รายการเอกสารองค์ความรู้ฉบับสมบูรณ์

สามารถจ่ายก�ำลังไฟไดต้ อ่ เนือ่ งไมต่ ่ำ� กวา่ 5 kVA ใน
การใช้งานปกติ โดยใช้ไฟ 1 เฟส

4. ข้อมูลและกรรมวิธีวิเคราะห์ความชัน

4.1 อุปกรณ์และข้อมูล

รูปที่ 5 รถบรรทุกดัดแปลงสภาพพร้อมคอมพิวเตอร์ ประกอบด้วยคอมพิวเตอร์ระบบปฏิบัติการ
Windows 10 64-bit Core i7 RAM 8 GB
เครื่องคอมพิวเตอร์ปฏิบัติงาน (Workstation) ซอฟต์แวร์ระบบสารสนเทศทางภูมิศาสตร์ ArcMap
จ�ำนวน 1 หน่วย ซ่ึงมีหน่วยประมวลผลกลาง 4 แกน 10.5 และข้อมูลสารสนเทศภูมิศาสตร์ ได้แก่ แบบ
หลกั มีสญั ญาณนาฬิกา 2.7 GHz หน่วยประมวลผล จ�ำลองความสูงเชิงเลขและข้อมูลถนน แสดงดัง
กลางมีความจ�ำแบบ Cache Memory รวมในระดับ ตารางที่ 1
เดียวกันขนาด 8 MB มีหน่วยความจ�ำหลัก (RAM)
ชนิด DDR4 ขนาด 64 GB มีหน่วยจัดเก็บข้อมูล ตารางท่ี 1 รายละเอียดข้อมูลท่ีใช้ในการศึกษา
Solid State Drive 256 GB
ข้อมูล รายละเอียด แหล่งท่ีมา
เคร่ืองคอมพิวเตอร์ปฏิบัติงานแบบ Rack
Mountable จ�ำนวน 2 เครอื่ ง ซง่ึ มหี นว่ ยประมวลผล DEM (.tiff) ดาวเทียม ALOS Alaska Satellite
กลาง 8 แกนหลกั มีสัญญาณนาฬกิ า 1.7 GHz หน่วย ระบบ PALSAR Facility, Geophysical
ประมวลผลกลางมีความจ�ำแบบ Cache Memory ค ว า ม ล ะ เ อี ย ด Institute, University
รวมในระดับเดียวกันขนาด 10 MB มีหน่วยความ ภาพ 12.5 เมตร of Alaska Fairbanks
จ�ำหลัก (RAM) ชนิด DDR4 ขนาด 16 GB มีหน่วย ปี 2008 (assigned by NASA),
จัดเก็บข้อมูล Solid State Drive 256 GB มีหน่วย USA.
ประมวลผลเพ่ือแสดงภาพแยกจากแผงวงจรหลักที่
มีหน่วยความจ�ำขนาด 4 GB ชนิด DDR5 ถนน เส้นทางทใี่ ชใ้ นการ Google map/
(.kml) ศึกษา Google earth
อุปกรณ์กระจายสัญญาณ (L2 Switch)
จ�ำนวน 1 เครื่อง เคร่ืองจ่ายไฟฟ้าส�ำรอง (UPS) กรอบแนวความคิดของการวิจัยคร้ังน้ี คือ
จ�ำนวน 5 เคร่ือง ตู้จัดเก็บคอมพิวเตอร์และอุปกรณ์ การน�ำระบบสารสนเทศทางภูมิศาสตร์มาใช้ในการ
อปุ กรณค์ น้ หาเสน้ ทางเครอื ขา่ ยแบบไรส้ าย (Wireless ศึกษาหลักการวิเคราะห์เชิงพ้ืนท่ีในการค�ำนวณ
Router 3G/4G/LTE) จ�ำนวน 1 ชุด เครื่องปั่นไฟ ความลาดชันของถนนที่ใช้เดินรถบรรทุก 6 ล้อ
ดีเซล จ�ำนวน 1 เครื่อง เป็นชุดเครื่องก�ำเนิดไฟฟ้า ดัดแปลงสภาพ แสดงดังรูปที่ 6 โดยการน�ำชั้นข้อมูล
DEM มาวิเคราะห์หาความชัน และชั้นข้อมูลถนน
มาสร้างพื้นท่ีกันชน จากน้ันจึงค�ำนวณหาความชัน
ของถนน ในการนี้ก�ำหนดความชันของถนนท่ี

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 37

ไม่เหมาะสมในการเดินรถตามรายะเอียดของ โดยใช้ซอฟต์แวร์ ArcGIS เครื่องมือ Mosaic
รถบรรทุก 6 ล้อ ดัดแปลงสภาพ ท่ีมีความสามารถ to new raster ทั้งน้ี ภายหลังการต่อภาพต้องตรวจ
ข้ึนทางชันสูงสุด 21.61 องศา สอบให้จุดภาพที่เช่ือมต่อกันเป็นจุดเดียวกันของท้ัง
สองภาพข้างเคียงกัน เพื่อหลีกเลี่ยงความผิดพลาด
สะสมท่ีจะเกิดข้ึนในการค�ำนวณความชัน เนื่องจาก
ข้อมูล DEM ท่ีน�ำมาใช้พิสูจน์หลักการมีรายละเอียด
ของภาพเดิมหยาบอยู่แล้ว คือ 12.5 เมตร เม่ือเทียบ
กับความยาวรถ 8.14 เมตร

รปู ที่ 6 กรอบแนวคิดในการเสนอหลักการ น�ำข้อมูลถนนท่ีได้จาก Google My Maps
เข้าสู่ซอฟต์แวร์ระบบสารสนเทศทางภูมิศาสตร์ เพ่ือ
แปลงรูปแบบข้อมูล Keyhole Markup Language
(.kml) เป็น Shapefile (.shp) แสดงดังรูปที่ 8 โดย
ใช้เคร่ืองมือ Conversion KML to Layer

4.2 การเตรียมข้อมูล

น�ำข้อมูล DEM ท่ีได้จากดาวเทียม ALOS เข้าสู่ รปู ที่ 8 ข้อมูลถนนจาก Google maps©
ซอฟต์แวร์ GIS เพื่อท�ำการต่อ (Mosaic) ข้อมูลภาพ
แต่ละภาพ (Scene) ให้เป็นชุดข้อมูลแผนท่ี (Map)
ครอบคลุมพ้ืนท่ีถนนช่วงท่ีกล่าวไว้ในรูปท่ี 1 ขวา
ผลของการต่อภาพแสดงดังรูปที่ 7

4.3 การวิเคราะห์ข้อมูล

รูปที่ 7 การรวมข้อมูลภาพเข้าด้วยกันเป็นชุดข้อมูล น�ำข้อมูล DEM ท่ีได้จากการข้อ 4.2 มาสร้าง
ข้อมูลความชัน (Slope) เพื่อใช้ในการวิเคราะห์พ้ืน
ผิว (Surface Analysis) ของค่าตัวแปรความสูง (Z)
ในพ้ืนท่ีศึกษา จากนั้นข้อมูลถนน (.shp) จะถูกน�ำ
มาสร้างพ้ืนท่ีกันชน (Buffer) ระยะ 20 เมตร เพื่อ
ใช้เป็นตัวแทนความกว้างของถนน เน่ืองจากเป็น
ถนนทางหลวง 3 ช่องจราจรทั้งสองด้าน ในขณะ

38 วารสารวชิ าการเทคโนโลยปี ้องกันประเทศ ปีท่ี 2 ฉบบั ท่ี 6 กันยายน - ธันวาคม 2563

เดียวกันท�ำการแปลงข้อมลู ถนนจากเดิมท่เี ปน็ ขอ้ มูล
แบบเส้น (Line) ให้เป็นข้อมูลแบบจุด (Point) เพื่อ
ใช้เป็นตัวแทนในการแสดงค่าความชัน โดยก�ำหนด
ให้แสดงค่าความชันทุก ๆ 8.14 เมตร ตลอดทั้ง
เสน้ ทาง ซง่ึ เทา่ กบั ขนาดความยาวของรถบรรทกุ 6 ลอ้
ดัดแปลงสภาพ ท้ังน้ีส�ำหรับการศึกษาอ่ืน ๆ ควร
ก�ำหนดระยะห่างจุดความชันตามขนาดของรถยนต์
หรือข้อมูลท่ีใช้ในการศึกษา

น�ำข้อมูลผลการสรา้ งพ้ืนท่ีกนั ชน (Buffer) ของ รูปที่ 9 ความชันของพื้นที่ทั้งหมด
ถนนท่ีสร้างขึ้นมาท�ำการตัด (Clip) ข้อมูลความชัน
(Slope) ในบริเวณที่มีการซ้อนทับกัน เพ่ือให้สะดวก 5.2 การวิเคราะห์ข้อมูลถนน (.shp)
ต่อการวิเคราะห์และสังเกตค่าความชัน โดยค่าความ ใช้เครื่องมือ Conversion KML to Layer ใน
ชันที่ได้จากข้อมูล Slope เป็นข้อมูลแรสเตอร์ซ่ึงไม่
สะดวกในการเลือก (Selection) เม่ือเทียบกับข้อมูล การแปลงรูปแบบข้อมูลถนนจาก .kml เป็น .shp
เวกเตอร์ จึงท�ำการปรับปรุง (Manipulate) ข้อมูล เพื่อน�ำไปใช้ในการสร้าง Buffer โดยก�ำหนดพ้ืนท่ี
ความชันให้แก่ข้อมูลถนนแบบจุด โดยใช้เคร่ืองมือ ห่างจากเส้นถนนดา้ นละ 10 เมตร แสดงดงั รูปที่ 10
Add Surface Information จากนั้นน�ำข้อมูล Buffer ของถนนมาตัด (Clip)
5. ผลการวิเคราะห์ ข้อมูล Slope ในบริเวณที่มีการซ้อนทับกัน เพื่อให้
5.1 การวิเคราะห์ข้อมูล DEM เหลือเฉพาะพ้ืนท่ีศึกษาบริเวณถนนบนเนินเขาระยะ
ทาง 31.4 กิโลเมตรเท่านั้น จากการศึกษาพบว่า
น�ำข้อมูล DEM ความละเอียด 12.5 เมตร ค่าความชันในบริเวณ Buffer ของถนนอยู่ระหว่าง
มาต่อเข้าด้วยกันให้เป็นชุดข้อมูลแผนท่ี โดยแต่ละ 0-41.79 องศา แสดงดังรูปท่ี 11
ภาพจะต้องมีจ�ำนวนแบนด์และจ�ำนวนบิตเท่ากัน
ภายหลังจากการเชื่อมต่อภาพเข้าด้วยกัน ข้อมูล
DEM จะถูกน�ำมาใช้ในการสร้างข้อมูลความชัน โดย
ค�ำนวณการเปล่ียนแปลงค่าความสูงระหว่างเซลล์
หลักและเซลล์ใกล้เคียงท่ัวพื้นที่ เพื่อให้ทราบถึง
ความชัน (องศา) ในพื้นที่ จากการศึกษาพบว่าค่า
ความชันของพ้ืนที่ท้ังหมดอยู่ระหว่าง 0-82.7 องศา
แสดงดังรูปท่ี 9

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 39

รปู ที่ 10 ระยะการสร้าง Buffer และ 21.61-41.21 แสดงดังรูปที่ 14 (ทั้งน้ีในพ้ืนที่
การศึกษาอื่น ๆ ควรพิจารณาวิธี Reclassify ให้
เหมาะสมกับความถ่ีและการกระจายตัวของข้อมูล
เช่น กรณีเป็นพื้นที่ราบไม่มีความชันควรเลือกใช้
การ Reclassify แบบ Equal Interval คือ แบ่งชั้น
ข้อมูลให้มีความกว้างเท่ากัน) โดยให้ความส�ำคัญค่า
ความชันในช่วง 21.61-41.21 องศา เน่ืองจากขีด
ความสามารถขน้ึ ทางชนั ของรถเดมิ อยทู่ ี่ 21.61 องศา
ตามที่กล่าวไว้ในหัวข้อ 4.1 การไต่ข้ึนลงทางชัน
ภายหลงั การดดั แปลงสภาพจงึ ไมค่ วรเกนิ 21.61 องศา
เพ่ือลดอุปสรรคในการเดินรถและป้องกันอันตรายที่
อาจเกิดข้ึนขณะขับข้ึนลงเขา

