U S I N G D ATA M I N I N G T O P R E D I C T FOREST FIRES IN AN ARIZONA FOREST

Size: px
Start display at page:

Download "U S I N G D ATA M I N I N G T O P R E D I C T FOREST FIRES IN AN ARIZONA FOREST"

Transcription

1 Sanjeev Pandey CSC 600, Data Mining 5/15/2013 Table of Contents 1. Introduction 2. Study Area and Data Set 3. Data Mining Approaches 4. Results 5. Conclusions U S I N G D ATA M I N I N G T O P R E D I C T FOREST FIRES IN AN ARIZONA FOREST 1. Introduction Almost since the beginning of the period of westward expansion in the history of the United States, there has been an interest in predicting and preventing forest fires. These fires have had a major role in western forests for thousands of years, having many beneficial results that are important for forest ecology. However several factors have led to the rise of catastrophic, large-scale fires that until recently had been very rare. The causes of these changes are debatable, however the desire to prevent such fires from occurring is universal in cities and towns near fire-prone forests. A key component of preventing large-scale fires is early intervention. Primarily this is done through surveillance for early detection and having large numbers of fire crews and equipment available over a wide geographic area during the fire season for intervention. Given existing financial constraints on government funded land management agencies and the increasing intensities of fire seasons due to climate change and prolonged drought, there is increasing focus on improving the ability to predict when, where, and how extensive fire activity will be on a given day. There are already several methods in place for predicting fire behavior before a fire has started. One tool is the Forest Fire Weather Index, which is a Canadian empirical model developed in the 1970 s to estimate fire risk for a given day using lookup tables based upon meteorological information such as temperature, relative humidity, wind speed, and precipitation. There are also tools that use statistical models to correlate fire risk with historical weather and fire history data. One that is widely used for this purpose is called Fire Family Plus. This program is available for free from the Rocky Mountain Research Station of the US Forest Service and uses meteorological data and fire history to calculate fire danger rating indices, summarize data, and jointly analyze fire and weather data using classic statistical methods (Bradshaw and Tirmenstein 2010). While these methods can be effective, they do not attempt to quantitatively predict fire behavior based on inexpensive and readily available real-time weather data. Several studies have investigated the possibility of using data mining techniques in the study of forest fire prediction. Cortez and Morais (2006) used multiple data mining techniques including multiple regression, Decision Trees, Random Forest, Neural Networks, and a Support Vector Machine to predict burned area for forest fires in a Portuguese natural park using meteorological data from weather stations located throughout the park. They found that using a SVM model with four inputs including temperature, rain, relative humidity and wind speed, they were able to predict small fires. Another study by Stojanova et Pandey 1/8

2 al. (2011) found that using weather modeling information in conjunction with vegetation modeling and GIS data they were able to predict fire occurrence on a given day in three different regions of Slovenia. This project will apply Naive Bayes, J48 Decision Tree, and SVM algorithms using meteorological data and fire history data in two phases to build a classifier that in the first phase predicts the likelihood of a fire occurring on a day based on the weather conditions. If it is concluded that a day has a high likelihood of having at least one fire, the second phase classifier will predict the number of fires ignited on that day, the total acreage of burned area that resulted from fires that ignited on that day, and the average fire size for fires that ignited on that day. The results of this analysis will be compared to the results from the other data mining and forest fire studies to assess their effectiveness. 2. Study Area and Data Set The study area is the Coconino National Forest located in northern Arizona near the city of Flagstaff. The forest is a mix of desert, several different types of forests, riparian areas, and alpine tundra ranging in elevation from 2,600 ft to 12,633 ft. Fires have burned in the forest regularly since the time of European settlement in the 1850 s, with the months between May and July usually being the peak of the fire season. Weather data was sourced from the Western Regional Climate Center (WRCC) Remote Automated Weather Station (RAWS) Network, which makes standardized historical data from its national network of meteorological stations available on the Internet. Data from four different weather stations (Flagstaff, Greenbase, Mormon Lake, Sedona) located within the Coconino National Forest was initially used. Data ranged from 4/1/1968 to 12/31/2012 and included average and high/low daily measurements for temperature and relative humidity, average wind speed and direction, precipitation duration, precipitation amount, solar radiation, and a binary value for if the station was wet. There were a total of 14,800 data points, which is less than the expected number of days for the date range (16,345), so it is assumed that there are missing dates in the dataset. The WRCC also maintains a database of fire history for a number of US land management agencies in all states. For this project the data for the Coconino National Forest, was used which had complete fire history data from 1970 to Fire history data included a unique identifier number for each fire, ignition date, discovery date, containment date, acres burned, cause, name, subunit of the forest effected, and location (in latitude and longitude) in addition to an number of fields that were blank or mostly incomplete. In total, there were 19,193 recorded fires between 1970 and The raw data for both sets was in a non-standard text format so the program Fire Family Plus was used to import the data into a Microsoft Access database that could then be modified and exported to a csv file for use with WEKA. Preprocessing of the data utilized MS Access queries to eliminate some of the data fields from each dataset and merge the two into a single dataset using date of measurement from the weather data and ignition date for the fire history data as the key. The data was then imported into a Microsoft Excel spreadsheet for further preprocessing. A PivotTable was used to determine the total number of fires and total area burned for fires that started on the same day since there were many days during the peak of the fire season where multiple fires started on the same day. Once this was done, a field was added that calculated the average fire size for days with fire occurrence by dividing the total area burned by the total number of fires started on that day. A binary field was also added that indicated if a day had any fires at all. The data was then divided into two sets for the two phases. The first phase data set had Pandey 2/8

3 all of the weather data and the binary field for fire yes/no. The second phase data set had only weather data and fire characteristics (number, total size in acres, average size in acres) for days with fire. As recommended by Cortez and Morais (2006), a ( ln ( n + 1) ) transformation on the number of fires, acres burned, and average fire size was performed since the data tended towards days with fewer smaller fires. While preprocessing it was found that the Sedona, Greenbase, and Mormon Lake stations had incomplete data. It was also found that even though the fire history data had a field identifying the location within the forest where the fire occurred, it was difficult to properly correlate the closest weather stations to a specific fire. For these reasons, the dataset was filtered to use only data from the Flagstaff weather station. Flagstaff is located in the middle of the Coconino National Forest in terms of location and elevation, so its data is mostly the average for the entire forest as a whole. Once the dataset was loaded in WEKA, the weather data for both data sets was discretized using default settings (10 bins, not equal frequency). The fire characteristics for phase 1 were discretized into two bins (fire yes, fire no). The fire characteristics for phase 2 were discretized into 10 bins with equal frequency since most of the data was for low numbers and low acreages of fires for a day. Fig. 1: Raw weather data, phase 1 Pandey 3/8

