Background Information
Pollution, Global warming, and climate change have all had a negative effect on global weather patterns thus affecting certain industries like agriculture. With erratic and unpredictable amounts of rain, high summer temperatures and extreme winters being registered across the globe, farmers today are adopting modern technologies to help analyze historical data and reduce risks associated to climate change and weather patterns (Nelson, et al., 2009). Data mining is among the most powerful tools available today to analyze bulk data using powerful computer software programs. The programs analyze, organized and filter the data using algorithms to help determine important links and trends on the bulk data sets which may not be detectable to the naked eye in numeric form. Converting this data into visuals displays of the data allows for better analysis, allowing the stakeholders to make better decisions for the future. The report will focus on strategies that can be adapted by AFM Development Pvt Ltd to analysis past rainfall data so as to predict future rainfall. This is an important requirement in the agriculture industry as it allows the farmers identify the amount of rainfall that can be expected. This in tern allows farmers to determine the type of crops to be planted as well as how to managed existing crops (Turlapaty, 2010). Also known as data mining, this strategy of converting numeric data to visuals also allow organizations and individuals to determine important trends and patterns which may otherwise not noticeable when viewing raw numeric data. To achieve the best results the data shall be mined using tableau, orange and excel software which will to convert the data into visuals which can then be studied for better understanding and prediction of future rainfall patterns
Data shall be collected for the BALLARAT AERODROME (89002) station which was established in 1908. Missing data shall be filled with the average amount of rainfall registered during the month over the entire period to eliminate data analysis errors. The secure a clear understanding related to the data, it is important to use 3 data mining sets (Zani, Cerioli, Riani, & Vichi, 2007). Set 1 will see a visualization of all data since 1908. Set 2 shall see a simulation of data over the past 10 years and Set 3 will be a simulation of data for the past 36 months. It is critical to converting the data into three sets as this will help determine important trends which may have occurred at Ballarat Aerodrome over the past 100 years. It’s important to collect and report the data which will help AFM Development Pvt Ltd determine important trends in rainfall patterns which have occurred in the past thus helping determine future rainfall patterns. This data can then be used by the company to determine the amount of yield and production which can be expected in the coming year as well as determine other important information and statistics such as disease and damage which may occur as a result on too low or too high rainfall (IAHS, 2006). This makes it important to analyze available data using different tools which will assist analysis to determine common and recurring trends on the data. This information can then be used to deliver accurate prediction of rainfall the next year.
Reporting / Dashboards
To secure the best understanding linked to the rainfall patterns experienced at the Ballarat Aerodrome (89002) area, it is important for the data analyst to first run the data through data mining tools which will help report on historical trends recorded on the data over the past 100 years,10 years and over the past 12 months. This is important as it will allow the data analyst to identify patterns on rail fall patterns occurring in the region over different time frames. It is also important to run the data through different data mining tools which will help confirm the data being secured is accurate and can be used for further investigations (Pyle, 1999). For this, we shall be using Tableau, Rapid miner and MS excel which shall be used to run data simulation which will help translate the data into visuals which can be used to predict the rainfall among and patterns to determine the amount of rainfall that can be expected in the next year’s rainfall
To analyses a 100-year cycle of rainfall data we shall use rapid miner and MS excel which will help produce rainfall scatter data as well as the frequency at which certain amounts of rain are recorded. This is important as it will help determine a frequently recurring rainfall average over the 100-year cycle. MS Excel will be used to create a bar and line chart which can be used to follow the rainfall pattern occurring over the past 100 years. This will help the analyst determine years on which rainfall tends to increase and reduce rapidly which can be used to predict the following year’s rainfall amount (Soukup & Davidson, 2002). The following analysis was produced from the 100-year data simulation which can be used to determine rainfall frequency and the amount over the 100 year period.
