Purification function of wetlands : spatial modelling and pattern analysis of nutrient reduction in the Liaohe Delta

X. Li

Research output: Thesisinternal PhD, WU


The eutrophication of coastal seawater has been a serious problem for the last two decades in Eastern China. The purification function of natural wetlands at big river deltas provides a potential solution to cut down nutrient input into the sea. The purpose of this research is to give a quantified evaluation as to what extent the natural wetland can be used as a treatment system for nutrient enriched river water. By integrating processed-based mathematical models with GIS, a number of valuable results and conclusions have been obtained through this study.

A spatial simulation model has been established based on the field and literature data, to simulate the nutrient reduction and its distribution in the wetland of Liaohe Delta, China. A non-linear regression model is used for the nutrient reduction in the canal system, while Mander-&-Mauring's linear regression model is adopted for the reed fields. According to the simulation result, there is a "mutual compensation" for the nutrient reduction in the reed system and canal system, so that the total reduction rate remains relatively stable in spite of the input concentration change at the pumping station. It is 66% for total nitrogen and 90% for soluble reactive phosphorus. In combination with the canals, the present 80,000 ha of reed can remove about 3,200-4,000 tons of nitrogen and 80 tons of soluble reactive phosphorous during the irrigation period each year. But this is only 1/10 of its total reduction capacity, with water being the limitation factor.

Four spatial combinations of reed, canals and pumping stations are designed to investigate the effect of pattern on nutrient reduction: 1) canal density, 2) reed area size, 3) reed shrinking pattern and 4) pumping station position. The simulation results indicate that each factor brings less than 10% deviation in total nutrient reduction rate, though the absolute reduction quantity can be different. If the reed area is stable, it is better to remain a low canal density, and keep the pumping station near the border of the reed area. Generally speaking, smaller reed area close to the pumping station is more efficient in nutrient reduction than larger, scattered ones. The shrinkage pattern of land transformation for the reed is most recommended in keeping a high reduction rate for the nutrients. The present reed area can accept at least 4 times more water in spring.

The relationship between landscape structure and nutrient reduction is measured with the help of some landscape indices. Not all the landscape pattern indices are closely related to the nutrient reduction of the wetland system. Therefore the ability of pattern indices to characterize the effect of pattern change on function is rather limited. Redundancies also exist among similar indices. Landscape indices should be chosen according to the purpose of the study, based on the criteria of simplicity, generality and ecological meaningfulness.

The research work is a combination of landscape ecology, wetland ecology and GIS technology. The spatial model developed is also applicable for other areas with similar situations. The results will contribute to a sustainable landscape planning in the study area. It is concluded that the natural wetlands have a great potential to be used for reducing nutrient input into the sea.

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Bregt, Arnold, Promotor
  • Jongman, R.H.G., Promotor, External person
Award date5 Jan 2000
Place of PublicationS.l.
Print ISBNs9789058081650
Publication statusPublished - 2000


  • water treatment
  • wetlands
  • landscape
  • nutrients
  • waste water treatment
  • hydrobiology
  • physical models
  • geographical information systems
  • china
  • nature


Dive into the research topics of 'Purification function of wetlands : spatial modelling and pattern analysis of nutrient reduction in the Liaohe Delta'. Together they form a unique fingerprint.

Cite this