SEARCH

SEARCH BY CITATION

Keywords:

  • nutrient pollution;
  • hydrology;
  • climate change impact;
  • water quality

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES

Climate change, together with the increasing influence of anthropogenic activities in the atmosphere, has impacted ecosystems enormously, escalating much concern for pollution control and management. Proper assessment of the effect of temperature rise due to global warming and the subsequent changes in hydrology on catchment biogeochemical processes is essential for accurate prediction of pollution level. Not many studies have been carried out on the aspect of climate change-driven nutrient pollution assessment, and inability of many existing modelling tools persists. With these research needs in focus, this study has made an assessment of climate change impact on nutrient pollution by applying a distributed hydrological modelling tool with newly developed nutrient modules that incorporate climate-based description of catchment and in-stream processes. The model has been applied to the Latrobe River basin in Australia and calibrated and validated prior to simulation of climate change scenarios. The scenarios have been developed on the basis of the Intergovernmental Panel on Climate Change projections of higher-emission scenario A1F1 for this region. The model has predicted the impact of temperature rise on the nutrient-transformation processes and the subsequent releases into waterways for projected high-flow and low-flow situations. The results show that climate change is likely to contribute to increase in nutrient pollution in the waterways. The results have been used to assess future stream water quality condition in the study area. Copyright © 2012 John Wiley & Sons, Ltd.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES

The impacts of human activities on the natural environment have reached extreme levels with far-reaching consequences, many of which are just being identified (Flower, 2009). Of these consequences, human-induced climate change has the potential to have negative impact on the ecosystem from local to global levels (IPCC, 2008).

The climate system is likely to be changing faster than it was thought before, indicating more serious risks (Steffen, 2009). There is an urgent need for a development of adaptation strategies that enhance resilience and develop capacity to cope with new climate conditions. For ecosystem management and pollution control, the development of such strategies will depend on the ability of societies to select and adopt the best management practices with consideration of climate change effects. The global nutrient cycle has been altered substantially with the fact that anthropogenically derived nitrogen (N) in the 2000s is ten times higher than that in the 1860s (Meybeck, 1982; Bennett et al., 2001; Galloway et al., 2004; Boyer et al., 2006; Lu et al., 2009). The change in land use, over-exploitation of natural resources, and heavy use of chemical fertilizers and combustion of fossil fuels for supply of food and energy have been responsible for excessive pollution in waterways. Proper assessment and quantification of these impacts on nutrient pollution can therefore be useful for developing effective planning and management strategies.

Nutrients are major indicators of water quality. Dissolved inorganic nitrogen (DIN) and dissolved inorganic phosphorus (DIP or PO4-P) are better nutrient indicators than total nitrogen and total phosphorus, and the ratio of DIN to DIP is a good predictor of algal blooms (Harris, 2001) or the production and growth of phytoplankton in inland and marine waters (Turner et al., 2003; Dagg et al., 2004; Imteaz et al., 2009). Hence, it is necessary to understand the forms of nutrients, riverine fluxes and the nutrient-transformation processes for determining the complex relationships between human activities, nutrient export and their related effects (Lu et al., 2009). Imteaz et al. (2003) studied the effect of lake inflow parameters on lake water quality. However, not many models are currently available that can effectively capture the catchment characteristics and link these to anthropogenic and climate effects in modelling nutrient exports in the riverine system.

The climate-driven N load to waterways was investigated by Whitehead et al. (2006) by applying a catchment nutrient dynamics model called INCA-N in a lowland chalk stream of the Kennet River in the UK, a tributary of the Thames River. Atmospheric deposition and its control has been one of the major focuses in this modelling. However, in Australia, atmospheric deposition is not considered as a source of nutrient (Harris, 2001). Although the export of nutrient into waterways was relatively low in Australia in the past, the situation has worsen in the recent decades due to land use change (excessive land clearance) and poor land and stream management, and it has become a major concern at national level (Tiller and Newall, 1995). The temperature is expected to rise largely in this continent. The average temperature in Australia is projected to rise by 1 to 2⋅5 °C by 2070 under low-emission scenario and 2⋅2 to 5 °C under high-emission scenario. The total rainfall is likely to decrease over much of the area. It is likely that the future climate pattern will be a long antecedent dry period followed by short, intense rainfall events (CSIRO and BoM, 2007). An investigation of water quality under these projected changes in climatic conditions, which is likely to increase the pollution level in the river system, is of interest to different stakeholders including river basin managers.