Analyzing the data clearly demonstrates that rainfall patterns are relatedly stable in the Ballarat Aerodrome region with major rainfall amount fluctuation occurring over 5-year cycles. This results in delivering adequate time for farmers to prepare for low or excessive rainfall during the period. Reviewing the 100-year data also reveals that the region has never experienced a major drought with the lowest amount of rain registered in the region is 300 mm per annum and the average being between 650 -680 mm per annum. The data does show a sudden spite in 1939, 1950 and a 1975 when rainfall was above 900mm per annum which is not a major factor of concern since it still falls within the expected rainfall amounts for the Cotton crop (Dehnavi, 2015). Major concerns are linked to low rainfall due to the cotton not setting fruit which results in low yield resulting in losses of revenue from cotton wool and seed sales.
Research
The yearly analysis is important towards determining the monthly rainfall patterns which can be used to determine months when irrigation may be required to retain cotton plant health and development. To perform this analysis a 10 year set of data requires being analyses to determine the regularity of rainfall and frequency of erratic rain amounts (Nettleton, 2014). This will help determine the driest months thus allowing for alternative irrigation methods to be put in place.
By reviewing the above data it’s clear that the Ballarat Aerodrome region tends to register low rainfall January and February where our average rainfall levels can be expected in March, April June, October, November, and December. The highest rainfall comes in July, August, and September. Armed with this data we cannot move to review the monthly analyses using tableau which would help pinpoint specific amounts of rainfall registered during the past year to determine if the rain patents fallow average expectancies so as to make a prediction for the next year (Abou-Nasr, Lessmann, Stahlbock, & Weiss, 2014). On the 10-year series, patterns shall be identified to help to make the next years prediction based on past data. The above charts clearly demonstrate each month has registers freak rainfall amounts but the data also demonstrates reduction after which high rainfall amounts can be expected over a 10-year cycle. Presently Ballarat Aerodrome is experiencing the lower levels of rain thus its likely observe high amounts of rainfall in the next year.
The ten-year rainfall cycle clearly demonstrates a repeating pattern in which therein fall amounts register freak high after which they gradually reduce then register another high signaling the start of another cycle. This is aimed at identifying important indicator as these allow the analyst to determine recurring pattern which can be used to develop accurate prediction of the next year’s rain pattern (Shurma & Sunderam.V.K, 2004).
The yearly analysis also demonstrates the same patterns on a monthly scale which can be considered as a confirmation of the data research and patterns thus delivering an accurate prediction of the next year’s rainfall prediction
Again the monthly data pattern can be observed on the January and February data but at a monthly scale thus allowing for a vivid prediction related to the number of years the rainfall patterns occur. According to the above analysis, rain fall patterns tend to occur over a 5 to 6-year cycle (Knapp, 2002). Reviewing the data also reveals Ballarat Aerodrome is at the end of a rainfall cycle which means the region has registered5 year lows thus it can be expected that high rainfall amounts can be registered during the next year.
100 year Cycle analysis – Century Data
This data can be confirmed using tableau models to report last year’s rainfall amounts which have been recorded at lows averaging 1.5 millimeters thus signaling an impending high amount of rain during the next one or two years (Osborne, 2012).
Armed with the above data a clear prediction related to expected rainfall amounts during the next year can be proposed. It is clear the Ballarat Aerodrome has a stable rainfall pattern but one which trend to follow a certain pattern on a 100, 10 and yearly cycle. This is important as the data cannot be translated accurately to predict rainfall amounts that can be expected over the next 12 months. Cotton is a drought resistant crop thus resulting in the crop being able to cope with low rainfall but high amounts of rainfall can easily damage the crop presently (Kumar & Varaiya, 2015).
The analysis clearly shows Ballarat Aerodrome is at the end of a low rain cycle and the region can expect erratic high rains in the next 12-24 months. It’s not possible to determine exactly when the high will occur but an 80% chance towards high irruptive rainfall being registered next year is highly likely. This may be postponed to the next year thus AFM Development Pvt Ltd will need to prepare to protect the cotton crop from excessive rain and water logging which may damage the crops roots resulting in loss of plants.