4 Fig 2: Discretized weather data, phase 1 Fig 3: Discretized data for phase 1, predicting fire occurrence (blue is no fire, red is fire) Pandey 4/8

5 Fig 4: Discretized data for phase 2, fire attributes only. Bottom row is transformed data (ln (n+1)) 3. Data Mining Approaches! The data mining algorithms used were all implemented from within the WEKA machine learning software suite. The classifier methods used include Naive Bayes, J48 decision trees, Logistic Regression, and Support Vector Machines. Classifiers were run with default settings in all cases and using 10-fold cross validation.! For Phase 1, Naive Bayes and J48 were run along with the Simple Logistic algorithm to provide comparison to the results of a statistical model such as the one used in the Fire Family Plus software. For Phase 2, Naive Bayes, J48, and SVM (using the built-in WEKA classify function SMOReg) were run. 4. Results The first phase classifiers had the following results: Accuracy Precision Recall J % Naive Bayes 75.7% Simple Logistic 76.2% Table 1: Phase 1 classifier results Since phase 1 of the project bore a strong resemblance to the Stojanova et al. (2011) Slovenian study, comparison was made to their results. In that study, the researchers stated that since the purpose of the study was to estimate the risk of fire outbreak with a probability value for the occurrence of fire given the Pandey 5/8

6 values of the input data, accuracy of the classifier should be considered in conjunction with other quality measures such as precision, recall, Kappa, and Area Under the ROC Curve. Higher precision indicates fewer false positive results, which is important in the case of predicting fire outbreaks since more false positives can lead to a waste of resources. Higher recall indicates that positive results are correctly predicted. This is also very important in predicting fire outbreaks since it means that fewer fires occur on days that are predicted as non-fire days. In my results, all of the classifiers performed with nearly equal accuracy. The J48 was slightly better than the other two models and the Naive Bayes performed slightly worse than the other two models. For the precision indicator, all three models performed virtually the same, but in terms of recall the J48 and Simple Logistic models performed better than the Naive Bayes. The accuracy, precision and recall values for this project s models were comparable to the results from the Slovenian study using the same classifier methods. The Slovenian study had 4 separate regions that were modeled and the results from my study generally fell within the range of values that they found. Their results were better in the best cases with some classifiers having 86% accuracy (Bagging J48), but also worse in some cases where they had accuracy as low as 63% (Naive Bayes). It must be noted that the input data used in this study was very different from the data used to construct the models in the Slovenian study since they used predicted weather data along with vegetation models and geographic data while this study only used historical weather data. Phase 2 of the project is comparable to the Cortez and Morais study which attempted to predict fire size using historical weather data. In that study, several Data Mining methods were employed including Random Forest, Decision Tree, Neural Network, Support Vector Machine, and Multiple Regression. Overall performance was measured by comparing the mean absolute deviation and root mean squared values for the classifiers. Accuracy was also included for comparison to results of phase 1. The results of this project did not seem to compare favorably to the results found in Cortez and Morais. N Size Average Size J48 43% (0.149) (0.273) 25% (0.169) (0.311) 44% (0.153) (0.277) NB 36% (0.137) (0.278) 23% (0.165) (0.299) 38% (0.152) (0.284) SVM N/A 29% (0.306) (0.404) N/A Table 2: Accuracy (Mean Absolute Error) (RMSE) of classifiers for 3 different attributes 5. Conclusion! In this project, a data mining approach was employed to predict occurrence of fire using daily weather data and the characteristics of fire occurrences on days with fire using that same weather data. The models were able to predict fire occurrence with 76% accuracy, which is approximately equal to the predictive accuracy of statistical models using the same data. The models were less accurate in predicting characteristics of fire on days where fires occurred, including area burned, number of fires, and average fire size. Pandey 6/8

7 ! The findings for predicting fire occurrence were in line with other studies. This is probably because fire is mostly limited to only part of the year with winters having little to no fire activity and summers having the bulk of the fire activity. Since many of the weather indicators are functions of season, it makes sense that they could in turn predict fire occurrence with some accuracy.! The findings for predicting characteristics of fire, on the other hand, were inconclusive. There are several possible causes for the difficulty in modeling the fire characteristics. One reason could be that the total size burned may not necessarily relate to the size burned on the day of ignition. For example, a fire that burns for several days may be small for a number of days and then blow up in a single day. A better number to use for area burned might be a day-to-day total for a fire that burns over several days. However any value that is used for area burned has many other factors involved such as time to detection, resources allocated to fight the fire, and number of ignition points. Also, there may not be extensive data on day-today area burned for multi-day fires since this was difficult to track until fairly recently.! Performance of the model may also be strongly effected by the binning strategy used when discretizing the data. Initially I used 10 bins with unequal frequency for the attributes that the model attempts to predict, but the resulting classifiers seemed overly optimistic probably since a majority of the cases were in the first bin. I changed the discretizing function to use the same number of bins with equal frequency and the classifiers performed much worse. It may work better to reduce the number of bins, but that may also reduce the meaningfulness of the resulting model since the prediction will be of a lower resolution. The natural log transformation of the attributes had no discernible effect on the predictive accuracy of the models, probably since the data was discretized after the transformation which ended up making the untransformed data and transformed data virtually identical.! The models for phase 2 of the project were limited to only 5,754 input points. since it included only fire days Increasing the number of fire occurrences might be useful in training the classifier. To do this, one could include fire history information from other land management blocks with similar vegetation and geography. Good candidates could be the neighboring Kaibab, Prescott, and Apache-Sitgreaves National Forests along with other National Parks and Monuments, and Bureau of Land Management parcels in Arizona, New Mexico and the greater Southwest.! In comparing the results for prediction of fire characteristics with those of the study in Portugal, it could be that the models for the Coconino National Forest were simply overgeneralizing the diversity of the landscape. The Portuguese study was for an area that was 1/10th the size of the Coconino NF but with a higher density of weather stations. In this way, the classifier algorithms may be benefited by either limiting classification to the area in the direct proximity of a weather station or increase the number of weather stations used and correlate fires directly to a weather station near their ignition point. Fires that occurred outside the area of any possibly representative weather stations would not be used. However, using such an approach may lead to a model that is over-fitted to a specific region that would therefore have little use to land managers outside of that small area.! Overall a thorough comparison to Coretz and Morais could not be made since the analysis they used was much more sophisticated than the analysis for this project. In the end I was unable to faithfully reproduce their methods of data preprocessing and classifier development due to the use of different software to perform the analysis (WEKA vs. RMiner library for the R statistical environment) and a basic lack of understanding of their approach. I attempted to recreate their results since their dataset was publicly Pandey 7/8