Modelling the catchment processes in soil–water–climate conditions and the movement of pollutant in waterways with hydrological cycle is a very complex phenomenon that requires realistic details of pollution characteristics and the mechanism that determines the release of nutrient from catchment to the streams, which are not adequately considered in many existing modelling tools. Inability to predict nutrients in high spatial and temporal resolution is also an issue. Although techniques in some modelling tools such as E2 (Perraud et al., 2005; Argent et al., 2007) or WaterCAST (Cook et al., 2009) developed in Australia have been improved to describe storm event-based nutrient simulation on a daily basis, these are simple models, where nutrient release is a function of event mean concentration, a quite simplistic approach not suitable for quantifying input changes due to hydroclimatic variability and realistic prediction of climate change impact. The nutrient emission modelling based on soil transformation processes in soil–water–climate conditions is relatively rare in Australia. Laboratory-based analysis has been carried out to determine transformation rate for different land use. Paul et al. (2002) used an empirical model to predict N mineralization process in forest soil, which incorporated soil water and temperature modifier in estimating N mineralization through the incubation test of soil samples in the laboratory. Hossain et al. (2011) have developed and applied a catchment water quality model to simulate nutrients build-up and wash-off from a catchment. Similar tools are available for agricultural areas. However, transformation-based nutrient emission and its fate during river transport has not been studied.

This study has presented an application of a newly developed nutrient dynamics model for the prediction of climate change impact on nutrient pollution. Considering the limitation in the storm event-based approach, the study focused on transformation-based simulation for nutrient-generation process, accounting for soil-bound nutrient release and dynamic transport for river network system. The model was developed within the existing framework of a distributed hydrological modelling tool called Institute of Industrial Science Distributed Hydrological Model (IISDHM; Dutta et al., 2000). The model was tested and applied for calibration and verification in two catchment studies (Alam et al., 2009; Alam and Dutta, 2010). The model was applied to assess the impacts of climate change on waterways nutrient dynamics in the Latrobe River, Australia. Several scenarios were developed on the basis of the projection of the Intergovernmental Panel on Climate Change (IPCC) and other studies carried out for the region. The effects of soil–water–climate conditions on nutrient-transformation processes and the subsequent release into river water were estimated.

MODELLING APPROACH

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES

The IISDHM was used to describe the hydrology in the catchment and river network system. The nutrient release was estimated on a sub-catchment basis and used as boundary input for simulation of nutrient in the river network system.

Distributed hydrological model

The IISDHM, originally developed at the Institute of Industrial Sciences of the University of Tokyo, Japan, has been widely used in many hydrological analysis in different countries (Dutta et al., 2000; Jha et al., 2000; Dutta et al., 2006; Asokan and Dutta, 2008; Dutta and Nakayama, 2009). It describes the different components of the land phase of the hydrological cycle, namely (1) interception and evapotranspiration, (2) unsaturated zone, (3) ground water flow, (4) surface flow and (5) river flow. The model computes surface run-off and river flow by solving Saint-Venant equations for continuity and momentum, taking into account of interception and evapotranspiration, and infiltration loss. With the use of kinematic wave approximation of the Saint-Venant equations in a finite difference scheme, the catchment run-off is calculated. The model deals with river flow separately where surface run-off enters into the river grids as lateral flow during simulation. The river component can also be run stand alone by extracting boundary conditions for flow from the surface run-off simulation. Both the surface component and the river component have been built using an explicit solution scheme, which is conditionally stable by satisfying Courant stability condition. The Courant condition is achieved when the time step is less than the time taken for a wave to travel the distance between two grids (Chow et al., 1988).

Catchment model of nutrients

The model considers external input of chemical fertilizer, plant uptake, and mineralization and immobilization of N for different land uses in determining nutrient generation. The reaction processes are temperature and soil moisture dependent. The equations describing these processes (Table 1) are adopted from the INCA-N model (Whitehead et al., 1998a). However, the original approach used in the INCA-N model has been modified to suit the modelling approach in this study. Major modifications include the introduction of an export function to determine the nutrient release based on flow capacity from the catchment, and the transfer of the release rates to the dynamic river transport module.

Table 1. Equations for nitrogen transformation process in soil layer (obtained from Whitehead et al., 1998a).
ProcessEquations
Plant uptake
  • display math(1)
where CN up = nitrogen uptake rate of plant (day−1), ui = plant growth index and XN = amount of ammonium and nitrate N
Mineralization–immobilization
  • display math(2)
where CN mina = nitrogen mineralization rate (g/day−1), SMI = soil moisture index, CN imm = nitrogen immobilization rate (day−1) and Xamm = amount of mineralized N
Nitrification
  • display math(3)
where Cnitri = nitrification rate (day−1) and inline image = amount of ammonium N
Denitrification
  • display math(4)
where Cdeni = denitrification rate (day−1) and inline image = amount of nitrate N
Temperature correction
  • display math(5)
where Cn = rate of reaction (day−1), Cr = rate of reaction (day−1) at 20 °C and θs = soil temperature (°C)

The soil moisture index (SMI) (Si) used in the aforementioned equations (Table 1) was determined on the basis of the function of soil moisture deficit (SMD), which becomes 0 when the deficit is maximum (i.e., SMD = SMDmax) and 1 when the soil is in saturation. The Si and the SMD can be calculated using the following equations presented by Whitehead et al. (1998a) and Finkele et al. (2006):

  • display math(6)
  • display math(7)

where Peff (rain–interception–run-off) = effective rainfall and ET = evapotranspiration.

