Very important and interesting materials

Use it

1. Wolkite University College of Engineering and Technology Hydraulic and Water Resources Engineering Department   Title: Quantifying Lake Dynamics: Detecting Changes in Water Level and Surface Area Using Remote Sensing Techniques, Case Study of Arekit Lake, Gurage Zone Thematic Area: Theme 5: National Resources Management and Biodiversity Conservation Sub-theme 2: National parks, Endogenous Plant and Animals, and Water Bodies Sub-theme 3: Environment, Bioremediation Sub-theme 4: Water R esources Thematic Research Proposal By: Dr.Bedassa Dessalegn (PhD) Mr.Girma Bekele (MSc)   November, 2023 Wolkite , Ethiopia

2. Outlines Introduction Background Research Objective Statement of Problem Methodology Expected Output Work Plan Budget Summary 2

3. Introduction R emote sensing (RS) approaches are crucial for detecting long-term lake dynamics. RS synthetic aperture radar (SAR) backscatter from lakes exhibits annual variation due to climate variations . W hen the lake is not frozen, the challenge in estimating lake surface area (LSA) is the variability in backscatter caused by lake surface roughness due to wind-induced. In the freeze-up period, the lake surface ice causes a challenge to distinguish water and land, since SAR backscatter from ice can exhibit very similar intensities to that from the surrounding land. 3

4. Cont... A dditional factor for monitoring lakes with SAR during; Winter season (lake frozen, or snow cover) can results greater radar backscatter Spring (snowmelt or wet snow) makes it challenging to distinguish the lake edge since SAR backscatter from wet snow on land and lake are very low, thus giving poor contrast between the surface types. A few studies have attempted on detection of LSA during this challenging season , and interpretation of SAR requires a greater degree of care to extract accurate information. 4

5. Cont... The study will involves the following: Uses SAR images RS to detect lake w ater l evel (LWL) and LSA Consider large SAR dataset in terms of; N umber of images and period of time covered, C onsistency of the method over many seasons of data and across a range of sensor resolutions Develop processing chain of SAR image that used on any lake as well as capable of recognizing the LSA of Arekit Lake. Develop the relationship between in-situ LWL and estimated LSA Evaluate the accuracy of SAR for LSA and LWL estimates. 5

6. Background Quantifying changes in LWL and LSA are important for: U nderstanding hydrological processes Assessing water resources Designing water resource projects A ssessing water resource potential, and Monitoring environmental changes RS approaches are emerged as powerful tools: For detecting and monitoring lake dynamics over a large spatial and temporal extents. To provide valuable data on LWL and LSA, which helps decision-makers. 6

7. Cont... R S approaches are more advantages than ground-based monitoring methods. It offer repeatable measurements , and assess temporal variations in lake dynamics RS provides a synoptic view (simultaneous observation of multiple lakes across vast areas). This broad coverage enables the identification of regional patterns and trends, and Provide a comprehensive understanding of hydrological dynamics. 7

8. Cont... Some of RS techniques used for detecting lake dynamics are; Synthetic aperture radar (SAR ) O ptical remote sensing Light detection and ranging ( LiDAR ) T hermal remote sensing S atellite-based altimetry, and M ultispectral/ hyperspectral imaging 8

9. Cont ... SAR provides high-resolution imagery that capture detailed information of LWL and LSA Optical remote sensing enables the analysis of water reflectance and shoreline changes LiDAR offers accurate elevation models to assess water surface variations Thermal helps identify temperature-related changes in LWL Altimeters measure the distance between the satellite and water's surface using laser, which allows to measure the LWL Multispectral and hyperspectral imaging provide insights into water quality parameters 9

10. Statement of Problem Understanding lake's characteristics, utilization , and potential impacts are important; T o develop effective strategies for sustainable management and policy-makers. L ack of quantitative scientific information on Arekit Lake dynamics poses a significant problem on its; U tilization, management, decision-making processes, etc. Filling this gap is paramount to ensure; U tilization and conservation of Lake Arekit , Ultimate safeguard of its ecological integrity and the well-being of the surrounding community. 10

11. Research Objectives General objective Detecting lake dynamics through quantifying lake water level and surface area changes using remote sensing techniques. Specific objectives Assessing in-situ measured lake dynamics like water level Identifying and analyzing drivers and causes of lake level and surface area changes Exploring long-term trends and variations in lake water levels and surface area 11

12. Methodology 12 Data SAR remote sensing approach SAR remote sensing ( Envisat ASAR, RADARSAT-2, and Sentinel-1A and B) is considered Due to ice cover and wind, other SAR datasets lead to poor contrast between the lake and the surrounding land and erroneous classifications. In-situ lake water level (LWL) Helps to derive regression equation between LSA and the in-situ LWL.