8 available, but I am not sure I interpreted the output from WEKA correctly since the results did not compare correctly to the published results. The authors described the data set as a difficult regression task, perhaps as a warning. 6. Works Cited Bradshaw, L.S. and Tirmenstein, D. (2010). FireFamilyPlus Version 4 User s Guide. Draft Edition. Boise, ID. Cortez, Paulo and A Morais. (2006). A Data Mining Approach to Predict Forest Fires using Meteorological! Data. Department of Information Systems/R&D Algoritmi Centre, University of Minho. D. Stojanova, P. Panov, A. Kobler, S. Dzeroski, and K. Taskova. (2011) Estimating the risk of fire outbreaks in the natural environment.. Data Min Knowl Disc, March 2012, Volume 24, Issue 2, pp Witten I, Frank E (2005) Data Mining: Practical Machine Learning Tools and Techniques. 2nd ed. Morgan! Kaufmann, San Francisco. Pandey 8/8

Will it rain tomorrow?

Will it rain tomorrow? Will it rain tomorrow? Bilal Ahmed - 561539 Department of Computing and Information Systems, The University of Melbourne, Victoria, Australia bahmad@student.unimelb.edu.au Abstract With the availability

More information

2011 National Seasonal Assessment Workshop for the Eastern, Southern, & Southwest Geographic Areas

2011 National Seasonal Assessment Workshop for the Eastern, Southern, & Southwest Geographic Areas 2011 National Seasonal Assessment Workshop for the Eastern, Southern, & Southwest Geographic Areas On January 11-13, 2011, wildland fire, weather, and climate met virtually for the ninth annual National

More information

FireFamilyPlus Version 5.0

FireFamilyPlus Version 5.0 FireFamilyPlus Version 5.0 Working with the new 2016 NFDRS model Objectives During this presentation, we will discuss Changes to FireFamilyPlus Data requirements for NFDRS2016 Quality control for data

More information

8-km Historical Datasets for FPA

8-km Historical Datasets for FPA Program for Climate, Ecosystem and Fire Applications 8-km Historical Datasets for FPA Project Report John T. Abatzoglou Timothy J. Brown Division of Atmospheric Sciences. CEFA Report 09-04 June 2009 8-km

More information

Predicting wildfire ignitions, escapes, and large fire activity using Predictive Service s 7-Day Fire Potential Outlook in the western USA

Predicting wildfire ignitions, escapes, and large fire activity using Predictive Service s 7-Day Fire Potential Outlook in the western USA http://dx.doi.org/10.14195/978-989-26-0884-6_135 Chapter 4 - Fire Risk Assessment and Climate Change Predicting wildfire ignitions, escapes, and large fire activity using Predictive Service s 7-Day Fire

More information

Arizona Climate Summary May 2012

Arizona Climate Summary May 2012 Arizona Climate Summary May 2012 Summary of conditions for April 2012 April 2012 Temperature and Precipitation Summary April 1 st 16 th : Although April began with another low pressure system sweeping

More information

Hierarchical models for the rainfall forecast DATA MINING APPROACH

Hierarchical models for the rainfall forecast DATA MINING APPROACH Hierarchical models for the rainfall forecast DATA MINING APPROACH Thanh-Nghi Do dtnghi@cit.ctu.edu.vn June - 2014 Introduction Problem large scale GCM small scale models Aim Statistical downscaling local

More information

4.5 Comparison of weather data from the Remote Automated Weather Station network and the North American Regional Reanalysis

4.5 Comparison of weather data from the Remote Automated Weather Station network and the North American Regional Reanalysis 4.5 Comparison of weather data from the Remote Automated Weather Station network and the North American Regional Reanalysis Beth L. Hall and Timothy. J. Brown DRI, Reno, NV ABSTRACT. The North American

More information

MODELING LIGHTNING AS AN IGNITION SOURCE OF RANGELAND WILDFIRE IN SOUTHEASTERN IDAHO

MODELING LIGHTNING AS AN IGNITION SOURCE OF RANGELAND WILDFIRE IN SOUTHEASTERN IDAHO MODELING LIGHTNING AS AN IGNITION SOURCE OF RANGELAND WILDFIRE IN SOUTHEASTERN IDAHO Keith T. Weber, Ben McMahan, Paul Johnson, and Glenn Russell GIS Training and Research Center Idaho State University

More information

The AIR Bushfire Model for Australia

The AIR Bushfire Model for Australia The AIR Bushfire Model for Australia In February 2009, amid tripledigit temperatures and drought conditions, fires broke out just north of Melbourne, Australia. Propelled by high winds, as many as 400

More information

Arizona Climate Summary May 2018 Summary of conditions for April 2018

Arizona Climate Summary May 2018 Summary of conditions for April 2018 April 2018 Temperature and Precipitation Summary Arizona Climate Summary May 2018 Summary of conditions for April 2018 April 1 st 16 th : April was a very dry month across Arizona. Several low pressure

More information

Shootout 2017 Problem Statement

Shootout 2017 Problem Statement Shootout 2017 Problem Statement Understanding Wildfires 1 Background In the fall of 2016, a number of wildfires occurred in the southeastern United States. Spanning over 5 states and several weeks, more

More information

FOREST FIRE HAZARD MODEL DEFINITION FOR LOCAL LAND USE (TUSCANY REGION)

FOREST FIRE HAZARD MODEL DEFINITION FOR LOCAL LAND USE (TUSCANY REGION) FOREST FIRE HAZARD MODEL DEFINITION FOR LOCAL LAND USE (TUSCANY REGION) C. Conese 3, L. Bonora 1, M. Romani 1, E. Checcacci 1 and E. Tesi 2 1 National Research Council - Institute of Biometeorology (CNR-

