Elsevier

Water Research

Volume 154, 1 May 2019, Pages 387-401
Water Research

Associating the spatial properties of a watershed with downstream Chl-a concentration using spatial analysis and generalized additive models

https://doi.org/10.1016/j.watres.2019.02.010Get rights and content

Highlights

  • The spatial factors of the watershed explained the inter-station Chl-a variation.

  • The Chl-a variation was well explained at a local scale with a 25 km radius.

  • Agricultural and urban land uses were the two most influencing man-made factors.

  • Proximity to instream weirs downstream explained high Chl-a concentrations.

  • Relative importance of spatial factors varied depending on scales.

Abstract

We examined the relationship between downstream algal growth potential and the spatial environmental factors of both upland areas and stream buffer zones using spatial analysis and generalized additive models (GAMs). The models employed site-representative concentrations of chlorophyll a (Chl-a) from a total of 688 national water quality monitoring stations and the spatial factors of the corresponding 688 watersheds. The spatial environmental factors included topography, climate, land use class, soil type, and proximity of the monitoring station to the weir downstream and wastewater treatment plants (WWTPs). The explanatory power (adjusted R2 or Radj2) of the models was used to compare different spatial influential scales defined by stream buffers and upstream circular buffers. The spatial environmental factors of the entire watershed area better explained the inter-station variation in Chl-a than did those of the stream buffer and/or upstream circular buffer areas. However, the spatial environmental factors of watershed areas more than 25 km upstream circular buffer zones had only minor influence on the explainability of the models with regards to the inter-station variation in Chl-a levels. Generally, land use patterns were more strongly related to the inter-station Chl-a variation than were point sources of pollutants such as WWTPs. The two most influencing land uses on the inter-station Chl-a variation were urban and agricultural land uses, with varying relative contributions depending on the spatial influential scale: In general relative contribution of urban land use was larger at a larger spatial influential scale while that of agricultural land use showed an opposite trend. In addition, the proximity to the weir downstream explained high Chl-a concentrations in the stream water. Relative importance and causal effects of the spatial environmental variables to instream Chl-a were established based on this national scale correlative analysis, leading to decision-making with the goal of controlling instream algal growth.

Introduction

Algal blooms have been of increasing concern worldwide due to climate change and the increase in anthropogenic nutrients in water bodies (O'Neil et al., 2012; Paerl and Otten, 2013). In many countries, including USA, France, Italy, the UK, and China, social and environmental problems caused by algal blooms in coastal and inland waters have also been reported (Hellio et al., 2004; Viaroli et al., 2005; Scanlan et al., 2007; Zubia et al., 2009; Liu et al., 2010; Perrot et al., 2014; Harke et al., 2016). In South Korea, sixteen instream weirs were constructed along the four major national rivers in 2012, as part of the National River Restoration Project aiming at flood control and water utilization. As a consequence, there have been an increasing number of episodic algal bloom events in the four major national rivers after weir construction, due to the increased hydraulic residence time of the rivers with the prolonged dry period during summer and autumn seasons (Srivastava et al., 2015; KME, 2015).

While algal growth in water bodies is a natural phenomenon, the high biomass resulting from excessive growth of algae is of major concern (Backer, 2002). Excessive algal growth is usually accompanied by a reduction in the diversity of algae due to the dominance of a few harmful algal species and cyanobacteria (Heisler et al., 2008; Anderson et al., 2012). In particular, the excess growth of harmful algae has numerous impacts on the environment: the mass mortality of wild and farmed fish and shellfish; the death and illness of humans caused by toxic seafood, unclean drinking water, and irrigation water contamination; the death and illness of marine mammals, seabirds, and other animals; and changes in trophic structure within water bodies (Anderson et al., 2002). Algal blooms also increase the turbidity of the water and restrict light penetration into the waterbody, negatively affecting the underwater habitats of benthic flora and fauna (Paolo and Laura, 2014), while also degrading aesthetics and leading to taste and odor problems in drinking water.