รปู ที่ 11 ความชันบริเวณ Buffer ของถนน

5.3 การก�ำหนดค่าความชันให้แก่ข้อมูลถนน

ใช้เครื่องมือ Generate Point สร้างข้อมูลจุด รูปที่ 12 การกำ�หนดช่วงค่าความชัน
โดยอ้างอิงจากข้อมูลถนน ก�ำหนดระยะห่างระหว่าง รปู ท่ี 13 การเพม่ิ ขอ้ มลู คณุ ลกั ษณะของความชนั ใหก้ บั ชน้ั ขอ้ มลู ถนน
จุด 8.14 เมตร เพื่อใช้ในการรองรับค่าความชันใน
บริเวณที่ซ้อนทับกัน แสดงดังรูปที่ 12 จากการศึกษา
พบว่ามีข้อมูลจุดที่ถูกสร้างข้ึนจ�ำนวน 11,348 จุด
จากนั้นท�ำการเพิ่มค่าความชันบริเวณท่ีซ้อนกันลง
สู่ข้อมูลถนนแบบจุดโดยใช้เคร่ืองมือ Add Surface
Information แสดงดังรูปที่ 13 และเน่ืองจาก
ข้อมูลจุดความชันมีการกระจายเป็นช่วงกว้างและ
มีจ�ำนวนมาก จึงท�ำการจ�ำแนกจุดข้อมูลความชัน
แบบ Natural Breaks (Jenks) ซึ่งเป็นการแบ่งตาม
จ�ำนวนค่าของ Record ที่มีในแต่ละช่วงข้อมูลออก
เปน็ 4 ระดบั คอื 0.00-4.83, 4.84-11.26, 11.27-21.60

40 วารสารวชิ าการเทคโนโลยีปอ้ งกนั ประเทศ ปีท่ี 2 ฉบับที่ 6 กนั ยายน - ธันวาคม 2563

จากการวเิ คราะหค์ วามลาดชนั ของถนน พบวา่ 6. สรุปและข้อเสนอแนะ
เส้นทางที่ใช้ในการเดินทางมีพื้นท่ีท่ีมีค่าความชัน 6.1 สรุปผลการน�ำเสนอหลักการ
เกินกว่าขีดความสามารถของรถบรรทุก 6 ล้อ
ดัดแปลงสภาพ จะสามารถเคล่ือนผ่านได้ตั้งแต่ 6.1.1 จากผลการวิเคราะห์ข้อมูล DEM
21.61 องศา ข้ึนไป ช่วงระหว่าง จ.อุตรดิตถ์ และ รายละเอียดจุดภาพ 12.5 เมตร ในหัวข้อ 5 ท�ำให้
จ.แพร่ ตามทางหลวงหมายเลข 11 แสดงดังรูป ทราบว่าหลักการวิเคราะห์เชิงพ้ืนท่ีด้วยการค�ำนวณ
ที่ 14 เร่ิมต้ังแต่ต�ำแหน่งแรกสุดของรูปที่ 14 ท่ี ความชันของถนนจากเครื่องมือ GIS สามารถน�ำไป
พิกัดจุดเร่ิมต้นละติจูด 17.824818 องศาเหนือ ประเมินอุปสรรคในการเดินรถและป้องกันอันตราย
ลองจิจูด 100.072276 องศาตะวันออก และจุด ที่อาจเกิดข้ึนขณะขับข้ึนลงเขา ในช่วงระยะทาง
สุดท้ายละติจูด 17.834883 องศาเหนือ ลองจิจูด 2.09 กิโลเมตร ที่มีความชัน 21.61 องศา ข้ึนไป และ
100.060016 องศาตะวันออก เป็นระยะทางทั้งสิ้น เกินขีดความสามารถในการใช้เดินรถบรรทุก 6 ล้อ
2.09 กิโลเมตร ดัดแปลงสภาพ ดังนั้น จึงควรหลีกเล่ียงเส้นทางใน
ช่วงดังกล่าว และเปล่ียนไปใช้เส้นทางอ่ืนท่ีมีความ
ชันไม่เกิน 21.61 องศา

รูปที่ 14 ความชันถนนที่เป็นอุปสรรคและอันตราย 6.1.2 จะเห็นได้ว่าหลักการวิเคราะห์ความ
รูปที่ 15 แบบจำ�ลองรถวางบนความชันถนน ลาดชันของถนนส�ำหรับการเคล่ือนย้ายรถบรรทุก
6 ล้อ ดัดแปลงสภาพ สามารถพิสูจน์โดยใช้ข้อมูล
DEM ที่รายละเอียด 12.5 เมตร ที่มีความละเอียดต�่ำ
ในปัจจุบัน ซ่ึงแสดงให้เห็นถึงอุปสรรคและอันตราย
ที่อาจเกิดขึ้นจริงหากต้องใช้เส้นทางดังกล่าว
ดังแสดงในรูปท่ี 15 โดยน�ำภาพรถขนาดมาตราส่วน
ถกู ตอ้ งมาใชป้ ระกอบในภาพ ทงั้ นี้ 1 ชอ่ งในต�ำแหนง่ ที่
เลอื ก (Object Identifier: OID) มคี วามยาว 8.14 เมตร
ท้ัง 7 จุด จากต�ำแหน่งที่เลือก OID มีความชันเกิน
37 องศา และมีค่า 41.2 องศา ที่จุดสูงสุดของเส้น
ทาง OID ที่ 3 และทงั้ หมดมคี วามชนั เกนิ 21.61 องศา
ซึ่งเกินขีดความสามารถในการขึ้นลงทางชันของรถ
บรรทุก 6 ล้อ ดัดแปลงสภาพ

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 41

6.2 ข้อเสนอแนะ 7. กิตติกรรมประกาศ

การศกึ ษาครง้ั นกี้ �ำหนดขอบเขตในการเลอื กชดุ ผู้เขียนขอขอบคุณคณะนักวิจัยโครงการ
ข้อมูลเฉพาะท่ีครอบคลุมเส้นทางที่ข้ึนลงเขาสูงชัน ประยุกต์ใช้แผนที่สถานการณ์ร่วมเพ่ือจ�ำลอง
ระยะทางท้งั ส้นิ 31.4 กโิ ลเมตร ระหว่าง จ.อุตรดิตถ์ ภารกิจช่วยเหลือทางทหารในสถานการณ์ฉุกเฉิน
และ จ.แพร่ เท่านั้น ซึ่งตามเส้นทางจริงจะมีปรากฏ ในการสนับสนุนข้อมูลเพื่อประกอบการวิเคราะห์
ความชันในชว่ ง จ.แพร่ เข้า จ.น่าน ซึง่ อยนู่ อกเหนอื 8. เอกสารอ้างอิง
ขอบเขตของบทความฉบับนี้ และเป็นอีกแนวทาง
ในการศึกษาเส้นทางหน่ึงก่อนการปฏิบัติภารกิจจริง [1] พระราชกฤษฎีกา แบ่งส่วนราชการและ
อีกท้ังหลักการน้ีสามารถน�ำไปวิเคราะห์หาเส้นทาง ก�ำหนดหน้าท่ีของส่วนราชการ กองบัญชาการ
ซ่ึงเป็นทางเลือกอื่นส�ำหรับการส่งมอบรถบรรทุก กองทพั ไทย กองทพั ไทย กระทรวงกลาโหม พ.ศ. 2552.
6 ล้อ ดัดแปลงสภาพ ของสถาบันเทคโนโลยีปอ้ งกนั ราชกิจจานุเบกษา. จ�ำนวน 6 น.
ประเทศ ทั้งน้ี ตอ้ งค�ำนึงถงึ ความถูกตอ้ งของขอ้ มูลท่ี
จะน�ำมาใช้ในการวเิ คราะห์ดว้ ย [2] ช�ำนาญ ขุมทรัพย์. 2561. แนวคิดระบบ
อ�ำนวยการปฏิบัติแบบเคลื่อนท่ีเพ่ือการบรรเทา
ข้อจ�ำกัดของรายละเอียดความถูกต้องของ ภัยพิบัติและสาธารณภัย วารสารสถาบันวิชาการ
ข้อมูลท่ีเลือกใช้คือ จุดภาพของ DEM ขนาด 12.5 ปอ้ งกันประเทศ. ปีที่ 9 ฉบับท่ี 1. น. 7 – 19.
เมตร ซึ่งมีความถูกต้องในแนวด่ิงน้อยอยู่แล้ว และ
เม่ือต้องท�ำกระบวนการปรับปรุง (Manipulation) [3] หนังสือหน่วยบัญชาการทหารพัฒนาด่วน
เพื่อให้ได้จุดความสูงระหว่าง 8.14 เมตร ตามขนาด ท่ีสุด ท่ี กห 0309/2568. ลง 11 กันยายน 2562.
ความยาวรถท่ี 8.14 เมตร จะเป็นการส่งต่อความ เร่ืองขอรับการสนับเครื่องมือโครงการประยุกต์ใช้
หยาบของข้อมูลขณะท�ำการวิเคราะห์ไปสู่ความ แผนที่สถานการณ์ร่วมเพื่อจ�ำลองภารกิจช่วยเหลือ
ชันท่ีได้ อย่างไรก็ตาม หลักการของการจ�ำลอง ทางทหารในสถานการณ์ฉุกเฉิน. จ�ำนวน 1 น.
ภูมิประเทศยิ่งได้ขนาดจุดภาพเล็กลงยิ่งเพ่ิมความ
ถูกต้องในแนวดิ่งยิ่งขึ้น ดังนั้น บทความฉบับนี้ [4] หนังสือสถาบันเทคโนโลยีป้องกันประเทศ
ต้องการชี้ให้เห็นความส�ำคัญของการเลือกเส้นทาง ท่ี สทป 5800/612. ลง 15 พฤษภาคม 2563. เร่ือง
เดินรถและอุปสรรคที่อาจเกิดข้ึนเมื่อต้องเดินรถแล้ว ตอบรับให้การสนับสนุนเคร่ืองมือโครงการประยุกต์
ยังเสนอแนะให้ใช้ข้อมูล DEM ในรายละเอียดภาพ ใช้แผนท่ีสถานการณ์ร่วมเพ่ือจ�ำลองภารกิจช่วย
ที่เล็กลง เพ่ือเพ่ิมความส�ำคัญในการน�ำ GIS ไปใช้ เหลือทางทหารในสถานการณ์ฉุกเฉิน. จ�ำนวน 1 น.
เป็นเคร่ืองมือช่วยตัดสินใจ เนื่องจากการใช้ข้อมูล
ที่มีความถูกต้องยิ่งข้ึนจะยิ่งแสดงให้เห็นถึงอุปสรรค [5] ช�ำนาญ ขุมทรัพย์. 2562. การถ่ายทอด
และอันตรายท่ีจะเกิดข้ึนได้ชัดเจนมากกว่าท่ีผู้เขียน เทคโนโลยีของโครงการวิจัยและพัฒนาสู่ภาคการ
ได้น�ำเสนอไว้ในหัวข้อ 6.1.2 ศึกษาและภาคอุตสาหกรรม วารสารสถาบนั วิชาการ
ป้องกันประเทศ. ปีท่ี 10 ฉบับท่ี 2. น. 12 – 25.