More information

NIDIS Intermountain West Drought Early Warning System May 1, 2018

NIDIS Intermountain West Drought Early Warning System May 1, 2018 NIDIS Intermountain West Drought Early Warning System May 1, 2018 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations. From top to bottom, and

More information

Arizona Climate Summary February 2012

Arizona Climate Summary February 2012 Arizona Climate Summary February 2012 Summary of conditions for January 2012 January 2012 Temperature and Precipitation Summary January 1 st 20 th : The New Year has started on a very dry note. The La

More information

10.5 PROBABLISTIC LIGHTNING FORECASTS AND FUEL DRYNESS LEVEL FORECASTS IN THE GRAPHICAL FOREAST EDITOR: EXPANDED DOMAIN AND DISTRIBUTION FOR 2009

10.5 PROBABLISTIC LIGHTNING FORECASTS AND FUEL DRYNESS LEVEL FORECASTS IN THE GRAPHICAL FOREAST EDITOR: EXPANDED DOMAIN AND DISTRIBUTION FOR 2009 10.5 PROBABLISTIC LIGHTNING FORECASTS AND FUEL DRYNESS LEVEL FORECASTS IN THE GRAPHICAL FOREAST EDITOR: EXPANDED DOMAIN AND DISTRIBUTION FOR 2009 Chris V. Gibson 1*, and P. D. Bothwell 2, S. Sharples 3,

More information

WMO Aeronautical Meteorology Scientific Conference 2017

WMO Aeronautical Meteorology Scientific Conference 2017 Session 1 Science underpinning meteorological observations, forecasts, advisories and warnings 1.6 Observation, nowcast and forecast of future needs 1.6.1 Advances in observing methods and use of observations

More information

Virtual Beach Building a GBM Model

Virtual Beach Building a GBM Model Virtual Beach 3.0.6 Building a GBM Model Building, Evaluating and Validating Anytime Nowcast Models In this module you will learn how to: A. Build and evaluate an anytime GBM model B. Optimize a GBM model

More information

Analysis of Data Mining Techniques for Weather Prediction

Analysis of Data Mining Techniques for Weather Prediction ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 Indian Journal of Science and Technology, Vol 9(38), DOI: 10.17485/ijst/2016/v9i38/101962, October 2016 Analysis of Data Mining Techniques for Weather

More information

CHAPTER 6 CONCLUSION AND FUTURE SCOPE

CHAPTER 6 CONCLUSION AND FUTURE SCOPE CHAPTER 6 CONCLUSION AND FUTURE SCOPE 146 CHAPTER 6 CONCLUSION AND FUTURE SCOPE 6.1 SUMMARY The first chapter of the thesis highlighted the need of accurate wind forecasting models in order to transform

More information

Chapter 02 Life on Land. Multiple Choice Questions

Chapter 02 Life on Land. Multiple Choice Questions Ecology: Concepts and Applications 7th Edition Test Bank Molles Download link all chapters TEST BANK for Ecology: Concepts and Applications 7th Edition by Manuel Molles https://testbankreal.com/download/ecology-concepts-applications-7thedition-test-bank-molles/

More information

WindNinja Tutorial 3: Point Initialization

WindNinja Tutorial 3: Point Initialization WindNinja Tutorial 3: Point Initialization 07/20/2017 Introduction Welcome to. This tutorial will step you through the process of running a WindNinja simulation that is initialized by location specific

More information

National Wildland Significant Fire Potential Outlook

National Wildland Significant Fire Potential Outlook National Wildland Significant Fire Potential Outlook National Interagency Fire Center Predictive Services Issued: September, 2007 Wildland Fire Outlook September through December 2007 Significant fire

More information

Predicting flight on-time performance

Predicting flight on-time performance 1 Predicting flight on-time performance Arjun Mathur, Aaron Nagao, Kenny Ng I. INTRODUCTION Time is money, and delayed flights are a frequent cause of frustration for both travellers and airline companies.

More information

Arizona Climate Summary

Arizona Climate Summary Arizona Climate Summary June 2012 Summary of conditions for May 2012 May 2012 Temperature and Precipitation Summary May 1 st 16 th : May began with a surface cold front sweeping across the western states

More information

Arizona Climate Summary March 2013

Arizona Climate Summary March 2013 Arizona Climate Summary March 2013 Summary of conditions for February 2013 February 2013 Temperature and Precipitation Summary February 1 st 16 th : February began with a ridge of high pressure over the

More information

Arizona Climate Summary May 2013

Arizona Climate Summary May 2013 Arizona Climate Summary May 2013 Summary of conditions for April 2013 April 2013 Temperature and Precipitation Summary April 1 st 16 th : The first week of April featured a ridge of high pressure over

More information

Predicting Fire Season Severity in the Pacific Northwest

Predicting Fire Season Severity in the Pacific Northwest Predicting Fire Season Severity in the Pacific Northwest Paul Werth 1 Abstract Projections of fire season severity that integrate historical weather and fire information can be used by fire managers when

More information

Click Prediction and Preference Ranking of RSS Feeds

Click Prediction and Preference Ranking of RSS Feeds Click Prediction and Preference Ranking of RSS Feeds 1 Introduction December 11, 2009 Steven Wu RSS (Really Simple Syndication) is a family of data formats used to publish frequently updated works. RSS

More information

NATIONAL HYDROGRAPHY DATASET (NHD) UPDATE PROJECT FOR US FOREST SERVICE REGION 3

NATIONAL HYDROGRAPHY DATASET (NHD) UPDATE PROJECT FOR US FOREST SERVICE REGION 3 NATIONAL HYDROGRAPHY DATASET (NHD) UPDATE PROJECT FOR US FOREST SERVICE REGION 3 Allison Moncada California State University, Northridge February 2018 July 2018 Advisor: Joel Osuna Center for Geospatial

More information

Physical Geography: Patterns, Processes, and Interactions, Grade 11, University/College Expectations

Physical Geography: Patterns, Processes, and Interactions, Grade 11, University/College Expectations Geographic Foundations: Space and Systems SSV.01 explain major theories of the origin and internal structure of the earth; Page 1 SSV.02 demonstrate an understanding of the principal features of the earth

More information

Predictive Analytics on Accident Data Using Rule Based and Discriminative Classifiers