13. Cont... Digital elevation model (DEM) Terrain correction is applied to the SAR images using DEM Shuttle Radar Topography Mission (SRTM) version type of DEM is used Climate data Response on soil degradation, expansion of agriculture and grazing land, etc., at Arekit Lake Software Generic Synthetic Aperture Radar (GSAR ) ArcGIS 13

14. Cont... Data processing Steps of geocoding SAR images using GSAR software is as follows: Calibration using an annotated lookup table from G round R ange D etected (GRD) high r esolution product Multi-looking and pixels averaging to suppress speckle Mapping the selected Universal Transverse Mercator (UTM)-projected output grid to the radar coordinates using DEM Projection of the backscatter product to the output grid using the radar coordinate mapping and cubic interpolation Geocoding generates mask files for radar shadow and layover Small portions of the lake are affected by layover/shadow. Imaging geometries in the multi-temporal post-processing avoid this problem. 14

15. Cont... A flow diagram of processing of SAR from geocoding images to final temporal LSA L1B image archive Geocoded VV, VH , radar mask images Classified images Corrected images Dataset reclassified and labeled-unfiltered LSA time series Extracting LSA time series LWL time series K-means clustering Layover/shadow correction Minimum/maximum masking area Regression equation Outlier filtering Calculating accuracy assessment Start End 15

16. Cont... M aximum area water mask is applied to each images; T o reduce the image area to be segmented and reduce computational time. M orphological dilation filter is considered to the mask; T o increase the proportion of surrounding land area T o ensure as enough water and non-water pixels is confined within the lake mask. After the implementation of masking; K-means clustering is applied to segment the unmasked part of the image. Once the entire dataset are segmented; P ost-processing and correcting the segmented images is the next step. M ulti-temporal method is used to correct the pixels which is; U nclassified due to areas of radar shadow/layover or incomplete coverage 16

17. Cont... M inimum area mask is applied to each image; To correct misclassifications due to wind or snow/ice at the lake surface . Correcting these effects increases the use of a greater proportion of the dataset instead of discarding poor classifications . This mask area is determined through mapping all pixels that are classified as water that mostly present in all June images . The area continually updated to take into account that the minimum area in June during one year is not a representative for other years This corrects misclassified pixels occurring within the center of the lake due to wind or other factors that increase surface roughness and/or backscatter. 17

18. Cont... 18 Minimum water mask area is utilized for summer and winter images For winter classifications pixels classified into class 1 or class 2 (mixed backscatter), R eset all class 2 pixels to class 1 if they are contained within the minimum mask; if class 2 pixels is outside of this mask belong to the land class . The d ifference between summer and winter images is that; P ixels with class 3 occurring within the minimum area always remain as class 3 in winter images, whereas in summer images always be reclassified to 1 . After these corrections, a detected water mask is constructed by; C hoosing only the lowest class and median filter to remove any remaining areas of speckle . The final water mask is used to calculate the LSA.

19. Expected Output Quantitative lake dynamics on LWL and LSA Long-term temporal trends of lake dynamics Relationship between in-situ LWL and LSA of SAR remote sensing Evaluation for the accuracy of SAR remote sensing Summary of key insights and implications of the quantified lake dynamics Suggestion on management strategies, monitoring approaches Further research needed to address the identified changes and support sustainable lake management. 19

20. Work Plan S.N     Job breakdown Tentative time frame Nov-Jan Feb-Apr Jun-Jul 1 Literature reviewing       2 Remote sensing (RS) data analysis       3 In-situ data collection and analysis       4 Processing RS image       5 Extracting LSA       6 Analyzing drivers of lake dynamics       7 Quantifying LWL       8 Developing the relationship of LSA with LWL and validating the results     9 Organizing the all research document and submitting       20

21. Budget Summary S. No Main Activity Description of activity Participant Cost (ETB) 1 Data collection       Field data collection of lake water level and surface area Bench mark determination and pointing measurement coordinates Researchers 350000 In-situ measurement Interview and team discussion with selected stakeholders and elders. The general condition of station Secondary data collection Historical data collection Researchers 75000 2 Data quality assessment Evaluating data quality Researchers 35000 3 Spatial data analysis Spatial model purchase and simulation Researchers 25000 4 Office work (model simulation) and verification Integrating spatial model output with WMO standards Researchers 15000 5 Analysis and documentation Validating model output and optimizing model output considering climate and financial factors Researchers 12000 Total     500000 21

22. Thank You !!!!