Algal growth in water bodies is dependent upon internal factors such as the chemical compositions of the water, nutrient cycles and availability, abundance of herbivore community, hydrologic disturbances, water temperature and light availability (Munn et al., 2010; Elliott, 2010; Riseng et al., 2004; Xie et al., 2012). Large scale regional factors such as climate conditions, external nutrient loadings and characteristics of the watershed can also affect algal production and biomass (Biggs, 1995; Leland et al., 2000; Urrea-Clos et al., 2014; Woelmer et al., 2016). A growing number of studies have sought to link watershed characteristics and algal biomass (e.g., Chlorophyll a) in downstream water bodies because watershed characteristics such as land use patterns can be more easily manipulated by managers than internal factors or other external influences such as climate variables (Woelmer et al., 2016).

The identification of the environmental drivers of algal growth is generally considered an important task in the management of water quality, and process-based or empirical models based around statistical methods can be used to do this task. Process-based models enable precise cause-and-effect relationships to be identified, but they typically require significant computational resources and are inherently limited in terms of describing complex natural phenomena. This is because, in a process-based model, the parameterization process that links the model to field data can be time consuming and expensive, and often the direct measurement of these parameters is impossible, decreasing the reliability of the model's results (Demeritt and Wainwrite, 2009). In contrast, empirical methods provide a way to search for patterns in large data sets and have often been preferred as an initial approach to the understanding of the causal effects involved in algal blooms. Empirical models allow hypotheses that explain the patterns found in the data to be generated and their implications to be discussed (Demeritt and Wainwrite, 2009; Shiffrin, 2016).

Empirical statistical methods have been widely used to build relationships between watershed characteristics such as land use patterns and algal biomass in downstream water bodies. Many regional studies reported positive correlations of algal biomass with agricultural land use (Brett et al., 2005; Migliaccio et al., 2007; Munn et al., 2010) and urban land use (Liu et al., 2011; Ding et al., 2015; Knoll et al., 2015). Some studies failed to detect statistical correlation of agricultural land use with downstream algal biomass due to interference from external factors, such as wind over shallow lakes in upland areas (Ma et al., 2016) or stronger influence of urban areas (Siliva and Williams, 2001; Ding et al., 2015). Negative correlations between forest land use and algal biomass is a common finding in past research because areas covered by forest or vegetation can delay or attenuate nutrient transport into receiving waters, decreasing algal growth potential downstream (Thomas et al., 2004; Mori et al., 2015).

The influences of watershed characteristics on the downstream water quality can be dependent on spatial scale. Land use patterns at local scales (500–1000 m) such as stream buffer areas or radial buffer regions often better explain downstream nutrient concentrations or algal biomass than those at the watershed scale (Johnson et al., 1997; Busse et al., 2006; Klose et al., 2012). The local effects of land use patterns on nutrients and algal biomass in the downstream water bodies can be enhanced during dry seasons with low delivery efficiency of nutrients due to low flows, or when the point sources dominate flows (Klose et al., 2012). Urban land use can be correlated with nutrients and algal biomass more strongly at local scales (500 m) while agricultural land use explain these water quality variables better at the watershed scale due to greater transport efficiency of agricultural nitrate in the watershed (Busse et al., 2006).

In this study, we assessed a large number of natural and anthropogenic spatial environmental factors that could potentially affect algal biomass in streams based on Chlorophyll a (Chl-a) concentration data measured at nationwide water quality monitoring stations in South Korea using spatial analysis and statistical regression. A generalized additive model (GAM) was employed for the statistical regression to assess the non-linear response of Chl-a to the spatial environmental factors. The spatial environmental factors were considered to be explanatory variables for differences in Chl-a concentration across different geographical locations. These factors include climate conditions, geographical features, land use, soil type, and environmental infrastructure facilities. In addition, the effect of the physical distance of upland pollutant sources from the water quality monitoring station on the instream algal growth potential was examined. The spatial environmental factors of different buffer regions (defined by the buffer radius and the stream buffer distance) within a watershed were compared by examining the increase of explained variance of Chl-a with buffer radius and stream buffer distance.

