Application of trait-based plant-soil feedback for agroecosystem optimization

TY - THES

T1 - Application of trait-based plant-soil feedback for agroecosystem optimization

AU - Bin, Zhaoqi

N1 - Data archiving: no data, PhD thesis

PY - 2026/1/23

Y1 - 2026/1/23

N2 - The Green Revolution increased global food production, but intensive agriculture now relies heavily on fertilizers and pesticides, resulting in soil degradation, pollution, resource depletion, and biodiversity loss. Sustainable agriculture therefore requires not only maintaining crop yields but also supporting multiple soil functions, such as biomass production, nutrient cycling, carbon storage, and disease suppressiveness, simultaneously, a concept known as soil multifunctionality. These functions are shaped by interactions between plants and soil biotic and abiotic properties, commonly described as plant–soil feedbacks (PSFs). Although PSF theory is increasingly applied to agriculture, its predictability remains limited, and its relationship with soil multifunctionality is poorly understood. In this thesis, I investigated whether plant traits can predict PSFs of crop species and how crops influence individual soil functions and overall soil multifunctionality. While trait-based PSF approaches have been widely studied in natural ecosystems, their application to agricultural crops remains rare. By combining a quantitative synthesis (Chapter 2), a greenhouse PSF experiment (Chapter 3), and a soil multifunctionality assessment (Chapter 4), I evaluated the potential of trait-based PSF approaches to inform crop rotation design that balances productivity and soil health. In Chapter 2, I synthesized published studies reporting both PSF and plant traits, supplemented with trait values from the TRY database. Across a wide range of species, trait–PSF relationships were generally weak, with root length showed the strongest association with PSF. Relationships were stronger within functional groups than across all species, indicating strong context dependence. Traits measured in situ explained more variation than database-derived traits, suggesting that local environmental conditions strongly influence trait expression relevant to PSF. Combining data sources did not improve predictions, highlighting the limited and context-dependent nature of trait-based PSF inference. In Chapter 3, I experimentally tested PSFs of twelve major crops under controlled greenhouse conditions and assessed whether root traits could predict crop PSFs. Results showed clear species-specific PSF: broad bean generally enhanced subsequent crop growth, whereas wheat had inhibitory effects. Root traits such as projected root area, root tip number, and root tissue density were the best PSF predictors, but their predictive power depended on whether traits were measured in sterilized or conspecific-conditioned soils. Moreover, crops differed in both their ability to condition soils and their responsiveness to soils conditioned by others, limiting the practical value of trait-based PSF predictions for crop rotation design. In Chapter 4, I extended the analysis beyond PSF to assess how the same twelve crops affected four soil functions and overall soil multifunctionality. No single crop consistently enhanced all functions. Instead, crops contributed in complementary ways: legumes promoted biomass production and nutrient cycling, while crucifers enhanced disease suppressiveness and carbon storage. Soil multifunctionality increased with functional evenness, indicating that crop diversity can enhance multiple soil functions simultaneously. Overall, this thesis shows that while PSFs are strong and species specific, trait-based PSF predictions are limited and highly context dependent. Sustainable crop rotation design therefore requires moving beyond yield-centered strategies toward functional complementarity among crops to support long-term soil multifunctionality and agricultural resilience.

AB - The Green Revolution increased global food production, but intensive agriculture now relies heavily on fertilizers and pesticides, resulting in soil degradation, pollution, resource depletion, and biodiversity loss. Sustainable agriculture therefore requires not only maintaining crop yields but also supporting multiple soil functions, such as biomass production, nutrient cycling, carbon storage, and disease suppressiveness, simultaneously, a concept known as soil multifunctionality. These functions are shaped by interactions between plants and soil biotic and abiotic properties, commonly described as plant–soil feedbacks (PSFs). Although PSF theory is increasingly applied to agriculture, its predictability remains limited, and its relationship with soil multifunctionality is poorly understood. In this thesis, I investigated whether plant traits can predict PSFs of crop species and how crops influence individual soil functions and overall soil multifunctionality. While trait-based PSF approaches have been widely studied in natural ecosystems, their application to agricultural crops remains rare. By combining a quantitative synthesis (Chapter 2), a greenhouse PSF experiment (Chapter 3), and a soil multifunctionality assessment (Chapter 4), I evaluated the potential of trait-based PSF approaches to inform crop rotation design that balances productivity and soil health. In Chapter 2, I synthesized published studies reporting both PSF and plant traits, supplemented with trait values from the TRY database. Across a wide range of species, trait–PSF relationships were generally weak, with root length showed the strongest association with PSF. Relationships were stronger within functional groups than across all species, indicating strong context dependence. Traits measured in situ explained more variation than database-derived traits, suggesting that local environmental conditions strongly influence trait expression relevant to PSF. Combining data sources did not improve predictions, highlighting the limited and context-dependent nature of trait-based PSF inference. In Chapter 3, I experimentally tested PSFs of twelve major crops under controlled greenhouse conditions and assessed whether root traits could predict crop PSFs. Results showed clear species-specific PSF: broad bean generally enhanced subsequent crop growth, whereas wheat had inhibitory effects. Root traits such as projected root area, root tip number, and root tissue density were the best PSF predictors, but their predictive power depended on whether traits were measured in sterilized or conspecific-conditioned soils. Moreover, crops differed in both their ability to condition soils and their responsiveness to soils conditioned by others, limiting the practical value of trait-based PSF predictions for crop rotation design. In Chapter 4, I extended the analysis beyond PSF to assess how the same twelve crops affected four soil functions and overall soil multifunctionality. No single crop consistently enhanced all functions. Instead, crops contributed in complementary ways: legumes promoted biomass production and nutrient cycling, while crucifers enhanced disease suppressiveness and carbon storage. Soil multifunctionality increased with functional evenness, indicating that crop diversity can enhance multiple soil functions simultaneously. Overall, this thesis shows that while PSFs are strong and species specific, trait-based PSF predictions are limited and highly context dependent. Sustainable crop rotation design therefore requires moving beyond yield-centered strategies toward functional complementarity among crops to support long-term soil multifunctionality and agricultural resilience.

U2 - 10.18174/681532

DO - 10.18174/681532

M3 - PhD thesis

T3 - NIOO Thesis 233

PB - Wageningen University

CY - Wageningen

ER -