[6] สัญญาจ้างปรับปรุงรถควบคุมภาคพ้ืน
เคล่ือนท่ีส�ำหรับการพัฒนาระบบควบคุมและ

42 วารสารวชิ าการเทคโนโลยปี ้องกนั ประเทศ ปีท่ี 2 ฉบบั ท่ี 6 กันยายน - ธนั วาคม 2563

สั่งการในสถานการณฉ์ กุ เฉนิ . 2560. ส�ำหรบั โครงการ Fauzi, Rusnah Muhamad. 2015. Geographic
ประยุกต์ใช้แผนท่ีสถานการณ์ร่วมเพ่ือจ�ำลองภารกิจ Information System (GIS) modeling approach
ชว่ ยเหลอื ทางทหารในสถานการณฉ์ กุ เฉนิ . สญั ญาเลข to determine the fastest delivery routes.
ที่ 62/CTH00080 ลง 29 ธันวาคม 2560. สถาบัน Saudi Journal of Biological Sciences (2016)
เทคโนโลยปี ้องกนั ประเทศ. 10 น. 23, 555 – 564.

[7] ไพศาล จ้ีฟู. 2561. การพัฒนาโปรแกรม [14] Caliskan, E,ediroglu, S., Yildirim, V.
ประยุกต์ส�ำหรับระบบสารสนเทศภูมิศาสตร์บนเว็บ. 2018. Determination forest road routes via
พมิ พค์ รง้ั ที่ 1. กรงุ เทพฯ : ส�ำนกั พมิ พแ์ หง่ จฬุ าลงกรณ์ GIS-based spatial multi-criterion decision
มหาวิทยาลยั , 175 น. methods. Applied Ecology and Environmental
Research. 17(1):759 - 779. DOI: http://dx.doi.
[8] สรรคใ์ จ กล่นิ ดาว. 2542. ระบบสารสนเทศ org/10.15666/aeer/1701_759779.
ภมู ศิ าสตร:์ หลกั การเบอ้ื งตน้ . พมิ พค์ รง้ั ท่ี 2. กรงุ เทพฯ
: โรงพมิ พม์ หาวทิ ยาลยั ธรรมศาสตร.์ 128 น. [15] Emad Basheer Salameh Dawwas.
GIS as a Tool for Route Location and Highway
[9] Paul Bolstad. 2016. GIS Fundamentals: Alignment. Thesis submitted in Partial
A First Text on Geographic Information Fulfillment of the Requirements for the Degree
Systems, Fifth Edition 5th edition. XanEdu of Master in Highway and Transportation
Publishing Inc.: Acton, MA. 770 p. Engineering, Faculty of Graduate Studies,
An-Najah National University, Nablus,
[10] Haiwen Du, A-Xing Zhu, and Yong Palestine. 140 p.
Wang. Spatial prediction of flea index of
transmitting plague based on environmental [16] Sayed Ahmed, Romani Farid Ibrahim,
similarity. Annals of GIS, Volume 26, Issue 3, and Hesham A. Hefny. 2017. GIS-Based Network
pp. 227 - 236. Analysis for the Roads Network of the Greater
Cairo Area. In Proceedings of the International
[11] Sohaib K M Abujayyab and Ismail Conference on Applied Research in Computer
Rakıp Karas. 2020. Landslide Susceptibility Science and Engineering ICAR'17, .Lebanon,
Mapping Using Shallow Neural Networks Model 22 - 06 - 2017. Available online ar published
at Refahiye District in Turkey. Turkish Journal at http://ceur-ws.org.
of Remote Sensing and GIS. Volume 1, Issue
2. pp. 61 - 77. [17] Hino Thailand. 2020. 6 - wheel truck;
FC9JJLA. Online available from: www.
[12] Yufan Zuo , Zhiyuan Liu and Xiao Fu. hinochairatchakarn.com/home/inventory/
2020. Measuring accessibility of bus system hino-500-dominator-fc9jela-fc9jjla-fc9jlla-175h
based on multi-source traffic data. Geo-spatial [Accessed 9 December 2020]
Information Science. Volume 23, Issue 3. pp. 248
- 257. DOI: 10.1080/10095020.2020.1783189

[13] Mohammad Abousaeidi, Rosmadi

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 43

บทความวิจัย

GIS for Vertical Takeoff and Landing Site Selection

Teeranai Srithamarong 1* and Phimraphas Ngamsantivongsa1
Received 22 October 2020: Revised 22 November 2020: Accepted 22 December 2020

Abstract

This research paper addresses the GIS analysis approach to the investigation of suitable
sites for a vertical takeoff and landing drone. The study manipulated GIS and terrain layers
and turned them into proper input before the spatial analysis that included slope, reclassify,
classify and buffer was applied to the individual layers. The output layers were weighted
and multi-criteria analyzed before those patches failing to comply with filtering out criteria
were discarded. Field survey for each suitable candidate site was conducted to cross-check
the proposed approach with the real world. Conclusion was extracted for the VTOL takeoff
and landing sites and discussion was provided with further study being suggested on the
mission simulation of selected takeoff and landing sites.
Keywords : GIS approach, Site selection, VTOL, Takeoff and landing

1 Knowledge and Publication Management Department - TKP, Defence Technology Institute.
* Corresponding author, E-mail: [email protected]

66 วารสารวิชาการเทคโนโลยีปอ้ งกนั ประเทศ ปีท่ี 2 ฉบบั ท่ี 6 กันยายน - ธนั วาคม 2563

1. Introduction System (GIS) technology was used to assess
Surveying for drone launch sites is the criteria requested to define the suitability
of land for housing [2]. The study in [3] was
troublesome when the study areas are under to develop a spatial model for land suitability
constraints of mountainous terrain. Even though assessment for wheat crop integrated with
it is plausible to use a sensor capable of GIS techniques. The proposed model allowed
an aerial survey system to help identify the obtaining results that corresponded with
areas of interest with growing research in the current conditions in the area. The land
machine and computer vision [1], there is also a evaluation procedure has also been applied
question of satellite navigation signal coverage by a GIS – based methodology. Integrating
and a requirement to help researchers to information with crop and soil requirements,
plot on a large scale map or other open the authors in [4] edited and managed land
sources such as Google Earth for flight mission suitability maps for specific purposes by means
planning. It can be seen from Figure 1 that of matching tables. With the final output
in aerial photography over Pua district, Nan aimed at creating military training scenarios
province in the Northern part of Thailand, to be included in a fire-arms training simulator
of an area of more than 300 km2, it is easier of the Royal Thai Army, GIS data was prepared
by exploring the target areas with unmanned and used for the Potential Surface Analysis
aerial surveying system. That will help in (PSA) in form of suitability map that revealed
planning for drone image acquisition more the potential of GIS vector layers that suited
quickly. Those areas will also be locations drug-trafficking routes [5]. However, the
of the ground-based survey to assess the GIS-based approach has been barely applied
accuracy of digital terrain models in the final to the survey of takeoff and landing site
phase of accuracy assessment. Therefore, selection for VTOL drone mapping.
flat and level sites are needed for vertical
takeoff and landing (VTOL) drone in general
for example DJI Phantom Drone.

In this current study, prior to launching Figure 1 The study area with mountainous terrain
a survey drone for VTOL site selection, it
can be more economical to investigate the
terrain nature of potential candidates for
the launching site. Geographical Information

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 67

The results of the study in [6] showed use vector layers and 2020 satellite raster layer.
that using an unmanned aerial system for The road network was manually updated using
topographic mapping and calculating volumes GIS basemap available online. The land use
was more time and cost efficient than land data were of 2016 product whose rural study
surveying, with no loss in accuracy, but only area underwent some urbanization. The 30 m
when performed over bare earth terrain, Landsat 8 imagery was of 2020 acquisition
suggesting that care be taken for the and selected to contain a few cloud-covered
topographic mapping of the densely covered patches. The 1:50,000 topographic map covering
terrain. This current study was expected to the study area was in elevated ranges for an
further extend the cost efficiency of VTOL overall understanding of selected terrain of
drone mapping by proposing a GIS – based the study area (see Figure 2). The Digital Elevation
approach for VTOL takeoff and landing site Model or DEM of 12.5 m was a product of
selection. The selection was performed prior to Advanced Land Observing Satellite (ALOS) in
the still needed field survey. The significance Phased Array type L-band Synthetic Aperture
of the study lied in the commercial mission Radar (PALSAR).
planner that was conducted using available
QGroundControl or Google Earth terrain data Figure 2 Topographic representation of the study area
that was inadequate to meet the standard
required for fine/small scale digital terrain 2.2 Vertical Takeoff and Landing Requirements
model for very precise engineering study [7].
With updated GIS data of the study within
GIS functionality before further spatial
analysis and multi-criteria analysis being applied.
Upon obtaining a suitability map for VTOL
takeoff and landing sites, the field survey was
conducted for every selected site to ensure
proper distribution over the 300 km2 study area.
2. Data Preparation

2.1 Geospatial Data As indicated in [6], the mission path must
be free of obstructions for at least 200 m in
GIS layers included 2016 road and land each horizontal direction. The takeoff and

68 วารสารวิชาการเทคโนโลยีปอ้ งกันประเทศ ปีที่ 2 ฉบบั ท่ี 6 กันยายน - ธนั วาคม 2563

land sites must consist of a level, flat surface to Suitable with Weight 2, Residence to
that is free of obstructions for at least 5 x 5 m Least suitable with Weight 1, and Water
2.3 Data Manipulation and Class Weighting to Unsuitable with Weight 0. Rationale under
this rating was that the Miscellaneous class
DEM was applied with slope creation to contained abandoned and unused areas
create a slope map. From [8], standard slope that were the most suitable for site selection.
descriptors are provided where level to nearly Agriculture and Forest was a Suitable
level at slope of 0 - 2% or at approximate candidate for site selection with subject to
degree of 0 – 1.1 is used as the most suitable field survey. Residential and urban areas
for the selection in Table 1 and the slope were a compromising issue best validated
results in degree of Figure 3 left. on site. Water bodies could cause severe
damage to the drone if unfortunate
Land use map was manipulated as takeoff and landing took place.
shown in Figure 3 middle with reclassify
function to rank Miscellaneous class to Most
suitable with Weight 3, Agriculture and Forest

Table 1 Slope suitability guidance.*

Slope Approximate Terminology Slope suitability Weight
(%) degrees Level - Nearly level
0 - 2 0 - 1.1 Most suitable 3
Suitable 2
2 - 9 1.1 - 5 Very gentle – gentle slope 1
Least suitable 0
9 - 15 5 - 8.5 Moderate slope Unsuitable

>15 >8.5 Strong slope
* Adapted from [8]

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 69

Figure 3 Results of the manipulation
and class weighting

In Figure 3 right, the road network layer Figure 4 NDVI map derived from Landsat 8 imagery
was buffered based on the accessibility of a
grownup man to carry the VTOL drone gear to 2.4 Rasterization and Data Resampling
a launch site and to create 200 m interval from The reclassified Land use and Road
either side of the road based on the 200 m
horizontal clearance requirement as the Most maps were tabulated with suitability and
suitable with Weight 3, from 200 to 300 m weighting columns, the latter of which were
either side of the road center as the Suitable numerical values of the rasterization process.
with Weight 2, from 300 m to 400 m as the The weighting column was for algebraic
Least suitable with Weight 1, and from 400 m operation during the raster overlay step. A
and beyond as the Unsuitable with Weight 0. rasterization process was applied to the
Satellite imagery was analyzed to obtain reclassified Land use and buffer Road vectors.
Normalized Difference Vegetation Index or The final pixel size at 15 m was arrived to
NDVI and categorized into 4 classes with the maintain as much close accuracy as possible
lowest NDVI range from -0.057 to 0.070 as to the 12.5 m DEM resolution. This size was
the Most suitable and Weight 3 based on plausible for the original 30 m Landsat of
the notion that low NDVI values resulted the NDVI map product.
from non- to the less- forest cover of the
studied patch. The NDVI range from 0.071 to The weighted NDVI and Slope raster
0.20 was rated Suitable and Weight 2, from layers were applied by revaluing the pixel
0.21 to 0.33 was rated Least suitable and with the value of the Class Weighting. The
Weight 1, and from 0.34 to 0.45 was rated revalued NDVI and Slope maps were
Unsuitable and Weight 0. The NDVI map was resampled to 15 m pixel size appropriate
shown in Figure 4.

