METHODOLOGY

A mixed method of research: Qualitative and quantitative methods

are used.

Combination of secondary and primary method shall be used.

Qualitative method include interviews, questionnares that supports

quantitative method and their findings.

DATA SOURCES

Primary data sources

 Include data generated through
questionnaires ,interviews ,survey ,observations etc

Secondary data sources

 Include existing data generated by verified and official entities or
agencies
Fig. 1. Flowchart describing Workflow
of Study
STEPS INVOLVES IN CALCULATING ROOF AREA

 Access the Google Earth interface

 Select desired area/district and zoom in to it
 Select 50 spatially well-differentiated buildings within the
boundary of the district
 Select the “Measure” tool on the Google Earth interface and
sequentially click on the boundaries of a roof of the building
 Close the loop of linear measurement by clicking on the first
selected point
 The area and perimeter of the closed loop is presented in a box
 Repeat steps 4 through 6 for 49 other buildings
 Calculate the average roof area of the 50 roofs (total area/50)
 Study Area
 Nepal is a landlocked country which covers total area of147,181
sq.km
 Nepal lies in 26°N to 31°N latitudes and 80°E to 89°E longitudes
 Nepal extends roughly 200km from north to south and 900 km from
east to west
 Nepal is divided into Himal,Hill and Terai and varies at elevation of
60 m above sea level (south) to 8848 m above sea level (Mt. Everest,
north)
 Nepal has a Extreme topographical variation over a short distance
and Results in significant spatial variation in rainfall distribution
 Monsoon season contributes approximately 80% of total annual
rainfall in Nepal
 Other seasons collectively account for the remaining 20%
throughout the year (Upadhyay et al., 2022)
Fig. 2. Study Area and Distribution of Rain Gauge Stations
Rain Gauge Data and Satellite Precipitation
Data
 Daily rainfall data from 30 rain gauge stations spread throughout
Nepal were obtained with Precise coordinates obtained from the
Department of Hydrology and Metrology (DHM), including
latitude, longitude, and elevation
 chirps daily rainfall data series were obtained by extracting the
daily rainfall over the pixel in which each rain gauge station was
installed.
 Ground station data and CHIRPS data aggregated to monthly and
annual time steps to validate the accuracy of CHIRPS-derived
data by comparing it with observation data from ground stations.
 To evaluate the overall accuracy and reliability of CHIRPS rainfall
data
 DHM considered as a reliable authority for obtaining precise
location details of the rain gauge stations
DHM Rain Gauge Data

 Daily rainfall observation data gathered for the past 30 years (1993
to 2023
 30 rain gauge stations operated and maintained by the Department of
Hydrology and Metrology (DHM), Nepal
 Efforts made to ensure diverse representation by selecting well-
differentiated station locations
 Distribution of stations designed to capture varied geographic and
climatic characteristics
 Evaluation of observational data for its suitability and reliability for
use in the study
 Numerous gaps identified in the observational data for every station
 Reasons for gaps include equipment downtime, observer's neglect in
recording data, resulting in smaller gaps
 Faulty equipment leading to abnormal functioning or
installation/removal of stations during the study period, resulting in
larger data gaps.
CHIRPS
 Quasi-global rainfall dataset covering 50°N to 50°S latitude range
and all longitudes
 Temporal coverage spans from 1981 to near-present
 Primary computing time step is the pentad (five-day interval)
 Produces precipitation data in various time steps by aggregating
(pentadal to monthly or dekadal) or disaggregating (pentadal to
daily) pentadal data
 This study specifically utilizes daily CHIRPS precipitation data
 Aggregates the daily data to derive monthly, seasonal, and annual
precipitation datasets
 CHIRPS data is employed for assessing precipitation patterns and
trends in the specified study
 Generation of Daily Data from Monthly Data
for Monthly Analysis

 Preference for daily rainfall data in modeling rainwater harvesting

systems (Heggen, 1993; Fewkes and Butler, 2000; Thomas, 2002;
Karin et al., 2015).
 Monthly averaged data can be used but with reduced accuracy
(Thomas, 2002).
 Transformation of monthly average data to daily data involves
uniformly spreading each month's rainfall across its days.
 In this study, monthly scale analysis of rainwater harvesting
reservoirs utilizes daily data derived from mean monthly rainfall
distribution.
 The obtained daily data is then subjected to analysis similar to daily
data from CHIRPS and ground observation stations.
 Behavioral Model
 Behavioral models have been used by researchers for simulating
rainwater harvesting system performance.
 Jenkins et al. introduced a behavioral model for rainwater collection
systems in1978 i.e yield before spillage(YBS) and yield after
spillage(YAS)
 Fewkes and Butler conducted subsequent research in 2002 and
concluded that the YAS algorithm provides a conservative estimate
of system performance compared to the YBS algorithm.
 The operating rule of the YAS algorithm is as follows:

Yt is the yield from the tank during the time interval t .