Predictive Analytics on Accident Data Using Rule Based and Discriminative Classifiers Advances in Computational Sciences and Technology ISSN 0973-6107 Volume 10, Number 3 (2017) pp. 461-469 Research India Publications http://www.ripublication.com Predictive Analytics on Accident Data Using

More information

National Wildland Significant Fire Potential Outlook

National Wildland Significant Fire Potential Outlook National Wildland Significant Fire Potential Outlook National Interagency Fire Center Predictive Services Issued: April 1, 2008 Next Issue: May 1, 2008 Wildland Fire Outlook April 2008 through July 2008

More information

Prediction of Snow Water Equivalent in the Snake River Basin

Prediction of Snow Water Equivalent in the Snake River Basin Hobbs et al. Seasonal Forecasting 1 Jon Hobbs Steve Guimond Nate Snook Meteorology 455 Seasonal Forecasting Prediction of Snow Water Equivalent in the Snake River Basin Abstract Mountainous regions of

More information

Elementary Social Studies Content Area Assessment Task. Winter 2010 EDUC 203

Elementary Social Studies Content Area Assessment Task. Winter 2010 EDUC 203 1 Elementary Social Studies Content Area Assessment Task Winter 2010 EDUC 203 Students: Fourth grade class in Watsonville, CA. Approximately 90% Latino, primarily of Mexicandescent. About 80% participate

More information

To Predict Rain Fall in Desert Area of Rajasthan Using Data Mining Techniques

To Predict Rain Fall in Desert Area of Rajasthan Using Data Mining Techniques To Predict Rain Fall in Desert Area of Rajasthan Using Data Mining Techniques Peeyush Vyas Asst. Professor, CE/IT Department of Vadodara Institute of Engineering, Vadodara Abstract: Weather forecasting

More information

Fire Susceptibility Analysis Carson National Forest New Mexico. Can a geographic information system (GIS) be used to analyze the susceptibility of

Fire Susceptibility Analysis Carson National Forest New Mexico. Can a geographic information system (GIS) be used to analyze the susceptibility of 1 David Werth Fire Susceptibility Analysis Carson National Forest New Mexico Can a geographic information system (GIS) be used to analyze the susceptibility of Carson National Forest, New Mexico to forest

More information

Quantifying Weather Risk Analysis

Quantifying Weather Risk Analysis Quantifying Weather Risk Analysis Now that an index has been selected and calibrated, it can be used to conduct a more thorough risk analysis. The objective of such a risk analysis is to gain a better

More information

Development of Empirical Weather Forecasting Techniques for Soaring Flight

Development of Empirical Weather Forecasting Techniques for Soaring Flight Development of Empirical Weather Forecasting Techniques for Soaring Flight George S. Young Department of Meteorology The Pennsylvania State University University Park, Pennsylvania 16802 young@meteo.psu.edu

More information

Colorado Alpine Dust Deposition and Associated Continental Winds 1

Colorado Alpine Dust Deposition and Associated Continental Winds 1 Hydrology Days 2010 Colorado Alpine Dust Deposition and Associated Continental Winds 1 Morgan Phillips 2 Colorado Climate Center and Bureau of Land Management Abstract. The winter and early spring of 2008-2009

More information

Improvement of the National Hydrography Dataset for US Forest Service Region 3 in Cooperation with the National Forest Service

Improvement of the National Hydrography Dataset for US Forest Service Region 3 in Cooperation with the National Forest Service Improvement of the National Hydrography Dataset for US Forest Service Region 3 in Cooperation with the National Forest Service Christian L. DeCasas California State University, Northridge November 7 th,

More information

Meteorology. Chapter 15 Worksheet 1

Meteorology. Chapter 15 Worksheet 1 Chapter 15 Worksheet 1 Meteorology Name: Circle the letter that corresponds to the correct answer 1) The Tropic of Cancer and the Arctic Circle are examples of locations determined by: a) measuring systems.

More information

The Road to Data in Baltimore

The Road to Data in Baltimore Creating a parcel level database from high resolution imagery By Austin Troy and Weiqi Zhou University of Vermont, Rubenstein School of Natural Resources State and local planning agencies are increasingly

More information

FireFamily Plus User s Guide

FireFamily Plus User s Guide United States Department of Agriculture Forest Service Rocky Mountain Research Station General Technical Report RMRS-GTR-67-WWW September 2000 FireFamily Plus User s Guide Version 2.0 Larry Bradshaw Erin

More information

Internet Engineering Jacek Mazurkiewicz, PhD

Internet Engineering Jacek Mazurkiewicz, PhD Internet Engineering Jacek Mazurkiewicz, PhD Softcomputing Part 11: SoftComputing Used for Big Data Problems Agenda Climate Changes Prediction System Based on Weather Big Data Visualisation Natural Language

More information

Arizona Climate Summary April 2018 Summary of conditions for March 2018

Arizona Climate Summary April 2018 Summary of conditions for March 2018 Arizona Climate Summary April 2018 Summary of conditions for March 2018 March 2018 Temperature and Precipitation Summary March 1 st 16 th : March began with rainfall in southeastern Arizona as the tail

More information

Twitter s Effectiveness on Blackout Detection during Hurricane Sandy

Twitter s Effectiveness on Blackout Detection during Hurricane Sandy Twitter s Effectiveness on Blackout Detection during Hurricane Sandy KJ Lee, Ju-young Shin & Reza Zadeh December, 03. Introduction Hurricane Sandy developed from the Caribbean stroke near Atlantic City,

More information

Climate Dataset: Aitik Closure Project. November 28 th & 29 th, 2018

Climate Dataset: Aitik Closure Project. November 28 th & 29 th, 2018 1 Climate Dataset: Aitik Closure Project November 28 th & 29 th, 2018 Climate Dataset: Aitik Closure Project 2 Early in the Closure Project, consensus was reached to assemble a long-term daily climate

More information

W. Douglas Wewer 1,2 *, Michael J. Jenkins 1

W. Douglas Wewer 1,2 *, Michael J. Jenkins 1 A PRACTITIONER S PERSPECTIVE: PROVIDING TIMELY AND RELEVANT MOUNTAIN WEATHER INFORMATION TO AVALANCHE WORKERS AT SNOWBASIN, A SUN VALLEY RESORT AND THE FOREST SERVICE UTAH AVALANCHE CENTER W. Douglas Wewer