70 วารสารวิชาการเทคโนโลยปี ้องกันประเทศ ปที ี่ 2 ฉบับท่ี 6 กนั ยายน - ธันวาคม 2563

to the final overlay step and matching with more influential factor on site selection than
the previous rasterized layers. Figure 5 the road buffer and NDVI layers because they
illustrates the Slope (far left), Land use (left), involved technical requirements and local safety,
Road (right), and NDVI (far right) raster layers. respectively. The road buffer and NDVI layers
shared equal percentage weight to the
analysis. The suitability map from multi-criteria
analysis was calculated by;

Suitability map (1)
(Slope * 3.5) + (Land use * 3.5) + (Road * 1.5) + (NDVI * 1.5)

=
10

Figure 5 Raster layers for suitability where Suitability map is the multi-criteria
analysis result, Slope is the weighted slope
3. Research Methodology map, Land use is the weighted land use
The proposed research methodology map, Road is the buffered and weighted
road layer map, and NDVI is the weighted
as shown in Figure 6 consisted of 3 steps NDVI map. Field Surveys were conducted
that were related to geospatial analysis and following the calculation results whose
performed mainly down the left side flow of selected areas were visited for observation
methodology. The VTOL mission simulation and photographic evidence.
was discussed for further studies. The data
preparation that involved the manipulation
of geospatial data to an analysis ready format.
The results were further weighted and
multicriteria-analyzed to obtain potential
candidates of launching site in the suitability
map. Practicality, transportation, expenditure
and safety were decisive criteria for the selection
of suitable VTOL takeoff and landing sites.

The weighted layers were ranked Figure 6 The proposed research methodology
according to their significance to the site
selection criteria. The slope suitability and
land use suitability were equally ranked the

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 71

for the survey operation without the need
to visit every patch for photographic and
positional collection.

Figure 7 The weighted overlay map

4. Results and Discussion

Approximately 64.31% of the calculated Figure 8 The weighted Most suitable patches overlay map
result hardly visible on Figure 7 was categorized
as Suitable sites illustrated in orange. The A spatial statistical method was applied
second largest areas were Moderately to determine a selected series of suitable
suitable and accounted for 35.45%. The patches for the field survey where a standard
Most and Least suitable areas shared almost distance was measured from the distribution of
respectively. Where safety to researcher data around the center of all data, see Figure 9.
lives and equipment was concerned, only
the Most suitable areas as shown in scattered There were 29 sites selected for the
blue patches of Figure 8 were adopted as survey and the distribution was within 8.3 km
candidates for takeoff and landing sites. in radius. Time and fuel consumption were
The Most suitable at 0.11% of the calculated much saved from the survey according to the
result hardly visible on Figure 7 was found adopted spatial distribution that yielded only
exclusively on residential areas that had 29 sampled points to perform the survey.
been dictated since the Data Manipulation
and Class Weighting process was embraced. Twenty-nine photos as shown in Figure 10
These areas were further sampled for field were taken with easy access to the locations
surveys. Some illustration of Most suitable
areas in blue of Figure 8 gave an idea of
spatial distribution that could be exploited

72 วารสารวชิ าการเทคโนโลยปี อ้ งกันประเทศ ปีท่ี 2 ฉบับที่ 6 กนั ยายน - ธนั วาคม 2563

resulted from the 200 m buffering data
manipulation. Some of the sites fell within
private properties but were accessible by
vehicle for photography. Together with high
reliability from the Land use layer, the photos
revealed the high suitability for the VTOL
takeoff and landing sites that responded to
the objective of the proposed approach. Of all
the 29 sites, there were 19 perfect sites for the
VTOL takeoff and landing mission, whereas 10
other sites were blended with construction,
water bodies and sparse vegetation considered
dangerous for the mission.

The topographic features found upon the Figure 9 The epicycle representing
survey illustrated in Figure 10 were summarized the center of the patches
in Table 2. There were two discussion points
worth consideration from the topographic Figure 10 Photos taken upon field surveys
features in the table. The Land use layer with
weighting percentage of 35% played a 5. Conclusion and Further study
significant role in some discrepancies between The research that adopted the GIS-based
the adopted approach and the real world. The
survey summary revealed that most of the approach for the VTOL takeoff and landing site
sites had withstood rare changes since 2016, selection had achieved the objective by
the year of land use production. However, the obtaining the flat and level sites. Four GIS and
fact that Pua district was one of Thailand terrain layers included 2016 road and land use
tourism destinations during the winter had vector layers and 2020 satellite raster layer were
undergone Land use changes in most of the manipulated prior to further spatial analysis
rest features with Residential category and
manmade Construction among others.

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 73

Table 2 Topographic features from the survey mission

No. Topographic Feature Survey Position in UTM
(X,Y)

1 Flat area and road in residential area 701823.2932 2127715.090

2 Abandoned and evenly vegetated area 703918.4949 2127259.551

3 Flat and unoccupied area 701830.2831 2127059.955

4 Flat area and paddy field 703893.8665 2126989.708

5 Vegetated and tree covered area 702884.6611 2126807.529

6 Sparse forest and scattered tree area 702799.9381 2126592.710

7 Flat area and paddy field 702066.6224 2126321.802

8 Flat and abandoned with tree and grassland 701892.2623 2126176.064

9 Flat and abandoned area 702379.5967 2124070.401

10 Flat and abandoned area 702330.9545 2122995.157

11 Flat and residential area 702073.3215 2122513.832

12 Flat and unoccupied area 702011.2941 2122382.184

13 Building and construction area 699032.0483 2121925.429

14 Flat area with tree and transmission pole 699145.9556 2121888.493

15 Flat area and road in residential area 703038.0582 2121421.105

and multi-criteria analysis. The Most suit- the 300 km2 study area. There were 29 sites
able areas accounted for 0.11% of the suitable selected for the survey and the distribution
areas. After obtaining the suitability map for was within 8.3 km, 19 sites of which were less
VTOL takeoff and landing sites, the necessary influenced by urbanization. The VTOL nature
field survey was conducted for every selected of drone in general i.e., DJI Phantom Drone
site to ensure proper distribution over can be of use with the results of this study.

74 วารสารวชิ าการเทคโนโลยปี อ้ งกนั ประเทศ ปีท่ี 2 ฉบับท่ี 6 กนั ยายน - ธันวาคม 2563

A simulation of the sites on mission planner model. CATENA, Volume 140, May 2016,
platform of the used VTOL drone is under pp 96-104. https://doi.org/10.1016/j.catena.
investigation during the time of publication of 2015.12.010.
the article.
6. Acknowledgments [4] Gallo, A., Spiandorello, M. and Bin,
C. 2014. GIS – based Methodology for Land
This research article forms part of key Suitability Evaluation in Veneto (NE Italy).
performance indicators of an ongoing project EQA – Environmental quality, 16 (2014) 1 - 7.
titled Applications of Common Operating
Picture for the Simulation of Military Assistance [5] Robert, O. P., Kumsap, C. and Janpengpen,
during Emergency and Communication Blackout A. 2018. Simulation of counter drugs operations
in the Defence Technology Institute, Thailand. based on geospatial technology for use in
a military training simulator. International
The authors appreciated help and valuable Journal of Simulation and Process Modelling.
input from Mobile Development Unit 31 in Nan Vol. 13, No. 4, pp. 402 - 415.
province with regard to activities in the Flooded
Situation Testbed. Funding and support from [6] Fitzpatrick, B. P. 2016. Unmanned
the institute are acknowledged. Aerial Systems for Surveying and Mapping:
7. References Cost Comparison of UAS versus Traditional
Methods of Data Acquisition. A Thesis
[1] Kaawaase, K.S., Chi, F. Shuhong, J. and Presented to the Faculty of the USC Graduate
Ji, Q. B. 2011. A Review on Selected Target School University of Southern California. In
Tracking Algorithms. Information Technology Partial Fulfillment of the Requirements for
Journal, 10: 691 - 702. DOI: 10.3923/itj.2011.691.702. the Degree Master of Science (Geographic
Information Science and Technology). 49p.
[2] Joerin, F., Thériault, M. and Musy, A. 2001.
Using GIS and outranking multicriteria analysis [7] El-Ashmawy, K. L. A. 2016. Investigation
for land-use suitability assessment. International of the Accuracy of Google Earth Elevation
Journal of Geographical Information Science, Data. Artificial Satellites. Vol. 51, No. 3, pp 89 - 97.
Volume 15, 2001 - Issue 2, pp 153 - 174. DOI: https://doi.org/10.1515/arsa-2016-0008.

[3] Baroudy, A. A. E. 2016. Mapping and [8] Barcelona Field Studies Centre. Measuring
evaluating land suitability using a GIS-based Slope Steepness. https://geographyfieldwork.
com/SlopeSteepnessIndex.htm Online access
on 27 April 2020.

Defence Technology Academic Journal, Volume 2 Issue 6, September - December 2020 75

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

THE INTEGRATION OF GEO-INFORMATICS TECHNOLOGY WITH
UNIVERSAL SOIL LOSS EQUATION TO ANALYZE AREAS
PRONE TO SOIL EROSION IN NAN PROVINCE

Preecha Pradabmook1 and Teerawong Laosuwan2

1Defence Technology Institute, Office of the Permanent Secretary of Defence, Nonthaburi, Thailand
2Department of Physics, Faculty of Science, Mahasarakham University, Maha Sarakham, Thailand

E-Mail: [email protected]

ABSTRACT
Soil erosion is a natural disaster which frequently takes place in the Northern region of Thailand. Soil erosion

causes loss of lives and properties of residents. This study was aimed to integrate a geo-informatics technology with the
Universal Soil Loss Equation (USLE) in order to analyze areas which are prone to soil erosion in Nan Province, Thailand.
The operation was performed by analyzing 6 factors of USLE including Rainfall erosivity (R-factor), Soil erodibility (K-
factor), Slope length (L) and slope steepness (S), Cropping management (C), and Conservation practice (P) with overlay
analysis being adopted as the last method. It was found from the analysis that the severity of the soil-erosion prone areas of
Nan Province constituted 5 levels that included the least severity of 2,120.192 km2, the less severity of 2,728.851 km2, the
moderate severity of 2,937.822 km2, the much severity of 2,133.648 km2, and the most severity of 1,551.5584 km2. The
findings from this study can be embraced as a guideline to plan on the conservation and the management of land, and
applied in a decision making process related to the land use planning in Nan Province, Thailand.

Keywords: soil erosion; geo-informatics; universal soil loss equation; remote sensing.

