Minimising agricultural pollution to enhance water quality in Laguna de Bay (Philippines) and Mt Lofty Ranges (Australia)

This project aims to develop a comprehensive biophysical and hydrological characterisation of the Pagsanjan-Lumban sub-catchment of Laguna de Bay, and to identify and quantify the sources of pollutants (sediments, nutrients and pesticides) from agricultural activities and their potential impact on the health of Laguna de Bay in the Philippines. Complementary activities will be undertaken as part of a larger research programme to protect water bodies in the Mt Lofty Ranges in Australia.

Major activities undertaken during this reporting period in the Philippines and Australia include:

Philippines component
The project was officially launched on Monday the 11th Sept. 2006 by the Australian Ambassador to the Philippines, Mr Tony Hely, at the LLDA offices in Calauan, Laguna. This was attended by Drs Rai Kookana, Jim Cox, Ian Willet and Ms Danni Oliver and Ms Sue Cuddy from Australia, and by the majority of the staff involved with the project in the Philippines.
A tour of the Pagsanjan-Lumban sub-catchment (study site) was made on Sept. 9 to allow the Australian collaborators to understand the sources of contaminants to the rivers draining into Laguna de Bay.
A highly successful two-day workshop was held on Sept. 12 and 13 wherein the team went through all of the project objectives/activities and agreed upon staff involvement and activities that needed to be undertaken to meet the objectives.
As a follow up from the project launch and the workshop, a meeting between the LLDA and UPLB teams was held on October 10 to go through collaborative arrangements between the organisations and agree on specific assignments of individual team members. A coordination meeting was held on December 7 to follow up on individual commitments and contributions to the Annual Progress Report.
Meetings with members of the Pagsanjan-Lumban River Basin Management and Development Foundation, Inc. were held on September 8 and October 12 to discuss possible engagement of volunteers for specific project activities.
A training module on GIS and RS technology has been developed by the LLDA GIS Team.
The criteria for the selection of sites for instrumentation such as auto samplers have been developed by Dr. Sammy Contreras.
The description of sampling methodology and protocols for the eco-toxicological studies that will be conducted under the project has been developed by Ms Cristy Bajet.
Members of the project team from Philippines visited Australia under another project and visited Australian field sites and had discussions with colleagues in CSIRO and other agencies.

Australian Component
In the Lenswood Creek sub-catchment of the Mt Lofty Ranges, study sites were selected in consultation with stakeholders including Apple and Pear Growers Association, SA Water, EPA and other agencies.
Four auto-samplers and flow data loggers at three sites in the Mt Lofty Ranges have been installed.
A stakeholders meeting allowed several agencies to visit study sites after installation of auto samplers.
The analytical methods for pesticide analyses for over 60% of the chemicals used in the three landuses being studied, namely grapes, apples and cherries, have been developed and optimised.
Analyses have been made on the water collected so far for sediment, nutrients and pesticides. However, the very low rainfall this year has resulted in only a small number of water samples that could be collected.
Five growers have been interviewed so far about chemical use in the Mt Lofty Ranges and a PIRI assessment made of chemical usage at the three sites where the auto samplers are located.

This report summarises research that was conducted during the period 1 January 2007 to 31 May 2008.

Philippines component
1. Installation of auto-samplers
Initially in January 2007, Dr. Sammy Contreras and his staff conducted a reconnaissance survey of the watershed to identify possible sites where the auto-samplers and the water level loggers could be located (Appendix 1). The four sites selected in February 2007 were chosen to represent the major landuses in the sub-catchment, namely production of rice, vegetables and coconut and piggeries since the latter is a major source of nutrients. In early February 2007 Australian staff visited the Philippines and two auto-samplers and Odyssey water level loggers were installed, with the help of the Filipino team, at Salasad (rice production) and at Lucban (vegetable production). Several weeks later Filipino staff installed the other two auto-samplers at the final two sites at Caviniti (coconut production) and at Majayjay (piggery township) (Appendix 2). Due to problems with the ownership of land at the Majayjay site this auto-sampler had to be relocated in April 2008 further down the Initian Creek near the San Francisco bridge.

There have been various problems with the auto-samplers so only water from Lucban has been collected in 2007. In April 2008 Nigel Fleming, from the Australian team, spent a week recalibrating the water level loggers, performing general maintenance on the auto-samplers and trouble-shooting any problems.

During the visit by the Australian staff in February 2007, UPLB and LLDA staff members were trained in the operation of auto-samplers, re-setting, maintenance and trouble-shooting of auto-samplers and in downloading data from the Odyssey water level loggers. Detailed written instruction sheets were also provided for the operation and trouble-shooting of the auto-samplers and the water level loggers.

LLDA staff then trained the volunteers who collected the water samples daily at each site and stored them in the fridge until collection by project staff at the end of the week for transfer to the laboratory. Details of the collection procedure were written in English and Tagalog (the Philippines national language). Details of training in daily water collection from the auto-samplers are given in Appendix 3.

Commencement of analyses
The auto-sampler at Lucban is working well and samples are composited weekly in the laboratory for pesticide and nutrient analyses while TSS samples are analysed as daily samples. The collection of samples from Caviniti started on July 30 2007, but after couple of months the auto-sampler started malfunctioning. Bridge construction has forced many of the residents to move their laundry area from downstream of the site to upstream. Due to possible effects of detergents on phosphate analyses the samples collected by the auto-sampler during the daytime (noon and 6 pm) are kept separate from those collected at night (midnight and 6 am) to determine whether detergents are having an impact on phosphate levels in the water. Sample collection at the Salasad Creek has temporarily ceased because of the construction of a bridge immediately upstream which may impact on total suspended sediment (TSS) and pesticide results. Collection was resumed in January 2008. Consequenlty there is little data from this site. The auto-sampler and water level loggers at Majayjay have been relocated further downstream on the Initian creek where the piggery waste flows to the San Francisco bridge.