The calculations of catchment releases are performed by multiplying the export function with the net generation as shown in Equations (8)-(11).

  • display math(8)
  • display math(9)
  • display math(10)
  • display math(11)

where suffix t denotes the computational time level; NH4-N and NO3-N are ammonium and nitrate, respectively. inline image and inline imagedenote external inputs of NNH4 and NNO3, respectively.

The export function Uw is denoted by the following equation:

  • display math(12)

where a and b are export parameters and Q is discharge (m3/s).

Sediment yield and soil-bound nutrient modelling

Whereas the model determines inorganic nutrient release from the catchment with run-off, the soil-bound nutrients are calculated in relation to the soil erosion process. For this purpose, the widely known modified version of the Universal Soil Loss Equation (Williams, 1975; Williams and Berndt, 1977) was used to calculate sediment yield from the catchment. The soil-bound or organic nutrients for N are calculated on the basis of the sediment yield by using the following equations (Leon et al., 2001):

  • display math(13)
  • display math(14)

where NSED = nitrogen transported by sediment (kg), NSCN = soil nitrogen concentration, YSED = sediment yield, ER = nutrient enrichment function, m and y are enrichment coefficients and Tf = correction factor for soil texture (e.g. 0⋅85 for sand, 1⋅0 for silt, 1⋅15 for clay and 1⋅5 for peat). However, the enrichment coefficient term ER was set to 1, assuming that soil-bound nutrient is proportional to sediment yield.

In-stream processes and river transport modelling

The one-dimensional equation for advection and dispersion processes (Equation (15)) with chemical reaction (Chapra, 1997) was solved for in-stream modelling and transport in the river network system.

  • display math(15)

where V = element volume, c = nutrient concentration, Ac = element cross-section area, E = longitudinal dispersion coefficient, x = distance unit, t = time, U = average velocity, r = reaction rate, p = internal sources/sinks and s = external or lateral sources/sinks.

Equation (15) was solved using an explicit solution scheme in a finite difference method. The numerical inaccuracy due to explicit condition was overcome by employing a steady-state time-varying solution and by satisfying the stability criteria for Courant number and diffusion number (Chapra, 1997). A detailed description is provided in the study by Alam et al. (2009).

The reaction term (rc + p) in the mass balance (Equation (15)) is a general representation of each constituent's reaction and its interaction with other parameters, which are detailed in Table 2. In this case study application, nitrite has been combined with nitrate. The production of phytoplankton or algae, and their interaction with nutrients (source/sink), was ignored due to their insignificance in the study area.

Table 2. Equations for in-stream reaction process (after Chapra, 1997).
ProcessEquations
  1. A = Algae as Chll a (mg/l), NORG = nitrogen as organic matter, NNH4 = nitrogen as ammonium, NNO2 = nitrogen as nitrite, NNO3 = nitrogen as nitrate, μ, ρ, α1, β3, β1 and β2 = rates of reaction; σ3, σ1 and σ4 = rates of settling and re-suspension, H = water depth, F = attenuation factor.

Algae (A)
  • display math(16)
Organic nitrogen (ORG-N or NORG)
  • display math(17)
Ammonia nitrogen (NH4-N or NNH4)
  • display math(18)
Nitrite nitrogen (NO2-N or NNO2)
  • display math(19)
Nitrite nitrogen (NO3-N or NNO3)
  • display math(20)

Study area: Latrobe River basin

The Latrobe River basin is located in the Gippsland region of Victoria, Australia, which is a temperate climate region. The basin drains a catchment area of 4500 km2 in the central Gippsland area (Figure 1). The mean annual rainfall in the region is about 780 mm. The river bears significant socio-economic and environmental values for the region, where several large industries are located including four major electricity power generation industries and a large paper mill. The river and its tributaries are major sources of water supply for more than 80 000 urban consumers and 800 private irrigators, nationally significant wetlands, as well as aquatic flora and fauna, and provide about 25% of the river inflow to the Gippsland lakes (LVWSB, 1986). The river basin includes natural grazing pasture, mining area and cropping land.

image

Figure 1. Study area in the Latrobe River basin.