INTRODUCTION annually and the severity is increasing due to more land
Soil erosion is a geological natural phenomenon use which causes more invasion into mountainous areas
and more changes of area conditions (Plakayrungrassamee
caused by the movement of land and rocks along mountain et al., 2011; Pholkerd et al., 2012; Suk-ueng & Chantima,
slope or from a high to low area. There are several 2017). When considering the cause of soil erosion due to
elements or factors in combination that cause soil erosion abnormally heavy rain, it is a natural one which is
and influence the level of severity of soil erosion in a unavoidable. Other factors that cause soil erosion include
particular area (Nearing et al., 2017); it starts with one the crack of land, the slope gradient, the geography, and
factor to be followed by other factors. However, the first land use. On land use, the improvement and correction can
general key factor that causes soil erosion is the quantity be made by refraining from the invading into and
of rain (Vita et al., 1998; Guzzetti et al., 2008), in damaging the forest, and then use the land properly.
combination with other supporting factors such as Therefore, it is necessary to conduct the study in order to
geographic, geological and pedological characteristics. find factors that are causes and to perform assessment to
Those characteristics involve the property of soil and detect the area which is prone to soil erosion, so that the
rocks, the ways the land is used, and the land cover which problem could be further solved correctly. The analysis
could decrease the force of rain before falling onto the into the soil erosion is quite complicated and depends on
land surface and hold up the soil. When the mountain many factors in combination and each factor changes
slope area loses its balance from a heavy rain to the extent constantly (Ganasri & Ramesh, 2016; Conforti &
that makes the soil saturated with water, the physical force Buttafuoco, 2017).
of soil decreases. As a result, the weight of water in soil
increases, thus causing the soil to move down to damage The analysis into the soil erosion is complicated
the lower area (Panagos et al., 2014; Ozsahin et al., 2018; and dependent upon many combined factors, each of
Ozsahin & Eroglu, 2019). Soil erosion happens when the which changes constantly. Consequently, it is difficult to
mountain slope area loses its balance because when there assess soil erosion accurately without extended time of
is a heavy rain to the extent that makes the soil to be study and experiment; for example, in the US, Wischmeier
saturated with water, the physical force of soil decreases and smith (1965) had conducted the study related to the
and the weight of water in soil increases, thus causing the soil loss from 10,000 land plots/year for many decades to
soil to move down that damages the lower area (Zuazo & the extent that it was possible to predict soil loss by using
Pleguezuelo, 2008; Mateos et al., 2017; Cruz et al., 2019). a widely used equation called Universal Soil Loss
Soil erosion is a natural disaster that causes the loss of Equation (USLE). According to the study into related
lives and properties of residents in many countries documents, there were many researchers trying to find the
(Ighodaro et al., 2013; Burt & Weerasinghe, 2014; Belo et soil loss rate due to the washing of rainfall. In 1930, the
al., 2020; Senanayake et al., 2020). In Thailand, especially study and experiment were conducted by taking various
in the Northern region, which consists of steep and high factors that affected soil erosion and concluded as criteria
mountains, where the land use is without conservation of in form of a mathematical model. Subsequently, the
land and water, has been facing the soil erosion problem equation had been developed to assess the soil erosion by

823

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

many researchers that include Cook, (1936); Baver (1933), Figure-1. Nan Province.
Zingg (1940), Smith (1941), Smith and Whitt (1947),
Browning et al. (1947), Musgrave (1947), Van Doren and Data analysis by ULSE
Bartelli (1956), and Smith and Wischmeier (1957). It was On the analysis of area which is prone to soil
just a simple equation and was used by next generations of
land and water conservationists. The equation has been erosion b by using geo-informatics technology in
updated and improved to become USLE by Wischmeier combination with USLE under the Wischmeier method
and Smith (1965). This current study was aimed at (Equation 1) in this study, it is to take various factors that
integrating the geo-informatics technology with the affect soil erosion including R, K, L, S, C, P into
Universal Soil Loss Equation (USLE) in order to analyze consideration together (Figure-2) with procedures and
areas which are prone to soil erosion in Nan Province of methods as follows:
Thailand.
A = R × K × LS × C × P (1)
STUDY AREA
Nan Province (Figure-1) is located at latitude 18° Where;

46' 30'' N and longitude of 18° 46' 44'' E and averaged of A = Average annual soil loss (ton/ha/year)
2,112 meters above mean sea level. Much of the area is R = Rainfall erosivity factor
mountainous lying along the Northern and Southern line. K = Soil erodibility factor
Around the Northern and Eastern sides, it borders with LS = Slope length and slope steepness factor
Lao People’s Democratic Republic. The weather is C = Cropping management factor
tropical grassland with 3 seasons; including summer, P = Conservation practice factor
rainy, and winter, each of which is distinctively different.
Nan Province covers 11,472.076 km2 divided into 5,500.0
km2, forest and mountain or 47.74%, 4,502.37 km2
deteriorated forest or 39.24%, of 1,401.67 km2 agricultural
area or 12.22 %, and 69.64 km2 residential area and others
or 0.60%.

MATERIALS AND METHODS

Data Collection

Primary data
Primary data was collected from the study area. It

included data in general of the study area and data on land
use.

Secondary data
It was requested from government agencies and

reconstructed to a new database. The obtained data from
government agencies included the locational data of
measuring station and rainfall from the Thai
Meteorological Department, 30 m DEM from USGS, soil
of Nan Province, land use, and provincial boundary.

R Factor Analysis
The potential of rain that caused soil erosion was

calculated to R factor analysis by using the rainfall data
during 5 months i.e., May to September, from rainfall
measuring stations in Nan Province and nearby. It was
calculated to find the rainfall on a yearly basis in
millimeters of each station. The data, then, was used to
analyze and plot the graph of mean rainfall by
interpolation with Kriging method. The result was used to
calculate in a mathematical equation. The R factor was
determined from the average rainfall on a yearly basis by
Equation 2.

Y = 0.163X -0.0375 (2)

824

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

Y = Rainfall erosivity factor geo-informatics software package was integral for the
X = Average rainfall result that constructed the plotted graph of average rainfall
Where; from Kriging method interpolation. It was found from the
measured data that the minimum rainfall on a monthly
basis was 117.913 mm, while the maximum one was
224.570 mm, the rainfall mean was 171.242 mm. After
that, data were classified for R factor in 5 levels (Figure-3)
in order to assess the distribution of rainfall in the study
area.

Figure-2. USLE analysis. Figure-3. R factor analysis.

K Factor Analysis It was found from the study that the rainfall at the
K factor analysis is the calculation of possibility very low level (117.913 - 152.630 mm) was in Na Muen
District covering area of around 0.538 km2 equal to
of the soil erosion. In this study, the soil group map at 4.69 %; rainfall at the low level (152.630 - 174.318 mm)
1:50,000 scale from Land Development Department was was found in Na Noi District covering area of around
used as input. The comparison is made with the data upon 0.678 km2 equal to 5.92 %; the rainfall at moderate level
the classification of K factor under the soil group of Nan (174.318 - 192.367 mm) was found in Wiang Sa District
Province and that of K factor obtained from the geological and Ban Luang District covering the area of around
division of the Land Development Department. That 1,976.673 km2 equal to 17.23 %; the rainfall at high level
created the map that showed the factor in regard to the (192.367 -208.262 mm) was found is Mae Charim District,
possibility of soil erosion occurrence. Phu Phiang District, and Mueang Nan District, and Santi
Suk District covering the area of around 5,307.617 km2
L and S Factor Analysis equal to 46.27 %; and the rainfall at very high level
In this study, the calculation of slope length (L) (208.262 - 224.570 mm) was found in Bo Kluea, Pua, Tha
Wang Pha, Chiang Klang, Song Khwae, Thung Chang,
and slope steepness (S) factors were conducted upon 30 Chaloem Phra Kiat Districts covering area of around
meter DEM from USGS. 2,970.931 km2 equal to 25.90 %.

C Factor Analysis Result of K Factor Analysis
It is the calculation of factors concerned with the The Land Development Department collects soil

plant management. In this study, the land use It is the groups under the condition that similar characteristics,
calculation of factor concerning plant management. In this properties, potential of cultivation, as well as similar land
study, the land use map of 2017 at 1:25,000 scale was used management in one same group for the convenience in
where C factor was input. A omparison was made with the examining the characteristics of soil and land use and
factor that concerned plant management of the Land
Development Department. That created the map that
showed the factor of plant management.

P Factor Analysis
It is the calculation of factor concerning soil

conservation. In this study, the land use map at 1:25,000
scale of 2017 was used with P factor as input. A
comparison was made with the factor that concerned the
soil conservation of the Land Development Department.
That created the map that showed the factor of plant
management.

RESULTS AND DISCUSSIONS

Result of R Factor Analysis
The result of the analysis into the erosion was

calculated using the rainfall data on a yearly basis of the
study area and nearby from the rainfall measurement
station of Thai Meteorological Department. The use of

825

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

providing advices with proper land management for was found from the analysis that Nan Province had an L
farmers and those who are interested. More than 300 soil factor of 1% - 6%.
groups were regrouped into 62 soil groups. It was found
from the data that Nan Province lands mostly fall within When the data were divided into 5 levels (Figure-
62nd soil group with slope complex of about 80%. This 5) for the assessment of slope length, it was found that
62nd soil group slope complex was characterized to very slope length at very low level of 1% covered an area of
steep. approximately 618.130 km2 or equal to 5.39%, slope
length at the low level (1% - 2%) covered an area of
approximately 388,543 km2 or equal to 5.42%, slope
length at the moderate level (2% - 4%) covered an area of
approximately 509.022 km2 or equal to 4.44%, slope
length at the high level (4% - 5%) covered an area of
approximately 2,747.928 km2 or equal to 23.95 %, and
slope length at the very high level (5% - 6%) covered an
area of approximately 6,975.324 km2 equal to 60.880 %.

On part of the study into S factor, it was found
that Nan Province has S factor of 0.065 - 47.138. When
the data were divided into 5 levels (Figure-6) for the
assessment of slope steepness, it was found that slope
steepness at very low level of 0.065 - 4.864 covering an
area of approximately 2,834.048 km2 or equal to
24.7039%, slope steepness at low level of 4.684 - 10.956
covering an area of approximately 2,447.294 km2 or equal
to 21.332 %, slope steepness at moderate level of 10.956 -
16.679 covering an area of approximately 2,845.091 km2
or equal to 24.800 %, slope steepness at high level of

16.679 - 23.140 covering an area of approximately
2,292.542 km2 or equal to 19.983 %, and slope steepness
at very high level of 23.140 - 47.138 covering an area of
approximately 1,053.096 km2 or equal to 9.179 %.

Figure-4. R factor analysis. Figure-5. L factor analysis.

In agricultural areas, soil erosion was very intense
and lack of water was found. In some area, rock fragments
were found scattered around land surface. In the analysis
of the soil erodibility factor or K factor, the soil group
found in Nan Province was input by K factor within the
range of 0.06 – 0.35 while Thailand’s K factor was 0.04 -
0.56. After that, data were classified in 3 levels (Figure-4)
to determine the distribution of soil erodibility factor. It
was found from the study that K factor at low level (0.06 -
0.23) covered an area of approximately 40.697 km2 or
equal 0.35 %, K factor at moderate level (0.23 - 0.29)
covered an area of approximately 510.480 km2 or equal to
4.45 %, and K factor at high level (0.29 - 0.35) covered an
area of approximately 10,920.892 km2 or equal to 95.20 %.

Result of L and S Factor Analysis
Geographic characteristics are factors that affect

soil erosion. They triggers the gravity to play more role in
causing soil erosion. Two key characteristics of geography
include slope length (L) and slope steepness (S). In the
area where a high level of slope steepness (S) and a high
level of slope steepness (S) are present, the severity of
running water follows to thus cause increasingly more
erosion than in plain areas. In this study, the L factor and S
factor were extracted from the 30 meter DEM of USGS. It

826

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

Figure-6. S factor analysis. Figure-7. C factor analysis.