More information

Midterm: CS 6375 Spring 2015 Solutions

Midterm: CS 6375 Spring 2015 Solutions Midterm: CS 6375 Spring 2015 Solutions The exam is closed book. You are allowed a one-page cheat sheet. Answer the questions in the spaces provided on the question sheets. If you run out of room for an

More information

MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October,

MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October, MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October, 23 2013 The exam is closed book. You are allowed a one-page cheat sheet. Answer the questions in the spaces provided on the question sheets. If you run

More information

Geospatial Fire Behavior Modeling App to Manage Wildfire Risk Online. Kenyatta BaRaKa Jackson US Forest Service - Consultant

Geospatial Fire Behavior Modeling App to Manage Wildfire Risk Online. Kenyatta BaRaKa Jackson US Forest Service - Consultant Geospatial Fire Behavior Modeling App to Manage Wildfire Risk Online Kenyatta BaRaKa Jackson US Forest Service - Consultant Fire Behavior Modeling and Forest Fuel Management Modeling Fire Behavior is an

More information

Proceedings, International Snow Science Workshop, Innsbruck, Austria, 2018

Proceedings, International Snow Science Workshop, Innsbruck, Austria, 2018 RELEASE OF AVALANCHES ON PERSISTENT WEAK LAYERS IN RELATION TO LOADING EVENTS IN COLORADO, USA Jason Konigsberg 1, Spencer Logan 1, and Ethan Greene 1 1 Colorado Avalanche Information Center, Boulder,

More information

Fire danger. The skill provided by ECMWF ensemble prediction system. Francesca Di Giuseppe and Claudia Vitolo Forecast Department, ECMWF

Fire danger. The skill provided by ECMWF ensemble prediction system. Francesca Di Giuseppe and Claudia Vitolo Forecast Department, ECMWF Fire danger The skill provided by ECMWF ensemble prediction system Link to slides: https://goo.gl/qm15fk Francesca Di Giuseppe and Claudia Vitolo Forecast Department, ECMWF ECMWF August 2, 2016 Fire forecast

More information

Lake Tahoe Watershed Model. Lessons Learned through the Model Development Process

Lake Tahoe Watershed Model. Lessons Learned through the Model Development Process Lake Tahoe Watershed Model Lessons Learned through the Model Development Process Presentation Outline Discussion of Project Objectives Model Configuration/Special Considerations Data and Research Integration

More information

LAB 19. Lab 19. Differences in Regional Climate: Why Do Two Cities Located at the Same Latitude and Near a Body of Water Have Such Different Climates?

LAB 19. Lab 19. Differences in Regional Climate: Why Do Two Cities Located at the Same Latitude and Near a Body of Water Have Such Different Climates? Lab Handout Lab 19. Differences in Regional Climate: Why Do Two Cities Located at the Same Latitude and Near a Body of Water Have Such Different Climates? Introduction Weather describes the current atmospheric

More information

NIDIS Intermountain West Drought Early Warning System December 18, 2018

NIDIS Intermountain West Drought Early Warning System December 18, 2018 NIDIS Intermountain West Drought Early Warning System December 18, 2018 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations. From top to bottom,

More information

CLASSIFICATION NAIVE BAYES. NIKOLA MILIKIĆ UROŠ KRČADINAC

CLASSIFICATION NAIVE BAYES. NIKOLA MILIKIĆ UROŠ KRČADINAC CLASSIFICATION NAIVE BAYES NIKOLA MILIKIĆ nikola.milikic@fon.bg.ac.rs UROŠ KRČADINAC uros@krcadinac.com WHAT IS CLASSIFICATION? A supervised learning task of determining the class of an instance; it is

More information

Hourly Precipitation Data Documentation (text and csv version) February 2016

Hourly Precipitation Data Documentation (text and csv version) February 2016 I. Description Hourly Precipitation Data Documentation (text and csv version) February 2016 Hourly Precipitation Data (labeled Precipitation Hourly in Climate Data Online system) is a database that gives

More information

NIDIS Intermountain West Drought Early Warning System February 6, 2018

NIDIS Intermountain West Drought Early Warning System February 6, 2018 NIDIS Intermountain West Drought Early Warning System February 6, 2018 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations. From top to bottom,

More information

1 What Is Climate? TAKE A LOOK 2. Explain Why do areas near the equator tend to have high temperatures?

1 What Is Climate? TAKE A LOOK 2. Explain Why do areas near the equator tend to have high temperatures? CHAPTER 17 1 What Is Climate? SECTION Climate BEFORE YOU READ After you read this section, you should be able to answer these questions: What is climate? What factors affect climate? How do climates differ

More information

DATA SOURCES AND INPUT IN GIS. By Prof. A. Balasubramanian Centre for Advanced Studies in Earth Science, University of Mysore, Mysore

DATA SOURCES AND INPUT IN GIS. By Prof. A. Balasubramanian Centre for Advanced Studies in Earth Science, University of Mysore, Mysore DATA SOURCES AND INPUT IN GIS By Prof. A. Balasubramanian Centre for Advanced Studies in Earth Science, University of Mysore, Mysore 1 1. GIS stands for 'Geographic Information System'. It is a computer-based

More information

WeatherHawk Weather Station Protocol

WeatherHawk Weather Station Protocol WeatherHawk Weather Station Protocol Purpose To log atmosphere data using a WeatherHawk TM weather station Overview A weather station is setup to measure and record atmospheric measurements at 15 minute

More information

AN INTERNATIONAL SOLAR IRRADIANCE DATA INGEST SYSTEM FOR FORECASTING SOLAR POWER AND AGRICULTURAL CROP YIELDS

AN INTERNATIONAL SOLAR IRRADIANCE DATA INGEST SYSTEM FOR FORECASTING SOLAR POWER AND AGRICULTURAL CROP YIELDS AN INTERNATIONAL SOLAR IRRADIANCE DATA INGEST SYSTEM FOR FORECASTING SOLAR POWER AND AGRICULTURAL CROP YIELDS James Hall JHTech PO Box 877 Divide, CO 80814 Email: jameshall@jhtech.com Jeffrey Hall JHTech

More information

Spatial Optimization of CoCoRAHS Network in Tennessee. Joanne Logan Department of Biosystems Engineering and Soil Science University of Tennessee

Spatial Optimization of CoCoRAHS Network in Tennessee. Joanne Logan Department of Biosystems Engineering and Soil Science University of Tennessee Spatial Optimization of CoCoRAHS Network in Tennessee Joanne Logan Department of Biosystems Engineering and Soil Science University of Tennessee Abstract CoCoRaHS (Community Collaborative Rain, Hail and

