Indonesia: Java Flood One

Lead Research Organisation: UK Centre for Ecology & Hydrology
Department Name: Hydro-climate Risks

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Planned Impact

The objective of this project is to create social and economic benefits by developing a set of medium term flood forecast (MRFF) tools for the urban centres of Jakarta, Bandung and Surakarta on the island of Java in Indonesia. The skills and development associated with this project will be strongly embedded within Bandung Institute of Technology (ITB) and will depend on free software packages and cost effective data acquisition solutions such that ITB will be able to independently develop similar MRFF tools for other cities in Indonesia following the cessation of the NERC funding. It is envisaged that developed workflows and methodologies can also be applied to flood affected ODA countries beyond Indonesia in the future.

Flooding represents the most frequently occurring hydrometreological hazard in Indonesia, contributing to around 31% of all disaster events recorded by the National Disaster Management Agency. Impacts from fluvial flooding in Jakarta alone are estimated to cost around USD 321 million per year. The social benefits of MRFF are difficult to estimate but the economic benefit of flood awareness alone is of the order of USD 400 for every USD 1 invested.

The new MRFFs will bring benefits to communities beyond conventional early warning systems by enabling conservation of livelihoods as well as lives. Economic benefits associated with additional lead in time provided by medium term forecasts include avoidance of damage and loss associated with flood events and reduction of relief and rehabilitation costs. In addition to this, newly installed hydrological instruments along with local NGO training will provide additional facilities to improve accuracy for local early warning flood systems, imperative for effective evacuation planning. New sub-catchment scale medium range weather and stream flow forecasts will also help improve agricultural planning and output as well as increased water security. The project will also provide new facilities and training for ITB to determine dynamic topographical changes anywhere in Indonesia. Beyond flood forecasting, the ability to acquire such data will be of significant benefit to infrastructure and town planning agencies. Yielded value could include, for example, avoidance of new construction work in areas significantly affected by subsidence.

The research is designed using participatory approaches to enrol potential end users into co-developing outcomes and outputs to improve flood risk communication. The project will develop and run a series of training workshops and public engagement sessions to launch and cascade the new flood risk communication tools and additional benefits in the study areas. These sessions will be co-delivered by the UK and Indonesian project teams.

We will be working with a range of partners to deliver impacts including government departments such as BNBP/BPBD (Indonesia's Disaster Management Agency), IAHRI (Indonesian Agro-climate and Hydrological Research Institute), PT Reasuransi MAIPARK and local NGOs such as Jaga Balai and Ciliwung Medeka. These latter organisations represent local community groups who have set up their own flood warning system using real-time and historical visual observations.

Publications

10 25 50
 
Title Calibration metrics corresponding to randomly sampled parameter values for a suite of hydrological models 
Description For each combination of 21 rainfall-evapotranspiration inputs, 24 models and 38 river flow gauging stations: 100,000 sets of randomly sampled model parameter values, the Kling-Gupta Efficiency (KGE) between the modelled and observed daily mean flow corresponding to each randomly sampled parameter set, and, for each parameter, the Kolmogorov-Smirnov statistic between the full set of 100,000 values and the 1000 values corresponding to the highest 1000 KGE scores. Please note that the EIDC link supplied only includes the one of the 100,000 sets of randomly sampled model parameter values for each combination of 21 rainfall-evapotranspiration inputs, 24 models and 38 river flow gauging stations that gives the single highest KGE score in each case. 
Type Of Material Database/Collection of data 
Year Produced 2022 
Provided To Others? Yes  
Impact This research dataset is used to identify which model structures are best suited to estimation of daily river flows in tropical monsoon climates. Areas of the world with tropical monsoon climates are highly populated and subject to intense rainfall every year, but have very rarely been the focus of hydrological model development. An article using this research dataset is in preparation for the Journal of Hydrology. 
URL https://catalogue.ceh.ac.uk/documents/f6cec7d4-edee-44b8-8f44-86d4f12ac72d
 
Title Quality-controlled and machine-readable daily flow data for river gauging stations in Java, Indonesia 
Description Quality-controlled and machine-readable derivatives of original flow data provided by Indonesian project partners for 140 river flow gauging stations in Java, Indonesia. I do not have permission to share this database outside of this project as I do not own distribution rights to the original data that it was derived from. 
Type Of Material Data handling & control 
Year Produced 2019 
Provided To Others? No  
Impact This dataset was necessary in the development, evaluation and calibration of hydrological models for the Citarum and Ciliwung catchments, and subcatchemtns within them. 
 
Title Rainfall and evapotranspiration time-series from six sources for 56 catchments in Java 
Description Catchment-average rainfall time-series at daily resolution for 56 catchments in Java, Indonesia, from the following sources: AgMERRA, KNMI, MSWEP, PERSIANN, SACA&D, Yanto et al. Catchment-average evaporation time-series at daily resolution for 56 catchments in Java, Indonesia, from the following sources: SACA&D, Yanto et al. 
Type Of Material Data handling & control 
Year Produced 2021 
Provided To Others? No  
Impact Rainfall and evaporation time-series are used in 12 combinations (6x2) as input data to calibrate hydrological models at daily resolution. Uncertainty as a result of input data (as opposed to model structural error) is quantified through different calibration performance with different input data. 
 
Title Statistics and distribution parameters of catchment-average rainfall in Citarum basin, Java 
Description L-moments, mean and median annual values and fitted Generalized Extreme Value (GEV) parameters, assuming either stationarity or trends against time, for 1-10 day catchment-average accumulated rainfalls over the ~6000 km2 Citarum basin in Java. 
Type Of Material Data handling & control 
Year Produced 2020 
Provided To Others? No  
Impact This dataset can help to estimate or quantify trends (increases or decreases) in large rainfalls over the past 30-40 years and can form the basis of a depth-duration-frequency (DDF) model of the type used for flood risk assessments in the UK and other countries. 
 
Title Suite of hydrological models and calibration script 
Description Suite of 24 scripts in the R programming language, each implementing one hydrological model (using rainfall and evaporation time-series as inputs and producing time-series of river flow and internal model fluxes and storage states as outputs). Wrapper R script to calibrate (i.e. find optimal parameter values for) each of the models, using recorded flow data as the target value. The hydrological models were either created by me or based on Open Source interpretations of models not created by others. The calibration wrapper was created by me, incorporating one function each from the R packages "hydroGOF" and "hydromad". This suite expanded in early 2021 to work with input data at sub-daily resolution (if 24 hours is composed of an integer number of timesteps). Prior to this, the suite could only use daily input data. Corrections to the models were also made as a result of testing in 2020. Four more models were added over 2021 and early 2022 and additional corrections were made to one existing model. A wrapper R script for parameter sensitivity analysis was added during 2021 and updated in early 2022 to be used with the four models that were not part of the original suite. 
Type Of Material Computer model/algorithm 
Year Produced 2020 
Provided To Others? No  
Impact The best-performing members of this suite of hydrological models are able to estimate daily mean flow rates satisfactorily in different parts of the river network within the Citarum and Ciliwung river basins when properly parameterized. 
 
Description UKCEH News Post: New projects will help reduce devastating impact of extreme weather 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact A News item was added to UKCEH's website to advertise three new projects involving UKCEH, all related to the UKRI Call "Understanding of the Impacts of Hydrometeorological Hazards in South East Asia". I was asked to provide information about Java Flood One, which was written up by UKCEH's Media Relations Officer.
Year(s) Of Engagement Activity 2019,2020
URL https://www.ceh.ac.uk/news-and-media/news/new-projects-will-help-reduce-devastating-impact-extreme-w...