Influence of Long-Term Fertilization on Carbon, Nitrogen, and Phosphorus Allocation and Homeostasis in Cotton Under the Regulation of Phosphorus Availability

Ecological stoichiometry offers critical insights into nutrient dynamics and soil–plant interactions in agroecosystems. To explore the effects of long-term fertilization on soil–cotton C, N, P stoichiometry and stoichiometric homeostasis in arid gray desert soils, this study was conducted at a national gray desert soil monitoring station in Xinjiang (87°28′27″ E, 43°56′32″ N, elevation: 595 m a.s.l.)—an arid and semi-arid region with an annual mean temperature of 5–8 °C and annual precipitation of 100–200 mm. Established in 1989, the 31-year experiment adopted a wheat–maize–cotton annual rotation system with six treatments: CK (control, no fertilizer), N (nitrogen fertilizer alone), NK (nitrogen + potassium fertilizer), NP (nitrogen + phosphorus fertilizer), PK (phosphorus + potassium fertilizer), and NPK (nitrogen + phosphorus + potassium fertilizer). Key results showed that balanced NPK fertilization significantly increased soil organic carbon (SOC) by 22.7% and soil total phosphorus (STP) by 48.6% compared to CK, while the N-only treatment elevated soil N:P to 3.2 (a 68.4% increase vs. CK), indicating severe phosphorus limitation. For cotton, NPK increased seed phosphorus content by 68.2% (vs. N treatment) but reduced straw carbon content by 10.2% (vs. PK treatment), reflecting a carbon allocation trade-off from vegetative to reproductive organs under nutrient sufficiency. Stoichiometric homeostasis differed between organs: seeds maintained stricter carbon regulation (1/H = −0.40) than straw (1/H = −0.64), while straw exhibited more plastic N:P ratios (1/H = 1.95), highlighting organ-specific adaptive strategies to nutrient supply. Redundancy analysis confirmed that soil available phosphorus (AP) was the primary driver of cotton P uptake and yield formation. The seed cotton yield of NPK (5796.9 kg ha−1) was 111.7% higher than CK, with NP (N-P co-application) achieving a 94.7% yield increase vs. CK—only 7.9% lower than NPK, whereas single N application showed the lowest straw yield (5995.0 kg ha−1) and limited yield improvement. These findings demonstrate that long-term balanced NPK fertilization optimizes soil C-N-P stoichiometric balance by enhancing SOC sequestration and phosphorus retention, regulating cotton organ-specific stoichiometric homeostasis, and promoting efficient nutrient uptake and assimilate translocation. The study confirms that phosphorus is the key limiting factor in arid gray desert soil cotton systems, and balanced NPK supply is essential to mitigate stoichiometric imbalances and sustain soil fertility and productivity. This provides targeted practical guidance for rational fertilization management in arid agroecosystems, emphasizing the need to prioritize phosphorus supply and avoid single-nutrient application to maximize resource use efficiency.