Result of C Factor Analysis Result of P Factor Analysis
Plants as soil cover are key factors in preventing The conservation practice Factor is the factor that

the soil erosion since they help to absorb and reduce shows the capability of controlling soil erosion. It was
crashing force of rain, to slow down running water on soil calculated from the ratio of soil loss obtained from
surface, to help soil better hold together, to increase soil experimented land plot where there was a kind of
space so that water can flow down more, and to help to conservation and the soil loss obtained from the
promote activities of living organisms in soil. In this study, experimented land plot where the soil was plowed down
the soil group map at 1: 25,000 scale of Land the slope when other conditions stayed unchanged. In this
Development Department was used and C factor was study, the soil group map at 1:25,000 scale was used. After
input. The comparison was made with the data on the plant that, P factor was input. The comparison was made with
management of the Land Development Department. That the data on factors related to the plant management of the
created a map showing cropping management factors. Land Development Department. That created a map
When the data was divided into 5 levels (Figure-7) for the showing the conservation practice factor.
assessment of C factor, it was found that C factor in very
low level of 0 covering an area of approximately 139.286 When data was divided into 2 levels (Figure-8)
km2 or equal to 1.21%, C factor in low level of 0 - 0.02 for the assessment of P factor in the study area, it was
covering an area of approximately 10,185.410 km2 or found at P factor at low level of 0 - 0.098 covering an area
equal to 88.78%, C factor in moderate level of 0.02 - 0.048 of approximately 3.891 km2 or equal to 0.0339%, P factor
covering an area of approximately 241.878 km2 or equal to at high level of 0.098 - 1 covering an area of
2.11%, C factor in high level of 0.048 - 0.280 covering an approximately 1,458.581 km2 or equal to 99.966%.
area of approximately 503.209 km2 or equal to 4.39%, and
C factor in very high level of 0.280 - 0.340 covering an
area of approximately 402.289 km2 or equal to 3.51%.

827

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

Figure-8. P factor analysis. Figure-9. Soil erosion analysis.

Analysis Result of Areas Prone to Soil Erosion Table-1. Areas and risk levels.
The analysis into the areas that were prone to soil
No. Risk level Area
erosion was performed by an overlay analysis of R factor,
K factor, L and S factors, C factor, and P factor. Then the 1 Very low Km2 %
results were divided according to the severity of soil 2 low
erosion into 5 levels (Figure-9). The Figure-9 shows the 3 2,120.193 18.48
areas which are predominantly dark green and light green, 4 Moderate
many areas are lower part and middle part of the 5 High 2,728.854 23.79
province, with the possibility of soil erosion in the very Total
low and low level, covering the area of about 4,849.043 Very High 2,937.823 25.61
km2 equal to 42.27%; the area which is yellow is mostly
the lower part at the eastern region and western region 2,133.648 18.60
with the possibility of soil erosion at the moderate level
covering the area of about 2,937.822 km2 equal to 25.61%; 1,551.558 13.52
much of the orange area is middle region and some part of
it in the northern part has the possibility of the occurrence 11,472.076 100
of soil erosion at the high level covering the area of about
2,133.648 km2 equal to 18.60%; and much of the red area CONCLUSIONS
is northern part of the province with the possibility of soil The soil erosion in Thailand frequently takes
erosion at the very high level covering the area of about
1,551.558 km2 equal to 13.52%. The areas of the place in the Northern region of the country following
mentioned risk levels are also summarized in Table-1. heavy rains over mountains that are sources of rivers. The
severity of landslide depended upon the rainfall on the
mountain, the steepness of the mountain, the abundance of
the forest, and the geological characteristics of the
mountain. This study embraced the integration of geo-
informatics technology with the ULSE to analyze the areas
which were prone to soil erosion where soil erosion took
place every year.

According to the study, it could be concluded that
Nan Province had areas prone to soil erosion of about
3,685.206 km2 or equal to 57.73% due to its geography in
general which was characterized by forest and mountain
for almost 75 % and the plain area for 25 % or at forest
and mountain areas to plain area ratio of 3:1. Therefore,

828

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

many of the agricultural areas where plants were grown performance of Sheshegu community farmers in the
was on the mountain at the steepness level of more than Eastern Cape of South Africa. Journal of Agricultural
5% with a total area of 6,975.325 km2 or equal to 60.80%. Science. 5(5): 140-147.
In such area, land use should be changed from plants to
farm plants and further to perennial plants. Also, there Mateos E., Edeso J. M. & Ormaetxea L. 2017. Soil
should be measures in conserving soil and water deemed Erosion and Forests Biomass as Energy Resource in the
suitable for conditions of the area in order to reduce soil Basin of the Oka River in Biscay, Northern Spain. Forests.
erosion. 8(7): 258.

ACKNOWLEDGEMENTS Musgrave G.W. 1947. The quantitative evaluation of
This research was financially supported by factors. Journal of Soil and Water Conservation. 2(3): 133-
138.
Defence Technology Institute (Grant year 2020). Help and
support from disaster team management of the institute Nearing M., Xie Y., Liu B. & Ye Y. 2017. Natural and
were highly appreciated and acknowledged herewith. anthropogenic rates of soil erosion. International Soil and
Water Conservation Research. 5(2): 77-84.
REFERENCES
Ozsahin E., Duru U. & Eroglu I. 2018. Land Use and Land
Baver L.D. 1933. Somes soil factors effecting erosion. Cover Changes (LULCC), a Key to Understand Soil
Agricultural Engineering. 14(2): 51-52. Erosion Intensities in the Maritsa Basin. Wate. 10(3): 335.

Belo D., Ernawati R., Cahyadi T., Nurkhamim. & Amri N. Ozsahin E. & Eroglu I. 2019. Soil Erosion Risk
2020. Analysis of Land Erosion Due to Mining of Clay Assessment due to Land Use/Land Cover Changes
Material in Sidorejo Village, Sleman District, Yogyakarta. (LULCC) in Bulgaria from 1990 to 2015. Alinteri Journal
Geographia Technica. 15: 33-41. of Agriculture Sciences. 34(1): 1-8.

Browning G.W., Parish, C.L. & Glass J.A. 1947. A Panagos P., Meusburger K., Van Liedekerke M., Alewell
method for determining the use and limit rotation of C., Hiederer R. & Montanarella L. 2014. Assessing soil
rotation and conservation practices in control of soil erosion in Europe based on data collected through a
erosion in Iowa. Soil Science Society of America, European Network. Soil Science and Plant Nutrition. 60
Proceedings. 23: 246-249. (1): 15-29.

Burt T. & Weerasinghe K. 2014. Rainfall Distributions in Pholkerd R., Khunrattanasiri W. & Pattaratuma A. 2012.
Sri Lanka in Time and Space: An Analysis Based on Application of Remote Sensing and Geographic
Daily Rainfall Data. Climate. 2(4): 242-263. Information System for Soil Erosion Assessment in Huay
Nam Rit Watershed, Uttaradit Province. Thai Journal of
Conforti M. & Buttafuoco G. 2017. Assessing space–time Forestry. 31(2): 42-52.
variations of denudation processes and related soil loss
from 1955 to 2016 in southern Italy (Calabria region). Plakayrungrassamee S., Pantanahiran W. & Navanugraha
Environmental Earth Sciences. 76: 457-475. C. 2011. Soil Erosion Analysis Using Universal Soil Loss
Equation (USLE) to Estimate the Loss of Plant Nutrient in
Cook H.L. 1936. The nature and controlling variables of Huaimaeprachan Watershed. Journal of Social Sciences
the water erosion process. Soil Science Society of America Srinakharinwirot University. 14: 1-12. (In Thai)
Proceedings. 1: 487-494.
Pradhan B., Chaudhari A., Adinarayana J. & Buchroithner
Cruz D., María J., Benayas J., Ferreira G., Monteiro A. & M.F. 2012. Soil erosion assessment and its correlationwith
Schwartz G. 2019. Evaluation of Soil Erosion Process and landslide events using remote sensing data and GIS: A
Conservation Practices in the Paragominas-pa case study at Penang Island, Malaysia. Environmental
Municipality (Brazil). Geographia Technica. 14(1): 14-35. Monitoring and Assessment. 184: 715-727.

Ganasri B. P. & Ramesh H. 2016. Assessment of soil Senanayake S., Pradhan B., Huete A. & Brennan J. 2020.
erosion by RUSLE model using remote sensing and GIS - Assessing Soil Erosion Hazards Using Land-Use Change
A case study of Nethravathi Basin. Geoscience Frontiers. and Landslide Frequency Ratio Method: A Case Study of
7: 953-961. Sabaragamuwa Province, Sri Lanka. Remote Sensing.
12(9): 1483.
Guzzetti F., Peruccacci S., Rossi M. & Stark C. 2008. The
rainfall intensity-duration control of shallow landslides Smith D.D. 1941. Interpretation of soil conservation data
and debris flows: An update. Landslides. 5(1): 3-17. for fiels use. Agricultural Engineering. 22: 173-175.

Ighodaro I.D., Lategan F.S. & Yusuf, S.F. 2013. The
impact of soil erosion on agricultural potential and

829

VOL. 16, NO. 8, APRIL 2021 ISSN 1819-6608

ARPN Journal of Engineering and Applied Sciences

©2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.

www.arpnjournals.com

Smith D.D. and Whitt D.M. 1947. Estimating soil losses
form field areas and clay pan soil. Soil Science Society of
America, Proceedings. 12: 485-490.

Smith D.D. & Wichmeier W.H. 1957. Factor effecting
sheet and rill erosion. Transactions of the American
Geophysical Union. 38: 889-896.

Suk-ueng K. & Chantima K. 2017. Application of
Geographic Information System to Land Use Suitability
Assessment in Ban Nanglae Nai, Muang District, Chiang
Rai Province. Kasalongkham Research Journal. 11(3):
163-174. (In Thai)

Van Doren C.A. & L.J. Bartelli. 1956. A method of
forecasting soil loss. Agricultural Engineering. 37: 355 -
341.

Vita P., Paola R., Bathurst J., Borga M., Crosta G., Crozier
M., Glade T., Guzzetti F., Hansen A. & Wasowski J. 1998.
Rainfall-triggered landslides: A reference list.
Environmental Geology. 35(2): 219-233.

Wischmeier W.H. & Smith D. D. 1965. Predicting rainfall
erosion losses from cropland east of the Rocky Mountain:
guide for selection of practices for soil and water
conservation. Agr. Handbook No. 282.USDA,
Washington, D.C. p. 47.

Wischmeier W.H. & Smith D. D. 1978. Predicting rainfall
erosion losses. A guide to conservation planning.
Agr.Handbook No.537.USDA, Washington, D.C. p. 49.

Zingg A.W. 1940. Degree and lenght of land slope as it
effects soil loss in runoff. Agricultural Engineering. 21(2):
59-64.

Zuazo V.H.D. & Pleguezuelo C.R.R. 2008. Soil-erosion
and runoff prevention by plant covers. A review.
Agronomy for Sustainable Development. 28: 65-86.

830

GIS BASED ANALYSIS FOR EMERGENCY RELIEF AND RESCUE
AND DISASTER MITIGATION

Chamnan Kumsap1 Teeranai Srithamarong2 and Suriyawate Boonthalarath3

Defence Technology Institute
47/433 Moo 3, Ban Mai, Pak Kret, Nonthaburi 11120, Thailand

1Email: [email protected]
2Email: teeranai.s@ dti.or.th
3Email: suriyawate.b@ dti.or.th

ABSTRACT

This research was aimed to contribute GIS capabilities to military emergency relief and rescue and
disaster mitigation that require the detailed analysis of an area and environment prior to performing a mission.
At an ultimate goal of maintaining faulty mechanic equipment that consists of backhoes and tailgate trucks for
the Mobile Development Unit 31 of the Armed Forces Development Command to prevent and solve public and
disasters problems in Pua district of Nan province, the objective was to use GIS for a route selection mission of
Mobile Development Unit 31 in the mission of disaster prevention and solution. A sample road network database
covering Pua district was prepared and tested for the simulation of an optimal route selection based on an actual
landslide incident in the district reported by the news media. The Mobile Development Unit 31 was set as the
starting point of the routing while the landslide location was set as the target point. Field survey along the selected
route was presented as proof of concept. More factors dictating route selection were recommended for a more
accurate route selection.