More information

WindNinja Tutorial 3: Point Initialization

WindNinja Tutorial 3: Point Initialization WindNinja Tutorial 3: Point Initialization 6/27/2018 Introduction Welcome to WindNinja Tutorial 3: Point Initialization. This tutorial will step you through the process of downloading weather station data

More information

NIDIS Intermountain West Drought Early Warning System February 12, 2019

NIDIS Intermountain West Drought Early Warning System February 12, 2019 NIDIS Intermountain West Drought Early Warning System February 12, 2019 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations. From top to bottom,

More information

NIDIS Intermountain West Regional Drought Early Warning System February 7, 2017

NIDIS Intermountain West Regional Drought Early Warning System February 7, 2017 NIDIS Drought and Water Assessment NIDIS Intermountain West Regional Drought Early Warning System February 7, 2017 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS,

More information

NIDIS Intermountain West Drought Early Warning System July 18, 2017

NIDIS Intermountain West Drought Early Warning System July 18, 2017 NIDIS Drought and Water Assessment NIDIS Intermountain West Drought Early Warning System July 18, 2017 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet

More information

AERMOD Sensitivity to AERSURFACE Moisture Conditions and Temporal Resolution. Paper No Prepared By:

AERMOD Sensitivity to AERSURFACE Moisture Conditions and Temporal Resolution. Paper No Prepared By: AERMOD Sensitivity to AERSURFACE Moisture Conditions and Temporal Resolution Paper No. 33252 Prepared By: Anthony J Schroeder, CCM Managing Consultant TRINITY CONSULTANTS 7330 Woodland Drive Suite 225

More information

Making Our Cities Safer: A Study In Neighbhorhood Crime Patterns

Making Our Cities Safer: A Study In Neighbhorhood Crime Patterns Making Our Cities Safer: A Study In Neighbhorhood Crime Patterns Aly Kane alykane@stanford.edu Ariel Sagalovsky asagalov@stanford.edu Abstract Equipped with an understanding of the factors that influence

More information

NIDIS Intermountain West Drought Early Warning System April 16, 2019

NIDIS Intermountain West Drought Early Warning System April 16, 2019 NIDIS Intermountain West Drought Early Warning System April 16, 2019 Precipitation The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations. From top to bottom,

More information

NIDIS Intermountain West Drought Early Warning System April 18, 2017

NIDIS Intermountain West Drought Early Warning System April 18, 2017 1 of 11 4/18/2017 3:42 PM Precipitation NIDIS Intermountain West Drought Early Warning System April 18, 2017 The images above use daily precipitation statistics from NWS COOP, CoCoRaHS, and CoAgMet stations.

More information

CHAPTER 4: PREDICTION AND ESTIMATION OF RAINFALL DURING NORTHEAST MONSOON

CHAPTER 4: PREDICTION AND ESTIMATION OF RAINFALL DURING NORTHEAST MONSOON 43 CHAPTER 4: PREDICTION AND ESTIMATION OF RAINFALL DURING NORTHEAST MONSOON After analyzing the weather patterns and forecast of seasonal rainfall for Cauvery delta region, a method to predict the dry

More information

Spatial Effects on Current and Future Climate of Ipomopsis aggregata Populations in Colorado Patterns of Precipitation and Maximum Temperature

Spatial Effects on Current and Future Climate of Ipomopsis aggregata Populations in Colorado Patterns of Precipitation and Maximum Temperature A. Kenney GIS Project Spring 2010 Amanda Kenney GEO 386 Spring 2010 Spatial Effects on Current and Future Climate of Ipomopsis aggregata Populations in Colorado Patterns of Precipitation and Maximum Temperature

More information

About places and/or important events Landmarks Maps How the land is, hills or flat or mountain range Connected to maps World Different countries

About places and/or important events Landmarks Maps How the land is, hills or flat or mountain range Connected to maps World Different countries What do you think you know about geography? About places and/or important events Landmarks Maps How the land is, hills or flat or mountain range Connected to maps World Different countries What do you

More information

Arizona Climate Summary November 2015 Summary of conditions for October 2015

Arizona Climate Summary November 2015 Summary of conditions for October 2015 October 2015 Temperature and Precipitation Summary Arizona Climate Summary November 2015 Summary of conditions for October 2015 October 1 st 14 th : October began with high pressure over the southwest,

More information

Global Climates. Name Date

Global Climates. Name Date Global Climates Name Date No investigation of the atmosphere is complete without examining the global distribution of the major atmospheric elements and the impact that humans have on weather and climate.

More information

Climate Change and Biomes

Climate Change and Biomes Climate Change and Biomes Key Concepts: Greenhouse Gas WHAT YOU WILL LEARN Biome Climate zone Greenhouse gases 1. You will learn the difference between weather and climate. 2. You will analyze how climate

More information

Globally Estimating the Population Characteristics of Small Geographic Areas. Tom Fitzwater

Globally Estimating the Population Characteristics of Small Geographic Areas. Tom Fitzwater Globally Estimating the Population Characteristics of Small Geographic Areas Tom Fitzwater U.S. Census Bureau Population Division What we know 2 Where do people live? Difficult to measure and quantify.

More information

Weather and Climate of the Rogue Valley By Gregory V. Jones, Ph.D., Southern Oregon University

Weather and Climate of the Rogue Valley By Gregory V. Jones, Ph.D., Southern Oregon University Weather and Climate of the Rogue Valley By Gregory V. Jones, Ph.D., Southern Oregon University The Rogue Valley region is one of many intermountain valley areas along the west coast of the United States.

