MODELLING THE CHLOROPHYLL-A CONCENTRATION OF LAGUNA LAKE USING HIMAWARI-8 SATELLITE IMAGERY AND MACHINE LEARNING ALGORITHMS FOR NEAR REAL TIME MONITORING

Recent studies have investigated the use of satellite imaging combined with machine learning for modelling the Chlorophyll-a (Chl-a) concentration of bodies of water. However, most of these studies use satellite data that lack the temporal resolution needed to monitor dynamic changes in Chl-a in productive lakes like Laguna Lake. Thus, the aim of this paper is to present the methodology for modelling the Chl-a concentration of Laguna Lake in the Philippines using satellite imaging and machine learning algorithms. The methodology uses images from the Himawari-8 satellite, which have a spatial resolution of 0.5–2 km and are taken every 10 minutes. These are converted into a GeoTIFF format, where differences in spatial resolution are resolved. Additionally, radiometric correction, resampling, and filtering of the Himawari-8 bands to exclude cloud-contaminated pixels are performed. Subsequently, various regression and gradient boosting machine learning algorithms are applied onto the train dataset and evaluated, namely: Simple Linear Regression, Ridge Regression, Lasso Regression, and Light Gradient Boosting Model (LightGBM). The results of this study show that it is indeed possible to integrate algorithms in Machine Learning in modelling the near real-time variations in Chl-a content in a body of water, specifically in the case of Laguna Lake, to an acceptable margin of error. Specifically, the regression models performed similarly with a train RMSE of 1.44 and test RMSE of 2.51 for Simple Linear Regression and 2.48 for Ridge and Lasso Regression. The linear regression models exhibited a larger degree of overfitting than the LightGBM model, which had a 2.18 train RMSE.

A Transfer Learning Technique for Inland Chlorophyll-a Concentration Estimation Using Sentinel-3 Imagery

Chlorophyll-a (Chla) concentration, which serves as a phytoplankton substitute in inland waters, is one of the leading indicators for water quality. Generally, water samples are analyzed in professional laboratories, and Chla concentrations are measured regularly for the purpose of water quality monitoring. However, limited spatial water sampling and the labor-intensive nature of data collection make global and long-term monitoring difficult. The developments of remote-sensing optical sensors and technologies make the long-term monitoring of Chla concentrations for an entire water body more achievable. Many studies based on machine learning techniques, such as regression and artificial neural network (ANN) methods, have recently been proposed for Chla concentration estimation using optical satellite images. The methods based on machine learning can achieve accurate estimation. However, overfitting problems may arise because the in situ Chla dataset is generally insufficient to train a complicated machine learning model, which makes trained models inapplicable. In this study, an ANN model containing three convolutional and two fully connected layers with 4953 unknown parameters is designed. A transfer learning method, consisting of model pretraining, main-training, and fine-tuning stages, is proposed to ease the problem of insufficient in situ samples. In the model pretraining stage, the ANN model is pretrained and initialized using samples derived from an existing Chla concentration model. The pretrained ANN model is then fine-tuned using the proposed transfer learning technique with in situ samples collected in five different campaigns carried out during early 2019 from Laguna Lake, the Philippines. Before the transfer learning, data augmentation and rebalancing methods are conducted to enrich the variability and to near-uniformly distribute the in situ samples in Chla concentration space, respectively. To estimate the alleviation of model overfitting, the trained ANN model, using an in situ dataset from Laguna Lake, was tested using an in situ dataset from Lake Victoria, Uganda, obtained in 2019, which has a similar trophic state as Laguna Lake. The experimental results from Sentinel-3 imagery indicated that the overfitting problem was significantly alleviated and the trained ANN model outperformed related models in terms of the root-mean-squared error of the estimated Chla concentrations.

“Wild fish are a blessing”: changes in fishing practices and folk fish cuisine around Laguna Lake, Northern Philippines

Several coastal communities rely heavily on wild-caught fish for personal consumption and their livelihoods, thus being sensitive to the rapid global change affecting fish availability. However, in the last century, aquaculture has been increasingly adopted. To understand the uses and changes of wild-caught fish, we conducted 30 semi-structured interviews with fishers of Laguna Lake, Philippines. Fishermen, with up to 60 years’ experience, reported catching 31 fish species as a staple food. The taxa with the greatest variety of food uses were the farmed Oreochromis aureus, and the wild Channa striata and Cyprinus carpio. Fish was boiled, fried, grilled and dried, and over 20 different local dishes were reported. Fishers reported that local communities previously relied more on wild fish, while today a greater proportion of consumed fish comes from aquaculture fish species such as Oreochromis aureus and Hypophthalmichthys nobilis. Wild fish remains a crucial aspect of local gastronomic diversity, underpinning the biodiversity of the Laguna Lake, while also representing an important element for food sovereignty. The study stresses the need to sustain local ecological knowledge to ensure the ecological, social and economic sustainability of the communities.