Section snippets

Instream water quality data

In South Korea, the water quality of streams across the nation is monitored with a nationwide water quality monitoring system overseen by the Korea Ministry of Environment (KME). This system currently consists of 844 stations that monitor a maximum of 38 water quality parameters at 5- to 8-day intervals, including water temperature, pH, nutrient species, suspended solids, Chl-a, heavy metals, and toxic substances. It should be noted that this nation-wide monitoring system is not aimed at

Principal components associated with slope, land uses and soil types

From the PCA for SLOPE and the six land use classes at the watershed scale, five principal components (PCs) were extracted, explaining 97.72% of the total variation in the original data (Table S4). PC_L1 consists of several dominant variables; SLOPE and Forest were strongly negatively loaded and Agricultural was strongly positively loaded; Urban, Grassland, Bareland and Wetland were the dominant land uses in PC_L2, PC_L3, PC_L4, and PC_L5, respectively. The signs of the component loadings for

Explanatory power of the GAM for Chl-a

The greater explanatory power of the GAM using the Q3 concentration of Chl-a compared with Med (Table S8) indicates that the spatial factors of the influential region can better explain higher algal growth events than they can the seasonal average algal population in a stream. The explanatory power of the GAM for the inter-station Chl-a variation was dependent on the spatial influential scales. The explanatory power of the model gradually increased as the upstream buffer radius increased,

Conclusions

We examined the correlations between spatial environmental properties in influential upland regions within a watershed and downstream Chl-a concentration. Site-representative values for 17 spatial environmental variables in these regions were extracted from spatial analysis and used as explanatory variables for downstream Chl-a concentration. Generalized additive models (GAMs) were used to identify the significant explanatory variables using data from 688 water quality monitoring stations. From

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was supported by a grant (2015R1D1A1A01056753) from the National Research Foundation (NRF) of Korea.

References (91)

  • B. Lehner et al.

    Development and validation of a global database of lakes, reservoirs and wetlands

    J. Hydrol.

    (2004)
  • D.Y. Liu et al.

    Recurrence of the world's largest green-tide in 2009 in Yellow Sea, China: porphyra yezoensis aquaculture rafts confirmed as nursery for macroalgal blooms

    Mar. Pollut. Bull.

    (2010)
  • C. Ma et al.

    Establishment of physico-chemical variables and Chl a criteria based on land-use patterns and terrestrial ecosystem health

    Ecol. Eng.

    (2016)
  • G. Marra et al.

    Practical variable selection for generalized additive models

    Comput. Stat. Data Anal.

    (2011)
  • J. Nossent et al.

    Sobol's sensitivity analysis of a complex environmental model

    Environ. Model. Softw

    (2011)
  • J.M. O'Neil et al.

    The rise of harmful cyanobacteria blooms: the potential roles of eutrophication and climate change

    Harmful Algae

    (2012)
  • T. Perrot et al.

    Modelling green macroalgal blooms on the coasts of Brittany, France to enhance water quality management

    J. Mar. Syst.

    (2014)
  • P.J. Rice et al.

    Evaluation of individual and combined management practices to reduce the off-site transport of pesticides from golf course turf

    Sci. Total Environ.

    (2017)
  • C.M. Scanlan et al.

    The monitoring of opportunistic macroalgal blooms for the water framework directive

    Mar. Pollut. Bull.

    (2007)
  • I.M. Sobol

    Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates

    Math. Comput. Simlat.

    (2001)
  • C.R. Steinmann et al.

    A combined system of lagoon and constructed wetland for an effective wastewater treatment

    Water Res.

    (2003)
  • G. Urrea-Clos et al.

    Factors explaining the patterns of benthic chlorophyll-a distribution in a large agricultural Iberian watershed (Guadiana river)

    Ecol. Indicat.

    (2014)
  • K. Welde et al.

    Effect of land use land cover dynamics on hydrological response of watershed: case study of Tekeze Dam watershed, northern Ethiopia

    International Soil and Water Conservation Research

    (2017)
  • M. Zubia et al.

    Antioxidant and antitumoural activities of some Phaeophyta from Brittany coasts

    Food Chem.

    (2009)
  • L.S.K. Alfonso et al.

    Wet weather discharge characteristics of phosphorus and management implications in a mixed land-use watershed

    Desalination and Water Treatment

    (2015)
  • D.M. Anderson et al.