Download figure to PowerPoint

Figure 2 shows the nutrient levels in the river system from 2002 to 2008, which indicates a very high phenomenon at the end of June 2007 when a flash flood of high magnitude occurred, leading to an algal outbreak that lasted for several months in the downstream lakes system (DSE, 2009). This kind of phenomenon highlights the need for risk assessment in extreme events. This study has attempted to quantify the nutrient levels during this peak flood and predict the impacts due to climate change effects based on the scenarios developed for the region.

image

Figure 2. Nutrient level at Kilmany South, downstream of the Latrobe River (source: Victorian Water Resources Data Warehouse online portal).

Download figure to PowerPoint

The modelling application was limited to the upper sub-catchment of the Latrobe River basin. The land-use types in this area are mainly production forest and grazing pasture. The area is relatively hilly. The head water is just 60 km east of Melbourne. The river is incised through the mountain and stepped down from topographic elevation of 274 to 59 m before joining to the Narracan Lake. Willow Grove is the downstream gauging location for monitoring of hydrologic and water quality data; the site has been considered for model calibration, verification and impact assessment. Through the Narracan lake systems, the river water is highly regulated towards the downstream in the basin, and flow is restricted at a number of dam locations, which complicated modelling of the whole systems.

Climate change scenarios

The climate in Gippsland has observed distinctly measureable changes over the last century. The average temperature has increased by 0⋅8 °C, and the annual average rainfall is likely to decrease (Brooke and Hennessy, 2005). Assessments of climate change in the Gippsland region can be found in a number of studies (Brooke and Hennessy, 2005; CSIRO and BoM, 2007; Jones and Webb, 2008). Whereas the annual average rainfall is likely to decrease, a number of extreme events are projected to increase. Table 3 shows the projected changes in temperature and rainfall based on the IPCC emission scenarios for 2030 and 2070.

Table 3. Projection of climate change (rise and fall) in the Gippsland region (Jones and Webb, 2008).
 For 2030 (AIB scenario)For 2070 (lower-emission scenario B1)For 2070 (higher-emission scenario A1F1)
Range (10–90%)Range (10–90%)Range (10–90%)
Annual rainfall−8 to 0%−12 to 0%−22 to 0%
Temperature+0⋅5 to +1⋅1 °C+0⋅9 to +1⋅9 °C+1⋅7 to +3.6 °C

Three scenarios were developed for the assessment of the effect of temperature rise and changes in run-off on nutrient flux. The first scenario was designed to assess the effect of temperature rise on the nutrient biochemical process. The model was run for temperature change for higher-emission scenario A1F1 of the IPCC, that is, a 3⋅6 °C temperature rise.

Scenario 1:

effect of 3⋅6 °C temperature rise on nutrient-transformation process.

In the second and third scenarios, changes in surface hydrology on the catchment nutrient release have been incorporated in addition to the effect of temperature rise. An assessment of climate change impact on hydrology was carried out by the Department of Sustainability and Environment and CSIRO Atmospheric Research Centre at Melbourne (Chiew and McMahon, 2002; Jones and Durack, 2005), which determined likely changes in the flow regime in the study area. According to the study on stream flow assessments in Victorian State catchments, run-off is likely to increase by 20% in East Gippsland, and the minimum change across most of the state will range between −5 and −10% for the IPCC 2070 AIF1 projection. In the worst case, stream flow may reduce up to 50% in all catchments. On the basis of these projections for stream flow changes, Scenarios 2 and 3 were developed as follows:

Scenario 2:

50% reduction of flow with the conditions in Scenario 1

Scenario 3:

20% increase of flow with the conditions in Scenario 1

Soil and water temperature

The temperature data for model analysis was obtained from the Bureau of Meteorology. The records for soil temperature was only available at Sale station located at the downstream part of the Latrobe River, which have been correlated with air temperature data available at Willow Grove. Soil temperature data were recorded at depths of 10, 20, 50 and 100 cm taken at 09:00 and 15:00 h daily. The average values at 10 cm depth were used. The mean daily value has been used for comparison with air temperature. The correlation between air and soil temperature is shown in Figure 3. According to Watson (1980), annual, summer and winter soil and air temperature are all strongly correlated, particularly in the south-eastern part of Australia. A similar trend was also found in the study area, with soil temperature usually higher than the air temperature. To estimate soil temperature for the projected rise of 3⋅6 °C in air temperature, a temperature gradient was measured from the correlation curve (Figure 3). On the basis of the temperature gradient, present soil temperature was converted to future projections (Figure 3). Site-specific variation may also occur because of factors such as presence of vegetative cover and watering regime (Watson, 1980), but it was not considered in this study.

image

Figure 3. Correlation of air and soil temperature (left) and temperature profile at current and future projections (right).

