摘要

Owing to the severe shortages of fresh water in arid and semi-arid areas, growers are forced to apply deficit irrigation with fresh or saline water. To quantify the impacts of water deficit and salt stress individually and the interaction on tomato yield and quality, three pot experiments were conducted from spring 2016 to autumn 2017. The EPIC growth model was also used to simulate fruit growth process. Three irrigation treatments used were full irrigation, 2/3 and 1/2 of the full irrigation. Salt stress varied with the season: 0, 3 parts per thousand, 6 parts per thousand, 9 parts per thousand for 2016 season, 0, 3 parts per thousand for 2016-2017 season and 0, 2 parts per thousand, 3 parts per thousand, 4 parts per thousand for 2017 season. Decreases in single fruit weights were associated with increasing soil salt content. However, fruit quality parameters including CI (color index), TSS (total soluble solids) and TSSC (total soluble sugar content) improved significantly with increasing salt content of soil. In the absence of salt stress, the application of 2/3 of full irrigation showed a 18.9% and 1.0% increase of yield per plant, respectively in 2016 and 2017 seasons with comparison to full irrigation, mainly owing to the increase of fruits number per plant. Fruit quality also improved with increasing Fn (fruit firmness), CI, TSS and TSSC by 7.9%, 43.8%, 9.8% and 3.8% in 2016 season, and by 4.7%, 0.7%, 20.9% and 34.2% in 2017 season, respectively. Fruit quality parameters were more affected by salt stress than drought, the interactive impact of water and salt on fruit quality parameters was not significant. At mild water stress (2/3 of full irrigation) with moderate salt stress (salt content of 3 parts per thousand), although yield showed a decline of 20.3%-32.0%, fruit quality parameters of Fn, CI, TSS and TSSC increased by 4.9%-43.4% through 2016 and 2017 seasons. Two of van Genuchten and Hoffman models (i.e. non-linear and exponential reduction model) were used to evaluate the relations of relative yield and soil salt content with acceptable accuracy, R-2 of 0.989 and 0.971, respectively. Soil salt content at which yield decreased by 50% is 4.0 parts per thousand, and 4.7 parts per thousand, respectively in the non-linear reduction model and the exponential reduction model. The EPIC growth model simulated fruit growth process with acceptable accuracy for no stress, water stress and salt stress, with R-2 of 0.863, 0.839 and 0.895, respectively. The two yield reduction models and the relationship between fruit quality parameters and soil salt content showed that there are tradeoffs between tomato yield and fruit quality in saline soils.