Microbiological analyses were made of Majayjay water samples because of OHS concerns for the safety of the volunteers collecting the water samples. Appropriate protective gear has been provided for the volunteers to conduct the sampling.

Results from the survey of growers about pesticide use are given in Appendix 4. Data collected from 2007 sampling of water from Lucban are given in Appendix 5.

Installation of rain gauges
During the launch of the project in September 2006 several options were identified to overcome the paucity of data that was needed for the models currently used in the Philippines. An scoping document outlining how volunteers could help accumulate data, primarily TSS, rainfall and water depth measurements, for the project was produced.

It was decided that the paucity of rainfall data for the catchment could be overcome by engaging schools within the watershed in the collection of rainfall data. In early March 2007 rain gauges were installed in the nine (9) identified schools and on March 9 2007 two high school students and one teacher from each of the schools were trained in using the rainfall gauge to measure rainfall (Appendix 6). The rainfall data are collected by Roman Corpuz, LLDA, who visits each school monthly. However, considering the distance of the rain gauge sites from the LLDA Office, the use of text messaging was proposed. Basically, the volunteers will be provided with prepaid cards regularly and then they will be asked to send through the daily rainfall data by text messaging. The collected data shall be placed in the HYMOS database established at the LLDA Calauan Office for utilisation in models. The location of the rainfall gauges and the auto samplers in the catchment is shown in Fig. 1. Examples of the rainfall data obtained so far are given in Appendix 5.

At this stage volunteers have not been trained in determining TSS and stage height measurements.

4. Ecotoxicological testing
The effect of selected pesticides, namely deltamethrin, lambda cyhalothrin and malathion, on the growth of duckweed has been evaluated across a range of concentrations. The mortality of Tilapia fingerlings in field collected water from Majayjay diluted at 0%, 50% and 75% has also been tested. The results from these experiments are given in Appendix 7.

Sediment toxicity tests of pesticides are planned in the future using mollusk eggs which will be collected and exposed to sediment/soil spiked with pesticides. Hatching success will be measured at designated time intervals. Possible test species are the local kuhol (Pila luzonica) or the golden apple snail (Pomacea canaliculata), which was introduced into the country and is now a pest of rice.

5 Water balance and erosion modelling and GIS mapping.
There has been very little work done yet on the water balance and erosion modelling. Some preliminary result of the application of the grid-based distributed hydrological model WinBTOPMC model in the Pagsanjan River Basin is in general very encouraging, given the limited information available, specifically on observed discharge hydrograph, the hydrological impact of site geology, and other unaccounted factors (e.g river siltation/erosion, diversion loss, etc.) (Appendix 8). To the extent that more accurate information on input spatial datasets, such as a detailed soil and land cover maps, and accurate discharge records become available, it is expected that the model results could be further improved.

It is planned to use the model PCARES for the erosion modelling. PCARES is a GIS-based model that has been used in the Maupali River watershed. It requires time-series rainfall data which is not available in the Pagsanjan-Lumban catchment but during the mid-term review the LLDA staff were able to obtain time-series rainfall data from an automatic rainfall gauge located in a nearby catchment, Rizal. The erosion modelling should be finished by mid 2008.

(image withdrawn - see report) Fig. 1. Location of schools where rainfall gauges and autosamplers have been installed in the catchment.

Australian Component
1. Continued analysis of water samples for sediment, nutrient and pesticides.
The three sites that have been instrumented in the Mt Lofty Ranges have received intermittent rainfall since their establishment in July 2006. No water flowed into or out of the lower dam at the grapes site until May 21 2007 and no water flowed at the cherries site until April 28 2007. The apples site is the only one that has flowed regularly but not consistently.

All water samples that have been collected have been analysed for sediments, nutrients (total N, P, C, nitrate and metals) and pesticides. A summary of the data found so far is given in Appendix 9.

Completion of interviews with growers about chemical use.
A total of 21 growers have been interviewed covering four main landuses in the Mt Lofty Ranges, namely the production of apples, cherries, grapes and pears. A separate assessment of the relative risk of off-site transport of pesticides has been made for each individual grower using the tool, Pesticide Impact Rating index (PIRI), and a booklet outlining the findings produced and supplied to each grower.

From the general site and chemical usage information, gathered from the broad cross-section of growers, scenarios have been developed that are representative of the conditions in the Mt Lofty Ranges and these have been run through PIRI. From the results obtained from these scenarios information sheets, called "Chemical Reference Charts", have been produced and distributed to all growers involved in the survey for feedback. Once the format has been finalised the "Chemical Reference Charts" will be distributed to growers and other interested stakeholders in the region. These sheets will allow growers to compare the relative risk of off-site transport to surface water of different chemicals in a range of environments that are representative of those in the Mt Lofty Ranges.

Training farmers in the use of PIRI
Many growers have expressed interest in using PIRI so a proposal was written for conducting a training program and presented to the Apple and Pear Board, South Australia. We are still awaiting on comment on the proposal and the likelihood of obtaining funding to run the training course.