Download figure to PowerPoint

Analysis of water and air temperature was necessary to establish a correlation and estimate future projected water temperature. There were no actual measurements of stream temperature for this river system. However, temperature recorded with water samples at regular interval (mostly fortnightly) is available together with air temperature at several gauging locations, which were collected for this study, and data at Rosedale station were used for establishing a correlation. From the established correlation, both current and future water temperatures were computed. Figure 4 shows the correlation between air and water temperature, and current and projected temperature profiles.

image

Figure 4. Correlation of air and water temperature (left) and temperature profile at current and future projections (right).

Download figure to PowerPoint

Soil moisture index

The soil moisture condition was calculated on the basis of the rainfall and the run-off and termed as soil moisture deficit. This deficit was then indexed 0 to 1 for SMI by using the equation discussed previously. The SMI has been calculated for base condition and applied for all scenarios, which provide variation only for seasonal patterns on nutrient-transformation process.

Model set-up

The model consists of a digital elevation model (DEM) and a number of spatial data layers for hydrological simulation in distributed manner. The SRTM DEM (CGIAR-CSI, 2004) was used to generate 1-km resolution model grids. Flow direction and flow accumulation maps were used to delineate surface flow path and generation of the river network grids. The sinks of the DEM were eliminated to obtain a hydrologically corrected DEM by using the hydrological assessment tool in ARCGIS software (ESRI, 2000). The model needs hourly rainfall and daily evaporation data. The main calibration parameter was Manning's roughness coefficient for hydrologic simulation. With the use of run-off coefficients, the infiltration loss was accounted for in the run-off calculation.

The model requires input of external and internal sources of nutrient. External source is chemical fertilizer; internal sources are mineralization rate of nutrient from different land uses. The transformation rates or the reaction coefficients were assumed on the basis of the literature (Whitehead et al., 1998b).

Observed data for nutrient and suspended sediment

Nutrient data for the Latrobe River were obtained from the Victorian Water Warehouse database (www.vicwaterdata.net), an online data portal maintained by various Victorian government agencies. The water quality data at Willow Grove and at other locations are available on fortnightly basis.

Model performance

The performance of the model has been demonstrated in the study by Alam et al. (2009). The newly developed module has been checked for numerical accuracy, sensitivities of different parameters and model uncertainty through applications to two catchment studies including the Latrobe River. Here, only the results from the calibration and the verification are presented.

Hydrological events were simulated from June 2007 to January 2008 by dividing it into the calibration period of 1 June 2007 to 9 September 2007 and the verification period of 10 September 2007 to 10 January 2008. A constant base flow was assumed in the absence of sub-surface zone modelling. The simulated discharge agreed well with the observed data at Willow Grove as shown in Figure 5. The Nash Sutcliffe coefficient (NSC) value was 0⋅82, and the coefficient of determination (R2 value) was 0⋅85 for the calibration period. The model has over-predicted two peaks in the validation period perhaps because of the overly distribution of rainfall from one rainfall station data, which has produced negative NSC values, with an R2 value of 0⋅8. Further investigation is necessary.

image

Figure 5. Comparison of discharge (Q), SS, ORG-N and NO3-N level at Willow Grove station.

Download figure to PowerPoint

Although a number of records were very limited for observed suspended sediment (SS), organic nitrogen (ORG-N) and NO3-N, statistical measures indicate an acceptable performance of the model. The NSC value is 0⋅76, and R2 is 0⋅77 for NO3-N. Although the co-relation of SS and ORG-N with the observed data is not strongly satisfactory, the trends are within the allowable limit of the observed data.

Climate change simulation results

The climate change analysis was applied on the base condition at Willow Grove site for the entire simulation period and compared for impact assessment. Figure 6 shows a comparison of SS and ORG-N. About a 4% reduction is observed in the concentration level for SS and ORG-N at peak because of the reduction of flow for Scenario 2 although the overall increase occurs at about 4%. For Scenario 3, about a 3% increase occurs at peak, and the average is increased up to 12%; more importantly, this increase raised the concentration of low-flow period quite significantly.

image

Figure 6. Impact on SS (left) and ORG-N (right) due to change in run-off.

Download figure to PowerPoint

Figure 7 shows the moving average plots of NO3-N for the different conditions. It can be seen that the 3⋅6 °C temperature rise (Scenario 1) has increased the nutrient levels quite significantly with intensification of the nutrient-transformation processes. The difference between base line and Scenario 1 (solid line in Figure 7) is almost constant at around 36% for the entire simulation period. With the decrease of catchment run-off, the nutrient load decreased (Scenario 2), but the concentration level was still higher than 18% of the base condition. Scenario 3 results show the highest increase in nutrient level. About a 42% increase would occur in river nitrate level because of a 20% increase of run-off with 3⋅6 °C temperature rise. This assessment can be used further to identify the probable changes in water quality indicators, as demonstrated in the next paragraph.

image

Figure 7. Climate change effects on NO3-N concentration.