Microbial source tracking of fecal contamination in Laguna Lake, Philippines using the library-dependent method, rep-PCR

Laguna Lake is an economically important resource in the Philippines, with reports of declining water quality due to fecal pollution. Currently, monitoring methods rely on counting fecal indicator bacteria, which does not supply information on potential sources of contamination. In this study, we predicted sources of Escherichia coli in lake stations and tributaries by establishing a fecal source library composed of rep-PCR DNA fingerprints of human, cattle, swine, poultry, and sewage samples (n = 1,408). We also evaluated three statistical methods for predicting fecal contamination sources in surface waters. Random forest (RF) outperformed k-nearest neighbors and discriminant analysis of principal components in terms of average rates of correct classification in two- (84.85%), three- (82.45%), and five-way (74.77%) categorical splits. Overall, RF exhibited the most balanced prediction, which is crucial for disproportionate libraries. Source tracking of environmental isolates (n = 332) revealed the dominance of sewage (47.59%) followed by human sources (29.22%), poultry (12.65%), swine (7.23%), and cattle (3.31%) using RF. This study demonstrates the promising utility of a library-dependent method in augmenting current monitoring systems for source attribution of fecal contamination in Laguna Lake. This is also the first known report of microbial source tracking using rep-PCR conducted in surface waters of the Laguna Lake watershed.

Library-independent source tracking of fecal contamination in selected stations and tributaries of Laguna Lake, Philippines

Laguna Lake is the largest inland freshwater body in the Philippines. Although it is classified to be usable for agricultural and recreational purposes by the country's Department of Environment and Natural Resources (DENR), studies looking at lake ecology revealed severe fecal contamination which contributes to the deterioration of water quality. Determining the sources of fecal contamination is necessary for lake protection and management. This study utilized a library-independent method of microbial source tracking (LIM-MST) to identify sources of fecal contamination in selected Laguna Lake stations and tributaries. Genetic markers of the host-associated Escherichia coli, heat-labile toxin (LTIIA) and heat-stable II (STII), were used to identify cattle and swine fecal contaminations, respectively. Meanwhile, human mitochondrial DNA (mtDNA) was used to identify human fecal contamination. Results identified the presence of agricultural and human fecal contamination in Laguna Lake Stations 1 and 5, Mangangate River, and Alabang River. The selected sites are known to be surrounded by residential and industrial complexes, and most of their discharges find their way into the lake. The identification of the specific sources of fecal contamination will guide management practices that aim to regulate the discharges in order to improve the water quality of Laguna Lake.

Predominance of blaTEM and tetA genes in antibiotic-resistant Escherichia coli isolates from Laguna Lake, Philippines

Lakes are one of the sinks of antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs); however, information on ARB and ARGs in lakes in the Philippines is scarce. In this study, Escherichia coli was isolated from the largest freshwater lake in the Philippines, Laguna Lake, to detect antibiotic resistance and the presence of ARGs. Broth microdilution assay (BMA) and molecular identification of five environmentally prevalent ARGs (strA, blaCTX-M, blaSHV, blaTEM, and tetA) were performed. The majority (75.70%) of the isolates harbored at least one of the targeted antibiotic genes. Multiplex PCR detected about 49.07% of the isolates had genes for extended-spectrum β-lactamases (ESBL), which were mostly represented by blaTEM (47.66%). The genes strA and tetA were observed in this study with detection frequencies of 29.91 and 45.33%, respectively. About 95.69% of thermotolerant E. coli isolates were non-susceptible to six different antibiotics using BMA. Nearly 37% of the isolates were found to be multidrug-resistant (MDR) with most isolates resistant to ampicillin (81.72%). Furthermore, the occurrence of ESBL genes was significantly correlated with tetA genes (P = 0.013). To date, this study is the first to report on the presence of MDR and thermotolerant E. coli in Laguna Lake, Philippines.

Molecular surveillance of Cryptosporidium spp. for microbial source tracking of fecal contamination in Laguna Lake, Philippines

Water quality deterioration in source waters poses increased health, environmental, and economic risks. Here, we genotyped Cryptosporidium spp. obtained from water samples of Laguna Lake, Philippines, and its tributaries for the purpose of source-tracking fecal contamination. A total of 104 surface water samples were collected over a 1-year period (March 2018 to April 2019). Detection of Cryptosporidium was carried out using genus-specific primers targeting a fragment of the small subunit (SSU) rRNA gene. The study revealed 8 (14%) tributary samples and 1 (2.77%) lake sample positive for contamination. The species were determined to be C. parvum (n = 4), C. muris (n = 2), C. hominis (n = 1), C. galli (n = 1), C. baileyi (n = 1), C. suis (n = 1), as well as rat genotype IV (n = 1). Two species were detected in duck (C. baileyi) and cattle (C. parvum) fecal samples. The data presented suggest that Cryptosporidium contamination is likely to come from sewage or human feces as well as various agricultural sources (i.e. cattle, swine, and poultry). This information reveals the importance of mitigating fecal pollution in the lake system and minimizing health risks due to exposure to zoonotic Cryptosporidium species.