    Harmful algal blooms and eutrophication: nutrient sources, composition, and consequences

    Estuaries

    (2002)
  • D.M. Anderson et al.

    Progress in understanding harmful algal blooms: paradigm shifts and new technologies for research, monitoring, and management

    Annual Review of Marine Science

    (2012)
  • L.C. Backer

    Cyanobacterial harmful algal blooms (CyanoHABs): developing a public health response

    Lake Reservoir Manag.

    (2002)
  • B.J.F. Biggs

    The contribution of flood disturbance, catchme8nt geology and land use to the habitat template of periphyton in stream ecosystems

    Freshw. Biol.

    (1995)
  • G.L. Bowie et al.

    Rates, Constants, and Kinetics Formulations in Surface Water Quality Modelling

    (1985)
  • M.T. Brett et al.

    Non-point source impacts on stream nutrient concentration along a forest to urban gradient

    Environ. Manag.

    (2005)
  • L.B. Busse et al.

    Relationships among nutrients, algae, and land use in urbanized southern California streams

    Can. J. Fish. Aquat. Sci.

    (2006)
  • C.F. Cai et al.

    A preliminary study on the conditions and losses of nutrients of purple soils in Three-Gorge Reservoir area

    Geogr. Res.

    (1996)
  • Y. Cha et al.

    A Bayesian hierarchical approach to model seasonal algal variability along an upstream to downstream river gradient

    Water Resour. Res.

    (2016)
  • M. Chang et al.

    Estimates of low flows using watershed and climatic parameters

    Water Resour. Res.

    (1977)
  • L.S. Clesceri et al.

    Standard Methods for the Examination of Water and Wastewater

    (1999)
  • B. Debele et al.

    Coupling upland watershed and downstream waterbody hydrodynamic and water quality models (SWAT and CE_QUAL-W2) for better water resources management in complex river basins

    Environ. Model. Assess.

    (2008)
  • D. Demeritt et al.

    Models, modelling and geography

  • J. Ding et al.

    Impacts of land use on surface water quality in a subtropical river basin: a case study of the Dongjiang river basin

    Southeast Chia. Water

    (2015)
  • W.K. Dodds et al.

    Nitrogen and phosphorus relationships to benthic algal biomass in temperate streams

    Can. J. Fish. Aquat. Sci.

    (2002)
  • J. Elliott

    The seasonal sensitivity of cyanobacteria and other phytoplantion to changes in flushing rate and water temperature

    Glob. Chang. Biol.

    (2010)
  • T. Hastie et al.

    Generalized additive models

    Stat. Sci.

    (1986)
  • S. He et al.

    Partial regression method to fit a generalized additive model

    Environmetrics

    (2007)
  • C. Hellio et al.

    Seasonal variation of antifouling activities of marine algae from the Brittany coast (France)

    Mar. Biotechnol.

    (2004)
  • A.T.W.M. Hendriks et al.

    Rethinking wastewater treatment plant effluent standards: nutrient reduction or nutrient control?

    Environ. Sci. Technol.

    (2017)
  • Cited by (20)

    • Response of planktonic diversity and stability to environmental drivers in a shallow eutrophic lake

      2022, Ecological Indicators
      Citation Excerpt :

      Studies have shown that nutrients and pollutants usually come from human activities, and their levels in lakes decrease when they exceed a certain level (Zhang et al., 2022). In addition, the increase in temperature stimulates the growth and metabolic rate of phytoplankton under certain nutrient levels and increases productivity (Kim et al., 2019). Optimal light conditions are a prerequisite for the reproduction and growth of phytoplankton.

    • Mixed spatial scale effects of landscape structure on water quality in the Yellow River

      2022, Journal of Cleaner Production
      Citation Excerpt :

      These index are often used to describe the shape, fragmentation, aggregation, and heterogeneity of landscape (Liu et al., 2021; McGariga and Marks, 1995). The determination of the width of the riparian zone and the radius of the circular buffer zone comprehensively refers to previous studies (Liberoff et al., 2019; Kim et al., 2019). The mixed scales of 113 water quality were divided through following steps:

    View all citing articles on Scopus
    View full text