DISTRIBUTION ESTIMATION MODEL FOR HEAVY METALS CONTAMINANT OF LEAD (Pb) AND CADMIUM (Cd) IN SOIL USING INVERSE DISTANCE WEIGHT (IDW) AND ORDINARY KRIGING (CASE STUDY: RANCAEKEK)

Rancaekek located in Bandung district is one of the areas with severe environmental damage and it has been going on in a long time. The lack of waste management from industries has resulted soil contamination such as heavy metals which are classified as hazardous waste. Inverse Distance Weight (IDW)...

Full description

Saved in:
Bibliographic Details
Main Author: FERLIANDE (NIM: 25313005), JEFRI
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/22679
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:Rancaekek located in Bandung district is one of the areas with severe environmental damage and it has been going on in a long time. The lack of waste management from industries has resulted soil contamination such as heavy metals which are classified as hazardous waste. Inverse Distance Weight (IDW) as the estimation model that looking the effect of distance between sample and estimation point without considering spatial statistics. Whereas Ordinary Kriging (OK) is part of geostatistical model considering spatial statistics of the data. Furthermore, the both of models are compared to obtain the best model in study area. The study indicates that IDW is better than Ordinary Kriging for estimating lead (Pb) which gives RMSE value in cross validation and validation is 5.03 ppm and 6.15 ppm, respectively. In contrast, Ordinary Kriging is better than IDW for estimating cadmium (Cd) which gives RMSE value in cross validation and validation is 0.19 ppm and 0.07 ppm, respectively. Spatial correlation in Ordinary Kriging is obtained from semivariogram with major range is 1215 m at azimuth 900 and minor range is 450 m at azimuth 00. This research also uses GIS (Geographical Information System) Technology in order to provides estimation results more representative.