HANIYAH SITEPU_Research Plan Presentation

Research Plan

1. RESEARCH PLAN
IMPROVING STREAMFLOW PREDICTION
WITH ARTIFICIAL NEURAL NETWORK AND
MERGED OF SATELLITE-GAUGE RAINFALL
DATA IN PROGO WATERSHED
HANIYAH SITEPU

2. OUTLINE
Introduction
Location
Supporting Data & Tool
Method
1
2
3
4

3. Drought, Farmers in Brebes Use Waste Water for Irrigation
BEWARE! Dry season is here.
Temanggung is at risk from drought and
clean water crisis
Farmers in Magelang Start Pumping Water to Irrigate Fields
“Streamflow data is one of the most
important quantities in hydrology.
It provides fundamental records for
water resources management and
climate change monitoring. Even
very short data gaps in this
information can cause extremely
different analysis outputs.”
- Tencaliec, 2015.
INTRODUCTION

4. BACKGROUND
The availability of streamflow
data in Indonesia is still
inadequate, and the distribution
of rainfall stations is uneven
–Syaifullah, 2014
.
The necessity of accurate,
reliable, and improved
streamflow Data
The utilization of rainfall satellites
as an alternative to improve
rainfall data prediction
–Suryaningtyas, 2019
.

5. Why are this study
important?
Excellent and accurate estimation
of streamflow data is needed for
water resources management in the
Progo watershed, especially for
irrigation management.
To produce a good estimation of
streamflow, accurate rainfall
data is required. However, the
distribution of rainfall stations in
the Progo watershed is still
uneven.
Irrigated Area
Non-Irrigated Area
AWLR
Rain Gauge
The uneven distribution of rainfall does not
necessarily describe the rainfall that occurs in the
entire Progo watershed. Thus, a method is
needed to improve the accuracy of rainfall and
streamflow estimation in the Progo watershed.

6. References & Gap Analysis
Ayoub et al (2016) &
Hossaini et al (2022)
predicting streamflow
data with ANN.
What’s new with this study?

This study was conducted to
improve the prediction of
streamflow simulations
generated by ANN, by
combining observational and
satellite rainfall data as input
data.

In this study, the combination of
rainfall and satellite data was
carried out using the Kagan
Rodda method. Kagan rodda is
generally used to rationalise the
placement of rain stations, so the
use of kagan rodda in this study
is relatively new.
Zhu et al (2019) &
Mekomnen et al (2022)
merge satellite and
observation
precipitation data to
enchance the discharge
prediction.

7. 1.To evaluate wheter satellite rainfall data can be used to
estimate rainfall in the in Progo watershed.
3.To compare and analyze the streamflow prediction
results form with and without satellite.
2.To assess the ANN model’s suitability for predicting
streamflow in the Progo watershed.
Target of the
Research

8. Study location
Progo Watershed is is located between 7
o
12' to
8
o
04' east and 109
o
59' to 110
o
291' south. With
an area of 2370.6 km
2
, This watershed is the
source of water supply in Magelang City,
Temanggung, Wonosobo, Magelang, Semarang,
Boyolali, Sleman, Kulon Progo, and Bantul
Regency, With an area of 2370.6 km
2
.


In this study, the location will be focused on the
upstream area of Progo watershed. It is a section
of river flow measured by the Kalibawang
AWLR, with a total area of 1764,42 km
2
.

9. Study location
Progo Watershed is is located between
7
o
12' to 8
o
04' east and 109
o
59' to 110
o
291'
south. With an area of 2370.6 km
2
, This
watershed is the source of water supply in
Magelang City, Temanggung, Wonosobo,
Magelang, Semarang, Boyolali, Sleman,
Kulon Progo, and Bantul Regency, With an
area of 2370.6 km
2
.

PROGO’s
Upstream
section:
Progo’s Upstream
Watershed
AWLR
Rain Gauge

10. Data & Tools That Are Used
Progo Watershed Map Observation and Satellite Rainfall Data
AWLR/Measured
Streamflow Data
Climatology Data Rain Station/Gauge
Coordinate

11. Climatology
Data
Observation
Rainfall Data
Satellite
Rainfall
Data
Progo Watershed
Map, and coordinate
of station in it
Data Quaility
Test
Data Quality
Test
Calculation of
Evapotranspiration
Calculation of Areal Rainfall
Data (1
st
Scenario)
Creating Thiessen
Polygon
Streamflow
Prediction with
ANN modelling
AWLR
Data
Satellite Data
Callibration
Progo Watershed
Map, and coordinate
of station in it
Creating Thiessen
Polygon
Climatology
Data
Calculating
evapotranspiration
Calculation of Areal Rainfall
Data (2
nd
Scenario)
Define a new
rain station point
Streamflow
Prediction with
ANN modelling
Streamflow
Simulasion Result’s
evaluation
Comparing the 1
st

and 2
nd
streamflow
simulation.
ANN Target
Conclussion and
suggestion
Finish
Creating Kagan
Rodda Triangle
Observation
Rainfall Data
Data Quaility
Test SCENARIO 1 SCENARIO 2

12. RESEARCH TIMELINE
Research stage
2023 2024
June July August Sept Oct Nov Dec Jan Feb March Apr May June
Research activities
1 Collecting Data

2 Mapping and data compile
3 Perform data quality test
4
Evaluating the suitability of
satellite rainfall data


5 Analysis of Kagan Rodda

6
Calculation of data requirements
for ANN model (Areal rainfall,
evapotranspiration)


7
Modeling streamflow prediction
using artificial neural network


8
Comparing modeling results and
draw a conclusion


9 Writing thesis reports and journals

10 Thesis Seminar
11 Journal Publication