Download figure to PowerPoint

An analysis was carried out to determine actual water quality based on recorded data available from 1991 to 2008 with missing records in some years (www.vicwaterdata.net). Data were available at least once or twice a month during this period, giving more than 12 sample points for each month at Willow Grove. Figure 8 presents 90% exceedance probability curve and the maximum, median and minimum values of the observed nitrate levels in different months along with two curves representing increased median values by 18 and 42%, which correspond to projected increase in nitrate levels in Scenarios 2 and 3, respectively. The nitrate level at this site is well below the Australian Drinking Water Guidelines (ADWG, 2004), which is 50 mg/l. But the nitrate level during this period exceeded the default low-risk trigger value beyond which ecosystems become vulnerable. According to the Australian and New Zealand Guidelines for Fresh and Marine Water Quality (2000), the default low-risk trigger value is 0⋅015 mg/l for NOx (oxides of nitrogen) for upland river and 0⋅04 mg/l for lowland river in south-east Australia.

image

Figure 8. Statistical distribution of NO3-N level and the predicted changes due to climate change.

Download figure to PowerPoint

The key water quality indicator used very often in management decisions is percentile values at different level. Median values in Figure 8 show a central tendency of the pollution data and can be assumed to be representative of the overall nitrate level of the river system. For an 18% increase, the median values would be close to the 90th percentile and exceed those in June, July and October, whereas for a 42% increase, they are above the 90th percentile except for February, August and October. It is inferred from this finding that the median state nitrate level is likely to change to the 90th percentile state (approximately) because of climate change, which is very significant. The overall percentile level for the entire period is presented in Figure 9 for base and 18 and 42% increased levels. The percentile levels determined for climate condition from this graph can be used for decision making.

image

Figure 9. Percentile levels for NO3-N concentration.

Download figure to PowerPoint

Because the impact is mainly influenced by the effects of increase of temperature on transformation process and the change in nutrients release for run-off, whatever the input condition is in different land use, the downstream effect will be at the same rate provided that the sediment input rate will be the same, so it is predicted that the result will be applicable for the entire river basin.

CONCLUSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES

The paper has demonstrated the applicability of a newly developed nutrient dynamics and transport process model in the study of the effect of temperature rise and changes in flow regime on nutrient pollution in the Latrobe River, Australia. A process-based approach was followed for the development of the model, which has enabled climate-based simulation.

The projected changes of temperature and surface run-off have been used to assess the effect on nutrient pollution in the study area. It appears that temperature rise will have a significant impact on the catchment biogeochemical process, which will, in turn, intensify the nutrient level in the waterways. The model results have shown the following.

  • For a 3⋅6 °C rise in temperature, the nutrient process would be substantially intensified, which would lead to an increase of the nitrate level by 36% for the current hydrological condition.
  • Although the projected reduction in surface run-off would reduce the catchment nutrient release, it is likely that the nitrate level in the river system will be higher by 18% than the current level because of the effect of temperature rise.
  • The increase of surface run-off due to the projected short-term intensified rain events would lead to an increase of the nitrate level by 42%.

The model outputs were further used to determine the water quality indicator for the nitrate level in stream water, which could be useful in management decision making.

ACKNOWLEDGEMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES

The sources of the hydrologic, water quality and other spatial and temporal data set for the model are the online portal of the Victorian Water Resources Data Warehouse (http://www.vicwaterdata.net/vicwaterdata/home.aspx), the Bureau of Meteorology and other government agencies in Australia.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MODELLING APPROACH
  5. CONCLUSION
  6. ACKNOWLEDGEMENTS
  7. REFERENCES
  • Alam MJ, Dutta D, Nakayama K. 2009. Modelling nutrient dynamics and transport process in river basin: a case study – Saru River, Japan. In Proceedings of the 7th International Symposium on Ecohydraulics, Concepción, Chile, 12–16 January 2009, 110.
  • Alam MJ, Dutta D. 2010. Quantification of climate change impact on nutrient pollution: application of a dynamic model for Latrobe River basin in Australia. In Proceedings of the International Symposium on Coastal Zones and Climate Change: Assessing the Impacts and Developing Adaptation Strategies, 12–13 April 2010, Monash University, Australia.
  • Australian Drinking Water Guideline (ADWG). 2004. National water quality management strategy. Australian Drinking Water Guidelines 6, 2004, National Health and Medical Research Council and the Natural Resource Management Ministerial Council, 10–11 April 2003.
  • Australian and New Zealand Guidelines for Fresh and Marine Water Quality. 2000. Chapter 3, downloaded from http://www.mincos.gov.au, date accessed 28 December 2010.
  • Argent RM, Pexton HM, McAlister AB. 2007. The PortsE2 Decision Support System – An Application of E2. In MODSIM07 International Congress on Modelling and Simulation: Modelling and Simulation Society of Australia and New Zealand, Oxley L, Kulasiri D (eds). Christchurch: New Zealand 895901. ISBN: 978-0-9758400-4-7. downloaded from http://www.mssanz.org.au/MODSIM07/MODSIM07.htm
  • Asokan S, Dutta D. 2008. Analysis of water resources in the Mahanadi River basin, India under projected climate conditions. Hydrological Processes 22: 35893603.
  • Bennett EM, Carpenter SR, Caraco NF. 2001. Human impact on erodable phosphorus and eutrophication: a global perspective. Bioscience 51: 227234.
  • Brooke C, Hennessy K. 2005. Climate change impacts in Gippsland, CSIRO Marine and Atmospheric Research.
  • Boyer EW, Howarth RW, Galloway JN, Dentener FJ, Green PA, Vörösmarty CJ. 2006. Riverine nitrogen export from the continents to the coasts. Global Biogeochemical Cycles 20: GB1S91. DOI: 10.1029/2005GB002537
  • Chapra SC. 1997. Surface Water Quality Modelling. WCB McGraw Hill: USA.
  • Chow VT, Maidment DR, Mays LW. 1988. Applied Hydrology. McGraw-Hill, New York.
  • Chiew FHS, McMahon TA. 2002. Modelling the impacts of climate change on Australian stream flow. Hydrological Processes 16: 12351245.
  • CGIAR-CSI. 2004. Void-filled seamless SRTM data version 4. International Centre for Tropical Agriculture (CIAT). The CGIAR-CSI SRTM 90m database. http://srtm.csi.cgiar.org
  • Cook FJ, Jordan PW, Waters DK, Rahman JM. 2009. WaterCAST – whole of catchment hydrology model an overview. 18th World IMACS/MODSIM Congress, Cairns, Australia 13–17 July 2009. http://mssanz.org.au/modsim09, downloaded on 1/11/2010.
  • CSIRO, BoM. 2007. Climate change in Australia. Technical Report, CSIRO and Bureau of Meteorology through the Australian Climate Change science program. Downloaded from www.climatechangeinaustralia.gov.au as dated on 25 November 2010.
  • Dagg M, Benner R, Lohrenz S, Lawrence D. 2004. Transformation of dissolved and particulate materials on continental shelves influenced by large rivers: plume processes. Continental Shelf Research 24: 833858.
  • DSE. 2009. Gippsland region sustainable water strategy. Discussion Paper, Department of Sustainability and Environment (DSE), Victorian Government, Melbourne, downloaded from http://www.ourwater.vic.gov.au/programs/sws/gippsland/discussion-paper
  • Dutta D, Herath S, Musiake K. 2000. Flood inundation simulation in a river basin using a physically based distributed hydrologic model. Hydrological Processes 14: 497519.
  • Dutta D, Herath S, Musiake K. 2006. An application of a flood risk analysis system for impact analysis of a flood control plan in a river basin. Hydrological Processes 20(6): 13651384.
  • Dutta D, Nakayama K. 2009. Effects of spatial grid resolution on river flow and surface inundation simulation by physically based distributed modeling approach. Hydrological Processes 23: 534545.
  • ESRI. 2000. Hydrologic and Hydraulic Modelling Support with Geographic Information Systems, Maidment D, Djokic D (eds). ESRI Press: California.
  • Finkele K, Mills GA, Beard G, Jones DA. 2006. National gridded drought factors and comparison of two soil moisture deficit formulations used in prediction of forest fire danger index in Australia. Australian Meteorological Magazine 55: 183197.
  • Flower DJM. 2009. An integrated approach to modelling urban water systems. PhD thesis, Monash University, Australia.