More information

DROUGHT ASSESSMENT USING SATELLITE DERIVED METEOROLOGICAL PARAMETERS AND NDVI IN POTOHAR REGION

DROUGHT ASSESSMENT USING SATELLITE DERIVED METEOROLOGICAL PARAMETERS AND NDVI IN POTOHAR REGION DROUGHT ASSESSMENT USING SATELLITE DERIVED METEOROLOGICAL PARAMETERS AND NDVI IN POTOHAR REGION Researcher: Saad-ul-Haque Supervisor: Dr. Badar Ghauri Department of RS & GISc Institute of Space Technology

More information

APPEND AND MERGE. Colorado average snowfall amounts - Aggregated by County and Month

APPEND AND MERGE. Colorado average snowfall amounts - Aggregated by County and Month APPEND AND MERGE Climate change affects the environment is many ways. This past winter produced little snow in California which is likely to compound their drought. The lack of snow also negatively impacts

More information

Ad Placement Strategies

Ad Placement Strategies Case Study 1: Estimating Click Probabilities Tackling an Unknown Number of Features with Sketching Machine Learning for Big Data CSE547/STAT548, University of Washington Emily Fox 2014 Emily Fox January

More information

Predicting Floods in North Central Province of Sri Lanka using Machine Learning and Data Mining Methods

Predicting Floods in North Central Province of Sri Lanka using Machine Learning and Data Mining Methods Thilakarathne & Premachandra Predicting Floods in North Central Province of Sri Lanka using Machine Learning and Data Mining Methods H. Thilakarathne 1, K. Premachandra 2 1 Department of Physical Science,

More information

For most crops, 30% risk is probably too much, but this is for you decide.

For most crops, 30% risk is probably too much, but this is for you decide. Frost Probabilities Compiled by Dustin Blakey Farm Advisor, Inyo and Mono Counties Frost and Freeze Dates Every year frost will occur at a different date. It is impossible to know exactly when the last

More information

Comparing Climate Features

Comparing Climate Features Long-Term Projects Comparing Climate Features A graph of the monthly temperatures and amounts of precipitation for a region is called a climatograph. Climatographs can be used to compare the climates of

More information

Precipitation. Standardized Precipitation Index. NIDIS Intermountain West Regional Drought Early Warning System January 3, 2017

Precipitation. Standardized Precipitation Index. NIDIS Intermountain West Regional Drought Early Warning System January 3, 2017 1/3/2017 NIDIS Drought and Water Assessment NIDIS Intermountain West Regional Drought Early Warning System January 3, 2017 Precipitation The images above use daily precipitation statistics from NWS COOP,

More information

Applications: Introduction Task 1: Introduction to ArcCatalog Task 2: Introduction to ArcMap Challenge Question References

Applications: Introduction Task 1: Introduction to ArcCatalog Task 2: Introduction to ArcMap Challenge Question References CHAPTER 1 INTRODUCTION 1.1 GIS? 1.1.1 Components of a GIS 1.1.2 A Brief History of GIS 1.1.3 GIS Software Products Box 1.1 A List of GIS Software Producers and Their Main Products 1.2 GIS Applications

More information

Climate Outlook through 2100 South Florida Ecological Services Office Vero Beach, FL September 9, 2014

Climate Outlook through 2100 South Florida Ecological Services Office Vero Beach, FL September 9, 2014 Climate Outlook through 2100 South Florida Ecological Services Office Vero Beach, FL September 9, 2014 Short Term Drought Map: Short-term (

More information

ClimateBC version history

ClimateBC version history ClimateBC version history ClimateBC v5.60 (August 31, 2018) Improvements Historical monthly data from the Climate Research Unit (CRU ts4.01) for the years 1999-2016 have been replaced by our newly developed

More information

A Support Vector Regression Model for Forecasting Rainfall

A Support Vector Regression Model for Forecasting Rainfall A Support Vector Regression for Forecasting Nasimul Hasan 1, Nayan Chandra Nath 1, Risul Islam Rasel 2 Department of Computer Science and Engineering, International Islamic University Chittagong, Bangladesh

More information

USE OF SATELLITE INFORMATION IN THE HUNGARIAN NOWCASTING SYSTEM

USE OF SATELLITE INFORMATION IN THE HUNGARIAN NOWCASTING SYSTEM USE OF SATELLITE INFORMATION IN THE HUNGARIAN NOWCASTING SYSTEM Mária Putsay, Zsófia Kocsis and Ildikó Szenyán Hungarian Meteorological Service, Kitaibel Pál u. 1, H-1024, Budapest, Hungary Abstract The

More information

http://www.wrcc.dri.edu/csc/scenic/ USER GUIDE 2017 Introduction... 2 Overview Data... 3 Overview Analysis Tools... 4 Overview Monitoring Tools... 4 SCENIC structure and layout... 5... 5 Detailed Descriptions

More information

Climate Variables for Energy: WP2

Climate Variables for Energy: WP2 Climate Variables for Energy: WP2 Phil Jones CRU, UEA, Norwich, UK Within ECEM, WP2 provides climate data for numerous variables to feed into WP3, where ESCIIs will be used to produce energy-relevant series

More information

Arizona Climate Summary August 2013

Arizona Climate Summary August 2013 Arizona Climate Summary August 2013 Summary of conditions for July 2013 July 2013 Temperature and Precipitation Summary July 1 st 16 th : To start off the month of July, a high pressure system came in

More information

Statistical Forecast of the 2001 Western Wildfire Season Using Principal Components Regression. Experimental Long-Lead Forecast Bulletin

Statistical Forecast of the 2001 Western Wildfire Season Using Principal Components Regression. Experimental Long-Lead Forecast Bulletin Statistical Forecast of the 2001 Western Wildfire Season Using Principal Components Regression contributed by Anthony L. Westerling 1, Daniel R. Cayan 1,2, Alexander Gershunov 1, Michael D. Dettinger 2

More information

Exercie 5 Preparing GIS data for simulation with FARSITE REM407 GIS Applications in Fire Ecology and Management

Exercie 5 Preparing GIS data for simulation with FARSITE REM407 GIS Applications in Fire Ecology and Management Exercie 5 Preparing GIS data for simulation with FARSITE REM407 GIS Applications in Fire Ecology and Management Assignment: Make maps of the Fuelmodel layer and the clipped Canopy cover layer and submit

More information

Attachment E: CADP Design Shadow Analysis

Attachment E: CADP Design Shadow Analysis Attachment E: CADP Design Shadow Analysis June 6, 2016 TO: Don Lewis San Francisco Planning Department 1650 Mission Street, Suite 400 San Francisco, CA 94103 SUBJECT: 2060 Folsom Street 17 th & Folsom

More information

Application of Text Mining for Faster Weather Forecasting

Application of Text Mining for Faster Weather Forecasting International Journal of Computer Engineering and Information Technology VOL. 8, NO. 11, November 2016, 213 219 Available online at: www.ijceit.org E-ISSN 2412-8856 (Online) Application of Text Mining

More information