changes were noticed in soil water content with the changes in maturity cotton cultivars in the case of irrigated cotton under three hydroclimatic regimes (Figure 6c,f and

Figures S7c,f and S8e,f). There was almost no influence of maturity cultivars on ETc and soil water content under dryland cotton farming (Figure 7 and Figures S9 and S10). Although the current climate in the Texas Panhandle is suitable for cotton production, short-season cultivars are more promising for a yield increase. It is worth noting that the short-season cultivar is crucial for dryland management as it can mature early and reduce water stress duration relative to the full- and long-season cultivars in this semi-arid environment. Therefore, the selection of appropriate maturity cultivars is necessary in view of the challenging environment for cotton production in the Texas Panhandle.

**Figure 6.** Comparison of average monthly irrigation (**a**,**d**), crop evapotranspiration (ETc) (**b**,**e**), and soil water content (**c**,**f**) during dry and wet years using different maturity cultivars in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin.

**Figure 7.** Comparison of average monthly crop evapotranspiration (ETc) (**a**,**c**) and soil water content (**b**,**d**) during dry and wet years using different maturity cultivars in the dryland cotton HRUs in the Double Mountain Fork Brazos basin.

#### **4. Conclusions**

An assessment of some potential cultivation practices on water conservation and cotton production was performed in the DMFB basin in the semi-arid Texas Panhandle region using the SWAT-MAD model. Modeling results indicated that using a relatively small irrigation application depth for cotton resulted in increased seasonal irrigation and ETc under various hydroclimatic years. By contrast, the large irrigation application depth for cotton showed water-saving and yield-boosting effects. The early planting date demonstrated the potential for water conservation and yield increase for cotton under both irrigation and dryland conditions, which allowed for the accumulation of relatively high heat unit totals for crop physiological maturity. It is intuitive that the early planting date could favor a yield increase more in the normal and wet years compared to the dry years. Similar to early planting, the short-season cotton cultivar highlighted great potential for yield improvement under this climatic condition. In summary, larger irrigation application depths for cotton could primarily support groundwater conservation. Early planting of irrigated and dryland cotton might be considered for enhancing cotton yields and reducing water consumption in the Texas Panhandle, especially in wetter years. Additionally, using a short-season cultivar could be an option for further improving cotton production capacity and narrowing the yield gap in the Texas Panhandle. In this study, we did not completely consider the spatial variations in agricultural inputs/practices due to the limited information available. In addition, the spatial inconsistency with the actual field boundaries based on the HRU definition could result in some uncertainties. Therefore, the modeling results have a certain level of uncertainty when representing the real world. For these

reasons, producers should be cautious when interpreting our findings for decision making in their specific fields.

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/agriculture12010017/s1, Table S1: Default and calibrated values of hydrologic and cotton growth parameters using the SWAT-MAD model in the Double Mountain Fork Brazos basin. Figure S1: Change percentages of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under dry and wet years using different irrigation application depths relative to the baseline irrigation depth in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S2: Comparison of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under normal years using different irrigation application depths in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S3: Change percentages of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under dry and wet years using various planting dates relative to the baseline planting date in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S4: Comparison of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under normal years using various planting dates in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S5: Change percentages of average monthly crop evapotranspiration (ETc) and soil water content under dry and wet years using various planting dates relative to the baseline periods in the dryland cotton HRUs in the Double Mountain Fork Brazos basin. Figure S6: Comparison of average monthly crop evapotranspiration (ETc) and soil water content under normal years using various planting dates in the dryland cotton HRUs in the Double Mountain Fork Brazos basin. Figure S7: Change percentages of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under dry and wet years using diverse maturity cultivars relative to the baseline cultivar in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S8: Comparison of average monthly irrigation, crop evapotranspiration (ETc), and soil water content under normal years using diverse maturity cultivars in the irrigated cotton HRUs in the Double Mountain Fork Brazos basin. Figure S9: Change percentages of average monthly crop evapotranspiration (ETc) and soil water content under dry and wet years using diverse maturity cultivars relative to the baseline cultivar in the dryland cotton HRUs in the Double Mountain Fork Brazos basin. Figure S10: Comparison of average monthly crop evapotranspiration (ETc) and soil water content under normal years using diverse maturity cultivars in the dryland cotton HRUs in the Double Mountain Fork Brazos basin.

**Author Contributions:** Conceptualization, L.T., Y.Z., and Y.C.; methodology, L.T., Y.Z., and Y.C.; software, L.T., Y.Z., Y.C., and G.W.M.; data curation, L.T., Y.Z., Y.C., and S.A.; writing—original draft preparation, L.T.; visualization, L.T. and Y.Z.; investigation, L.T., Y.Z., G.W.M., S.A., and D.K.B.; supervision, Y.C.; writing—reviewing and editing, Y.Z., Y.C., G.W.M., S.A., and D.K.B. All authors have read and agreed to the published version of the paper.

**Funding:** This research was funded by the Chinese Universities Scientific Fund under award number 1191-15051002. The APC was funded by 1191-15051002. The research was also supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture under award numbers NIFA-2021-67019-33684 and NIFA-2012-67009-19595.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data sharing is not applicable.

**Acknowledgments:** This research was supported partially by the Ogallala Aquifer Program, a consortium between the USDA-Agricultural Research Service, Kansas State University, Texas A&M AgriLife Research, Texas A&M AgriLife Extension Service, Texas Tech University, and West Texas A&M University. We gratefully thank the anonymous reviewers for their valuable comments and suggestions for improving this paper.

**Conflicts of Interest:** The authors declare no conflict of interest.