  • Galloway JN, Dentener FJ, Capone DG, Boyer EW, Howarth RW, Seitzinger SP, Asner GP, Cleveland CC, Green PA, Holland EA, Karl DM, Michaels AF, Porter JH, Townsend AR, Vörösmarty CJ. 2004. Nitrogen cycles: past, present, and future. Biogeochemistry 70(2): 153226.
  • Harris G. 2001. A nutrient dynamics model for Australian waterways: land use, catchment biogeochemistry and water quality in Australian rivers, lakes and estuaries, Australia State of the Environment, Second Technical Paper Series (Inland Waters). Department of the Environment and Heritage, Canberra. http://www.ea.gov.au/soe/techpapers/index.html
  • Hossain I, Imteaz MA, Hossain MI. 2011. Application of build-up and wash-off models for an East-Australian catchment. International Journal of Civil & Environmental Engineering, World Academy of Science, Engineering and Technology 75(162): 942947.
  • Imteaz MA, Shanableh A, Asaeda T. 2009. Modelling multi-species algal bloom in a lake and inter-algal competitions. Journal of Water Science & Technology 60(10): 25992611.
  • Imteaz MA, Asaeda T, Lockington D. 2003. Modelling the effects of inflow parameters on lake water quality. Journal of Environmental Modeling & Assessment 8(2): 6370.
  • IPCC. 2008. Climate Change and Water, Bates B, Kundzewicz ZW, Wu S, Palutikof J (eds). IPCC Technical Paper VI, Inter Governmental Panel on Climate Change (IPCC), IPCC Secretariat: Geneva, 210 pp. ISBN: 978-92-9169-123-4.
  • Jha R, Herath S, Musiake K. 2000. River network solution for a distributed hydrological model and applications. Hydrological Processes 14: 575592.
  • Jones RN, Durack PJ. 2005. Estimating the Impacts of Climate Change on Victoria's Runoff Using a Hydrological Sensitivity Model. Prepared for the Department of Sustainability and Environment by CSIRO Atmospheric Research: Melbourne.
  • Jones R, Webb L. 2008. Sectoral and Regional Vulnerabilities to Climate Change in Victoria. CSIRO Marine and Atmospheric Research: Aspendale.
  • Leon LF, Soulis ED, Kouwen N, Farquhar GJ. 2001. Nonpoint source pollution: A distributed water quality modelling approach. Water Research 35(4): 9971007.
  • Lu F-H, Ni H-G, Liu F, Zeng EY. 2009. Occurrence of nutrients in riverine runoff of the Pearl River Delta, South China. Journal of Hydrology 376: 107115.
  • LVWSB. 1986. The Latrobe River Basin, Water Resources and Aquatic Environment: a Review, Introduction and Pictorial Overview. The Latrobe Valley Water and Sewerage Board (LVWSB): Traralgon, Vic3844, Australia.
  • Meybeck M. 1982. Carbon, nitrogen, and phosphorus transport by world rivers. American Journal of Science 282: 401405.
  • Perraud JM, Seaton SP, Rahman JM, Davis GP, Argent RM, Podger GM. 2005. The architecture of the E2 catchment modelling framework. In MODSIM 05 International Congress on Modelling and Simulation: Melbourne, Zerger A, Argent RM (eds). Modelling and Simulation Society of Australia: Canberra, 690696. Downloaded from http://www.mssanz.org.au/modsim05/index.htm
  • Paul KI, PolGlase PJ, O'Connell AMO, Carlyle JC, Smethurst PJ, Khanna PK. 2002. Soil nitrogen availability predictor (SNAP): a simple model for predicting mineralisation of nitrogen in forest soil. Australian Journal of Soil Research 40: 10111026.
  • Steffen W. 2009. Climate change 2009: faster change & more serious risk. Department of Climate Change, Australian Government, Canberra.
  • Tiller D, Newall P. 1995. Preliminary nutrient guideline for Victorian inland streams. Environment Protection Authority, Government of Victoria, Publication 478, downloaded from www.epa.vic.gov.au
  • Turner RE, Rabalais NN, Justic D, Dortch Q. 2003. Global patterns of dissolved N, P and Si in large rivers. Biogeochemistry 64: 297317.
  • Watson CL. 1980. Seasonal soil temperature regimes in south-eastern Australia, Australian Journal of Soil Research 18: 32531.
  • Whitehead PG, Wilby RL, Butterfield D, Wade AJ. 2006. Impacts of climate change on in-stream nitrogen in a lowland chalk stream: an appraisal of adaptation strategies. Science of the Total Environment 365(1–3): 260273.
  • Whitehead PG, Wilson EJ, Butterfield D. 1998a. A semi-distributed integrated nitrogen model for multiple source assessment in catchment. Part I. Model structure and process equations. Science of the Total Environment 210/211: 547558.
  • Whitehead PG, Wilson EJ, Butterfield D, Seed K. 1998b. A semi-distributed integrated flow and nitrogen model for multiple source assessment in catchments (INCA): part II – application to large river basins in South Wales and Eastern England. Science of the Total Environment 210/211: 55983.
  • Williams JR. 1975. Sediment-yield prediction with universal equation using runoff energy factor. Present and prospective technology for predicting sediment yields and sources, ARS-S-40, US Department of Agriculture. Agricultural Research Service 244252.
  • Williams JR, Berndt HD. 1977. Sediment yield prediction based on watershed hydrology. Transaction of American Society of Agricultural Engineers 20: 11001104.