Vt is the volume of rainwater in the tank at the end of time interval t
Qt is the volume of rainwater that enters the tank during time interval t
Dt is the demand during time interval t
S is the maximum storage capacity
Summary of Statistical Metric table
 Statistical Metrics
 Numerous studies employ observation data from rain gauge stations to
validate satellite-derived precipitation data
 Coefficient of determination is a statistical measure assessing the
strength of the relationship between observed and satellite-derived
precipitation data.
 Percentage Bias quantifies the percentage difference between observed
and satellite-derived values.
 RMSE-Observations Standard Deviation Ratio evaluate Ratio of Root
Mean Square Error to the Standard Deviation of Observations.
 Kling-Gupta Efficiency is a metric assessing the goodness-of-fit between
observed and modeled values.
 An adaptation of the Kling-Gupta Efficiency for enhanced accuracy
assessment.
 Nash-Sutcliffe Efficiency: Evaluates the predictive accuracy by comparing
the observed mean with the modeled mean.
 The use of multiple statistical metrics provides a comprehensive
evaluation of the accuracy of satellite-derived precipitation data
 These metrics collectively contribute to a detailed understanding of the
performance of satellite precipitation data in comparison to ground-
based observations
Fig.Summary of Performance Rating for Statistical Metrics
table
Reliability for Predetermined Tank Sizes
 Performance metrics are crucial for evaluating the effectiveness of
rainwater harvesting systems.
 Reliability, defined as the fraction of days satisfying water demand, is

a common and vital performance metric (Thomas, 2002).

 Other metrics like satisfaction and efficiency exist but reliability is

more widely adopted and easier to interpret.

 Reliability serves as a singular metric due to its widespread usage and

straightforward interpretation.
 Rainwater harvesting system performance is dependent on various

factors.
 Installation characteristics, such as rainwater harvesting tank storage

capacity, significantly influence system performance .

 The reliability of a rainwater harvesting system is naturally linked to

its storage capacity.

 Different systems with varying storage capacities will exhibit unique

reliability profiles.
 Understanding the impact of storage capacity on reliability is

important for accurately evaluating and comparing the performance

of different rainwater harvesting systems.
 An analysis of reliability was performed for the following

predetermined tank sizes:

0.5 m3 tank storage capacity
2 m3 tank storage capacity
4 m3 tank storage capacity
6 m3 tank storage capacity
8 m3 tank storage capacity
10 m3 tank storage capacity
Persons Per Household and Water Demand
 For performance analysis of rainwater harvesting systems, the study
considers an average household size of 4 persons.
 The study uses the assumed 4-person household size to calculate
both daily and monthly water demand.
 The World Health Organization (WHO) classifies access to water
into 4 different levels.
 WHO provides distinct volume recommendations for each access
level (WHO, 2020).
 WHO guidelines serve as a reference for determining the appropriate
water demand levels corresponding to different access categories.
 Utilizing WHO guidelines ensures that the study's analys is aligns
with international standards and recommendations.
 The recommended water volumes from WHO play a crucial role in
evaluating the effectiveness of rainwater harvesting systems in
meeting the diverse water needs across different access levels.
Fig;Summary of Water Access and Adequacy Recommendation by WHO table
 Runoff Coefficient
 Runoff coefficient represents the percentage of rainfall that runs off a
surface during precipitation.
 Runoff coefficient is influenced by the type of surface material.
 Different materials exhibit varying degrees of permeability and
water absorption.
 Various factors, including surface roughness, land use, and
vegetation cover, contribute to the determination of the runoff
coefficient.
 Understanding runoff coefficients is crucial in hydrology for
predicting surface runoff and managing water resources effectively.
 Engineers and urban planners use runoff coefficients to design
drainage systems and mitigate the impact of rainfall on surfaces.
 Managing runoff coefficients is essential for minimizing soil erosion,
preventing flooding, and preserving water quality in natural
ecosystems.
Runoff coefficient of various materaials
Roof Area Calculation

 The storage capacity of a rainwater harvesting reservoir relies, among

other factors, on the total roof area available for capturing rainfall.
 Roof areas for the study were obtained by averaging measurements
from 50 randomly selected, spatially distributed sample buildings in
each district.
 Google Earth Engine serves as the platform for acquiring roof area
data.
 Google Earth Engine provides an accessible and user-friendly
interface, simplifying remote sensing computations.
 Measurements obtained via Google Earth are acknowledged as
approximations.
 Cha and Park (2007) suggest that, despite limitations, Google Earth
can substitute traditional site data when accessibility is constrained.
 Researchers, including Thenkabail (2007) and Zhang et al. (2008),
have utilized Google Earth imagery in their studies, indicating its
validity and utility.
 Images from Google Earth were used to extract roof areas for 50
buildings per district.
 The average roof area for each district was calculated by taking the
mean of the 50 samples
 The average roof area obtained through this process is a key
parameter in analyzing the performance of rainwater harvesting
systems across different districts in the study.