1. INTRODUCTION

The Armed Forces Development Command is a military agency under the Ministry of
Defense. It is an ally member of the Department of Disaster Prevention and Mitigation under
the Ministry of Interior. It has an important duty in preventing and solving public problems and
disasters. Its direct report units are scattered throughout Thailand to reach the problems of the
people in every corner of the country. Therefore, they are the military unit that is faced with a
wide variety of public services and disasters according to the area of responsibility. Units in
the northern part are located in mountainous region with high mountain terrain, they often
encounter landslides. Most of the equipment under responsibility is mechanical such as
backhoes or tailgate trucks, etc. Most of them have been in use for more than 10 years and
therefore have deteriorated over time. Mobile Development Units have also attempted to
maintain their conditions to help the people. Therefore, if principles and technology can be
applied to enable the units to continue to operate the faulty equipment, the units will perform
the disaster prevention and mitigation mission in the best interest of the people in the area.

This research project is a collaboration between Defence Technology Institute and
Mobile Development Unit 31 or MDU31 of Armed Forces Development Command. The goal
is to use Geographic Information Systems (GIS) to transform geospatial data into a tool for
emergency relief and rescue and disaster mitigation. The database will be used for disaster
management which requires the detailed analysis of the area and environment prior to
performing the mission. This will contribute to the maintenance of MDU31’s faulty mechanic
equipment in order for the MDU31 to prevent and solve public and disasters problems in the
study area of Pua district in Nan province. The objective is to use GIS to support the MDU31
in the mission of disaster prevention and solution in response to landslides in the study area by
optimal route selection for the transport of the faulty mechanical equipment. The technology
to maintain the state of mechanic equipment will be introduced to the units and the principles
and processes will be tested in the actual problem area.

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

2. GIS IN DISASTER MANAGEMENT

Coppock (1995) conducted a brief survey of the diversity of such hazards and made an
attempt to review what had been written in the past, a task made difficult by the wide range of
interests involved. The review showed that, within the GIS field proper, relatively little had
been published and that, within the disciplines studying natural hazards, few papers described
operational systems that were applied routinely, four examples of which were summarized. van
Westen (2000) discussed that the collection and management of spatial data from remote
sensing and GIS were regarded proper to handle a large amount of data and had demonstrated
their usefulness in disaster prevention, preparedness and relief. The objectivity and
reproducibility of assessment were considerably improved by sequential imagery interpretation
with quantitative description of the factors and well defined analytical procedures and decision
rules, which were applied to come to the hazard assessment. In response to the previous
discussion, Johnson (2000) claimed that GIS was the foundation for emergency management.
As soon as potential emergency situations were identified, mitigation needs could be
determined and prioritized. Utilizing existing databases linked to geographic features in GIS
made quick displays of values at risk possible. Thus, the closest and quickest response units
could be selected, routed, and dispatched to an emergency once the location was known. The
review of Tomaszewski et al. (2015) provided interdisciplinary literature from a variety of
spatially-oriented disaster management fields and demonstrated progress in various aspects of
GIS for disaster response. They further concluded that a GIS for disaster response research
agenda and provided a list of resources for researchers new to GIS and spatial perspectives for
disaster management research.

3. GIS BASED DISASTER MITIGATION CONCEPT

Figure 1. Concept of GIS based analysis for disaster mitigation.

The ngagement of GIS and disaster mitigation is proposed for military disaster
management as shown on Figure 1. All geo-referenced data is handled in GIS with emphasis
on landslide data and previous records of the incidents. This GIS systematic approach can be
applied to other areas with frequently incurred disasters. The spatial analysis capability of GIS
plays a major on the GIS side of the management while a military decision making alternatives
is the output result of disaster mitigation component. Policy and missions will drive the
mitigation plan while regions under responsibility contributes how decision is made and
equipment allocated.

4. RESEARCH METHODOLOGY

The research methodology proposed in this project is illustrated in Figure 2. Four stages
are followed to implement GIS based disaster mitigation that returns optimal routes to dispatch
military equipment from MDU31 to landslide sites.

Figure 2. Use of GIS based disaster mitigation to access landslide sites.
4.1 GIS data preparation

GIS data preparation was to ensure essential geo-spatial layers are ready for further
analysis processes. Four types of data were central to the spatial analysis for a landslide risk
map. Geological and topological conditions were integral in nature while environmental ones
needed further GIS data manipulation before the analysis. Rainfalls were largely regarded as
dynamic especially precipitation and rain-induced landslides. Rain statistics were influential
factors to the magnitude of rainfall to landslide incidents.
4.2 Environment analysis for landslide-risk areas

The analysis of environments for landslide – risk areas to produce a landslide – risk map
took the summation of weighted 4 factors. GIS data layers describing geological and
topological conditions each carry a 30% combined weight percentage while those containing
environmental and rain conditions were each carry a 20%combined weight percentage. The
result map revealed those patches prone to landslides. Though a road layer was weighted in the
weighting process, it next provided accessibility to the mapped landslide sites.
4.3 Mitigation command and control

Prior to implementing this stage, a road network needed to complete connecting edges
and nodes so that the network analysis could be reiterated for starting and end points. Road
attributes describing surface, lane number, width intersections etc. were conditions that later

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

determined whether military vehicles and equipment on which they could be transported from
the analyzed starting to end points. Records of vehicle maintenance and regular checks were
data for command and control of the vehicles for the mission (Figure 3 left) in which the traffic
was completely blocked by the landslide (Figure 3 right).

Figure 3. Maintenance practices (left) before mission in the site (right).

4.4 Onsite mission operation

This three cascaded operation of figure 2 includes command and control from the
MDU31 center, holding a big picture situation map and complete military equipment database,
incident station commanding the mission upon the selected routes to the sites with an offline
copy of dataset from the center, and the team responding to the incident with mobile devices
to track the selected routes and handling the military equipment at the landslide sites.

5. CASE STUDY

In order to achieve the objective of this research article, case studies of a landslide
incident retrieved from online media was showcased, road network data was analyzed for the
route selection mission of MDU31 in disaster prevention and solution. The following describes
the case study extracted from the lowest portion of Figure 2 in response to the objective.

5.1 Landslide-prone study area

As part of the MDU31 landslide disaster management project, Pradabmook and
Laosuwan (2021) reported the research output that Nan Province had areas prone to soil erosion
of about 3,685.206 km2 or equal to 57.73% due to the topology characterized by forest and
mountain for almost 75 %. Where agricultural activities were found to be planted on the
mountain with steepness of more than 5% in a total area of 6,975.325 km2 or equal to 60.80%.

5.2 GIS road network

Yi et al (2012) calculated the shortest evacuation routes between affected points and
shelters or Origin - Destination ranking model where considerable roads and land features and
other environmental factors when the closest facilities and routes were selected, selection
criteria and approach methods could be suggested for future events. Likewise, in this research
the network of roads was formed by the connectivity of arc segments constituting an individual
road. Then, road network database consisted of Edge to connect components such as sections
or intersections, Junction to connect arcs, and Turn to define directions. Connectivity analysis
came in two types i.e. group connectivity and road connectivity within the same group. The
latter connectivity connected roads of the same group in two types namely Endpoint
connectivity for simulating object crossover and Vertex connectivity for dividing a line
segment into sub-segments. A snapshot of Pua road network dataset is shown on Figure 4.

Figure 4. Pua district road network dataset.
5.3 Actual landslide incident

According to Siamrath online (at https://siamrath.co.th/n/97454#) on 17 August 2019 at
16:34 Nan province local time, there were heavy rains day and night and 60 villages of Nan
province were at risk of flooding and landslide blocking the road linking Pua district to Bo
Kluea district. Along the road from Pua to Bo Kluea at the front gate to Doi Phu Kha National
Park, the road was blocked by sliding mountain.
6. ROUTE SELECTION AND VALIDATION
6.1 Route selection

Figure 5. Selected route and simulation for validation.
The road network analysis for route selection returned the route result as shown on Figure
5 left. The starting point of the route began at MDU31 (see the lower left), traversed along
National Highway No. 1080, National Highway No. 1258 and Nan Rural road No. 2047 to end
at the landslide incident area as reported online by the media. The total distance was measured
at 30.3927 km.
6.2 Selected route validation

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

Figure 5 right was a snapshot extracted from the flythrough simulation of the selected
route generated from the 5 cm. resolution mosaic of orthoimagery being draped on the DSM
of the same resolution. An arrow is to provide a visual link from the snapshot to the selected
route resulted from the road network analysis. Road surface was assumed to be concrete with
the sufficient road width to accommodate military vehicle to transport to the site. Road
characteristics input to GIS attributes were on the way in the project. Site ground survey could
have best validated the selection but the COVID-19 pandemic made it impossible.

7. RESULT AND DISCUSSION

The route of 30.3927 km. distance was selected from the dataset to demonstrate the
integrated GIS and military decision making for the MDU31 to access the actual landslide site.
The route was simulated to illustrate the road conditions sufficient for the transport of MDU31
vehicles and equipment to the blocked road of the landslide site. However, the complete use of
GIS based analysis for emergency relief and rescue and disaster management for optimal access
to landslide sites was subject to further studies of DTI ongoing project for MDU31. Road
conditions were recommended for the more accurate route selection. More surveys to update
the road dataset wwere under development as well as integral military decision making of
MDU31 for the disaster management. Other landslide sites as reported by the press will be
input to the analysis for solutions to test and evaluation of the dataset for road network analysis.

8. REFERENCES

Coppock, J.T., 1995. GIS and Natural Hazards: An overview from a Gis Perspective. In:
Carrara A., Guzzetti F. (eds) Geographical Information Systems in Assessing Natural
Hazards. Advances in Natural and Technological Hazards Research, vol 5. Springer,
Dordrecht. https://doi.org/10.1007/978-94-015-8404-3_2.

Johnson, R., 2000. GIS Technology for Disasters and Emergency Management. An ESRI
White Paper - May 2000. Redlands, USA. 12p.

Kumsap, C., 2018. Concept of Mobile C4ISR System for Disaster Relief. National Defense
Studies Institute Journal, January – April 2018, Vol 9 No.1, pp. 7 – 19.

Kumsap, C., Witheetrirong, Y., and Pratoomma, P., 2016. DTI's modeling and simulation
initiative project to strive for the HADR mission of Thailand's ministry of defence.
Proceedings of the 6th International Defence and Homeland Security Simulation
Workshop, September 26-28 2016, Cyprus, 44-51.

Pradabmook, P. and Laosuwan, T. 2021. The Integration of Geo-informatics Technology with
Universal Soil Loss Equation to Analyze Areas Prone to Soil Erosion in Nan Province. ARPN
Journal of Engineering and Applied Sciences, Vol. 16 No. 8, 823-830.

Robert, O. P., Kumsap, C. and Janpengpen, A., 2018. Simulation of counter drugs operations
based on geospatial technology for use in a military training simulator. International
Journal of Simulation and Process Modelling, Nol.13 No.4, pp. 402 - 415.

Tomaszewski, B., Judex, M.,Szarzynski, J., Radestock, C. and Wirkus, L., 2015. Geographic
Information Systems for Disaster Response: A Review. Journal of Homeland Security and
Emergency Management. June 2015. DOI: 10.1515/jhsem-2014-0082.

van Westen, C.J., 2000. Remote Sensing for Natural Disaster Management. International Archives of
Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. pp. 1609 - 1617.

Yi, C., Park, R. S., Murao, O., and Okamoto, E., 2012. Emergency management: Building an O-D
ranking model using GIS network analysis. Journal of Disaster Research, Vol.7 No.6, 793-802.

FLOOD RISK FIELD SURVEY USING MOBILE GIS IN PUA
SUBDISTRICT, PUA DISTRICT, NAN PROVINCE, THAILAND

Phaisarn Jeefoo1,* Watcharaporn Preedapirom 2 and Chamnan Kumsap 3

1 Research Unit of Spatial Innovation Development (RUSID), Geographic Information Science, School of
Information and Communication Technology (ICT), University of Phayao 19 Moo 2, Maeka, Muang, Phayao

56000 – Thailand
Email: [email protected] / [email protected]
2 Physiology, School of Medical Sciences, University of Phayao 19 Moo 2, Maeka, Muang, Phayao 56000 –

Thailand

Email: [email protected]

3 Defence Technology Institute, Office of the Permanent Secretary of Defence (Chaengwattana) 5th Floor,
47/433 Moo 3, Ban Mai, Pak Kret, Nonthaburi 11120 – Thailand
Email: [email protected]

ABSTRACT

This research paper presents the application of flood mobile field survey in Pua subdistrict, Pua district,
Nan province by using free and open source software. Geographic Information Systems (GIS) technology is
ideally suited as a tool for the presentation of data derived from continuous monitoring of locations and used to
support and deliver information to environmental managers and the public. Combined with Google API AppSheet,
it extended web capabilities to provide real-time data from notified activities. Both geographical data and
remotely sensed and geo-referenced image data were provided, and the ground truth of Google map remote
sensing was recognized and also further recommended for capability study. This application provided the
opportunity to visualize and grasp the current situation of the flood and thereby managed to offer prompt decision
making as an action plan immediately needed.

1. INTRODUCTION

Natural disaster compounded by climate change causes more than $500 billion in losses
every year (As Natural Disaster Rise, 2017). In particular, flooding is one of the most frequently
occurring natural catastrophic events (Sanyal and Lu, 2004) impacting human lives,
infrastructure and environment around the globe (Klema, 2014: Schumann and Moller, 2005:
Anusha and Bharathi, 2019). Floods are among the most devastating natural hazards in the
world and wildly distributed leading to significant economic and social damages than any other
natural phenomenon (DMSG, 2001; Haq et al., 2012). Climate changes and human-induced
land-use interventions are defined as important factors causing the flood hazard. There is a
mutual trigger situation that the urban areas are the most influenced areas from flooding and
also urbanization is the most important reason of the formation of flood (Ozkan & Tarhan,
2015). Remote Sensing has made substantial contribution in flood monitoring, mitigation and
damage assessment that leads the disaster management authorities to contribute significantly.
Geographic Information Systems (GIS) technology is ideally suited as a tool for the
presentation of data derived from continuous monitoring of locations and used to support and
deliver information to environmental managers and the public. GIS based spatial analysis and
visual elements are used frequently in recent years for the detection of flood hazard areas and
for the preparation of maps. GIS applications based on database and analysis tools have logical
and mathematical relationships between the layers (Kourgiala & Karatzas, 2011).

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

Mobile GIS is a mature technology which takes geospatial technology beyond the walls
of an office. Therefore, mobile applications have extended to field use which allows the user
easy access, storage, updates, analysis and real time visualization of field data. Till recently,
mobile GIS applications were mainly used as a navigation or location-aware system. Mobile
GIS technology nowadays offers a potential alternative to fill the gaps of traditional GIS
systems. With mobile GIS technology, officers and many other field workers have the potential
to access the enterprise geospatial data from the server-side to accomplish their tasks with high
level of accuracy. More importantly, it is also possible to update these geospatial enterprise
data in real time (Choosumrong et al., 2016: Jeefoo, 2019).

The main objective of this research was to develop the mobile GIS field survey by using
open source software for correcting flood risk hazards in Pua subdistrict, Pua district, Nan
province, Thailand

2. MATERIAL AND METHOD

2.1 Study area: Pua subdistrict, Pua district, Nan Province
Pua subdistrict, Pua district, Nan province in the northern part of Thailand (Figure 1) was

selected as the study area. Pua subdistrict comprises of 12 villages and covers an area of 23.9
sq.km. with geographical location between 19° 9’ N to 19° 12’ N and 100° 52’ 30’’ E to 100°
55’ 30’’ E. It is mostly covered with forested mountain, with an approximate elevation of 310
meters about mean sea level.

Figure 1. Geographic location of the study area.

2.2 Method
A smartphone running Android/iOS operating system was chosen to be a field device.

The chosen smartphone was used for sending the flood risk field survey data in real-time to the
base of operation to serve various purposes of field surveys. Real-time data availability
provides many advantages.

Figure 2 shows the architecture of the flood risk field survey application. The application
running on the device has two major modules: the map module and the survey module. The
map module is used for retrieving the location data from the Google Maps. This location will
be sent along with other types of data to the cloud server, and it can be used to pinpoint the

Short Paper Title

current location of the device when displaying a map. The second module is the survey module.
This module takes care of getting the information from the flood data collector including type
of the report, description, latitude-longitude and images.

Connect to your data Customize app Deploy to users

Data Base

(Location,
camera,
text report)

Flood data collection

Figure 2. Architecture of the flood risk field survey application.

When the field data collector fills in details by clicking on the SURVEY button, the data
will be sent to the cloud server via Wi-Fi network or mobile network (3G/4G/5G).

The system was being used in Pua subdistrict, Pua district, Nan province, Thailand. Field
data collectors had the basic information of all the flood or flash flood in the area database such
as elevation, slope, geography, climate, culture, etc. that was collected. However, they were
unable to identify the location of the flood situation.

Google Sheet created the flood database and triggered the build app on AppSheet website
(Figure 3 and Figure 4), https://www.appsheet.com/.

Figure 3. MAP page build app.

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

Figure 4. SURVEY page build app.
The server-side provided access to geospatial data and performed online spatial requests
such as find, spatial query, measure, and closeness analysis based on requests made by from
client-side. On the other hand, the user at the client-side could navigate and display through
separate GIS layers of the geospatial data hosted by the server-side.
Application of the mobile side of the system was concentrated on the mobile GIS
application. The previous application was used for field survey report from the geospatial field
survey. The GPS location in the smartphone was adept of pinpointing the current lat/long
location automatically. Once the existing location had been reached, the user would be able to
start inserting the data using the input form.
3. RESULTS
The application of the mobile-side of the system was concentrated on the mobile GIS
application. The previous application was used for flood situation report from the geospatial
filed survey. The collectors got access to the app, then identified their existing location. The
GPS location in the smartphone was adept of pinpointing the current location automatically.
Once the existing location was found, the user started inserting the data using the SURVEY
form.
Figure 4 below shows some screenshots of the application.

Short Paper Title

a) Main page, MAP b) Database table with c) phone, camera, location
reporter, age, gender, functions
address functions

d) photo input and location e) SURVEY page f) Database, reported

automatically

Figure 4. Screenshots of the flood risk field survey application.

The implementation of the flood data collection using mobile GIS field survey consisted of
the software used and details of item software version. The flood risk field survey using Mobile
GIS technology was designed and developed with a user-friendly main interface (Figure 4a).
The main screen of the application provides access to the reporting tools. Reporting tool for
field data includes data and image files of current location (Figure 4 (b, c, d, e, f)). The
application development environment and tool are shown in table 1.

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

Table 1. Application development environment and tool.

No. Flood Risk Field Survey using Mobile GIS

Software and Hardware Software

1. Server Cloud Server

2. Operating System Server Windows 10 Enterprise

3. Web Server AppSheet

4. Application Server AppSheet

5. Database Server Google Sheet

7. User Interface AppSheet

8. Client Web Browser Chrome, Firefox, Internet
Explorer, Safari

The web interface for flood risk field survey is shown in Figure 5. The collectors can
visualize the reporting points of real time field survey that send data, and they can make use of
that data for recording and analyzing purposes.

Figure 5. Web interface for flood risk field survey.

By clicking on a pinpoint which was the location of the house that was flooded on the
map, the information associated with the image such as latitude, longitude, flood status,
reporter, and date was automatically linked with geographic data such as names of subdistrict,
district, and province.

4. CONCLUSION

Google Maps provides its source of base map and user friendly applications. Freeware
products can be easily and quickly downloaded and installed. The interface is well organized
and easy to follow. Data recording tools are fairly user friendly, easy to figure out, and
supportive to users with multiple data forms for output and sharing. This is a good free mobile
tool, especially in the context of training others to use it, given its simple and easy to understand
design. The implemented mobile GIS platform provides the basic GIS functionalities and

Short Paper Title

location. The new generation of mobile network technology advances rapidly, and the storage
capacity of intelligent communication terminal increases substantially. So that the mobile GIS
has become the new hot spot following Desktop GIS and Web GIS (Wu, 2012: Jeefoo, 2014).
The client/server GIS framework that was developed was an independent application, which
could be run in every modern mobile smartphone without requiring any other additional
software. This application helped the field parties to gather data from flood risk field survey
and provided inputs for monitoring and protection.

5. ACKNOWLEDGEMENT

This research was supported by Defence Technology Institute (DTI), Thailand. Spatial
thanks go to the Mobile Development Unit 31 (MDU31) of Pua district, Nan province for
supporting essential data and information.

6. REFERENCES

Anusha, N., and Bharathi, B. 2019. Flood detection and flood mapping using multi-temporal synthetic
aperture radar and optical data. The Egyptian Journal of Remote Sensing and Space Sciences,
https://doi.org/10.1016/j.ejrs.2019.01.001 (accessed 9 July 2021)

As Natural Disaster Rise, Countries Call for Action on Resilient Crisis Recovery Planning, 2017.
https://www.worldbank.org/en/news/feature/2017/06/06/as-natural-disasters-rise-countries-call-
for-action-on-resilient-crisis-recovery-planning (accessed 10 July 2021).

Choosumrong, S., Raghavan, V., Jeefoo, P., & Vaddadi, N. (2016). Development of Service Orinted
Web-GIS Platform for Monitoring and Evaluation using FOSS4G. International Journal of
Geoinformatics, 12(3), 67-77.

DMSG, 2001. The Use of Earth Observing Satellites for Hazard Support: Assessments & Scenarios.
Committee on Earth Observation Satellites Disaster Management Support Group, Final Report,
NOAA, Dept. Commerce, USA.

Haq, M., Akhtar, M., Muhammad, S., Paras, S., & Rahmatullah, J. (2012). Techniques of Remote
Sensing and GIS for flood monitoring and damage assessment: A case study of Sindh province,
Pakistan. The Egyptian Journal of Remote Sensing and Space Sciences 15, 135-141.

Jeefoo, P. 2014. International Conference on Information Science & Application (ICISA), Real-time
field survey using android-based interface of mobile GIS.
https://ieeexplore.ieee.org/document/6847455 (accessed 11 July 2021)

Jeefoo, P. 2019. Wildfire field survey using mobile GIS technology in Nan province. The 4th
International Conference on Digital Arts, Media and Technology and 2nd ECTI Northern Section
Conference on Electrical, Electronics, Computer and Telecommunications Engineering, Nan,
Thailand. https://ieeexplore.ieee.org/document/8692291

Klemas, V., 2014. Remote sensing of floods and flood-prone areas: an overview. J. Coast. Res. 31 (4),
1005-1013.

Kourgiala, N., & Karatzas, G. (2011). Flood management and a GIS modelling method to assess flood-
hazard areas – a case study. Hydrological Sciences Journal, 56(2), 212-224.

Ozkan, S. P., and Tarhan, C. 2015. Detection of Flood Hazard in Urban Areas Using GIS: Izmir Case.
Procedia Technology 22, 373-381.

Sanyal, J., and Lu, X.X., 2004. Application of remote sensing in flood management with special
reference to monsoon Asia: a review. Nat. Hazards 33, 283-301.

International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2021

Schumann, G.-J.P., and Moller, D.K., 2015. Microwave remote sensing of flood inundation. Phys.
Chem. Earth 83-84, 84-95.

Wu, L. 2012. 2nd International Conference on Remote Sensing, Environment and Transportation
Engineering, Research and development of mobile forestry GIS based on intelligent terminal,
IEEE, 978-1-4673-0875-5/12. https://ieeexplore.ieee.org/document/6260685 (accessed 11 July
2021).

Short Paper Title