In a circumstance that rain may not register sudden high; it is likely to register the same amount of rain as in 2017. Low rain fall does not have a major effect on cotton crop as the crop goes in to dormancy (Bajaj, 2012). On the other hand protection needs to be provided for cotton crops during the high rain period. AFM Development Pvt Ltd is therefore advised to prepare for high rains in the next 12 months which could damage the cotton crop root system resulting in permanent crop damage (Jin & Lin, 2011). The company will need to develop flood and waterlogging countermeasures to ensure the maximum amount of water is draining away from the crop and farm to prevent crop root damage.
The erratic weather pattern is likely to last just for one year before stabilizing to average rainfall amounts but it’s important to understand that the excessive rainfall could result in AFM Development Pvt Ltd registering major losses due to crop damage (Midmore, 2015). Mitigation plans for waterlogging and drainage must, therefore, be implemented immediately.
10 Year analysis – Decade Data
Conclusion
By performing data mining and analysis of historical data AFM Development Pvt Ltd is capable of collecting important information related to rainfall patterns at the Ballarat Aerodrome area. This knowledge can then be used to determine important trends the Ballarat Aerodrome region can expect during the next 12 to 24 months thus allowing businesses and industry to prepare in advance and help prevent unexpected damage. Data analysis is a powerful tool which can be used by businesses and organization to determine important patterns which would otherwise go undetected when presented with numeric datasets.
Re: 12-24 MONTH RAINFALL FORECAST – BALLARAT AERODROME (89002)
After performing an in-depth review of the historical rainfall data for Ballarat Aerodrome (89002) retrieved from the https://www.bom.gov.au/climate/data/. We are pleased to present to above finding with regard to the expected rainfall pattern for the next 12-24 months
The Data has been broken down and run through data mining tools to deliver simplified and translatable information which can be used by the company staff and management to study predict rainfall patterns.
After analyzing the data we have come the following conclusion:
Rainfall Forecast – Heavy Rainfall Expected (12-24 months).
Recommendations: After analysis of the data we forecast heavy rain during the next 12-24 months. AFM Development Pvt Ltd is therefore advised to consider pre-cautionary measures to protect cotton crop from water logging which can damage the crops root system and loss of cotton plants.
We request you to also review the data and raise queries and points which can also be used to develop alternative simulation so as refine the finding and secure a more accurate rainfall forecast for the next 12-24 months.
References:
Abou-Nasr, M., Lessmann, S., Stahlbock, R., & Weiss, G. M. (2014). Real World Data Mining Applications. Springer.
Bajaj, Y. (2012). Cotton. New Delhi: Springer Science & Business Media.
Dehnavi, S. (2015). Water Resources Availability, National Food Security Strategies and Farmers’ Reactions in Darab, Iran. kassel university press .
IAHS. (2006). Climate Variability and Change–hydrological Impacts. Flow Regimes from International Experimental and Network Data (Project).
Jin, D., & Lin, S. (2011). Advances in Computer Science, Intelligent Systems and Environment, Volume 2. Springer Science & Business Media.
Knapp, B. (2002). Elements of Geographical Hydrology. Routledge.
Kumar, P. R., & Varaiya, P. (2015). Stochastic Systems: Estimation, Identification, and Adaptive Control. SIAM.
Midmore, D. J. (2015). Principles of Tropical Horticulture. Oxfordshire: CABI.
Nelson, G. C., Rosegrant, M. W., Koo, J., Robertson, Richard, S., Timothy, et al. (2009). Climate Change: Impact on Agriculture and Costs of Adaptation. Washington DC: Intl Food Policy Res Ins.
Nettleton, D. (2014). Commercial Data Mining: Processing, Analysis and Modeling for Predictive Analytics Projects. Elsevier.
Osborne, P. L. (2012). Tropical Ecosystems and Ecological Concepts. Cambridge University Press.
Pyle, D. (1999). Data Preparation for Data Mining, Volume 1. Morgan Kaufmann.
Shurma, R., & Sunderam.V.K. (2004). Water Resources Development and Management. Mittal Publications.
Soukup, T., & Davidson, I. (2002). Visual Data Mining: Techniques and Tools for Data Visualization and Mining. John Wiley & Sons.
Turlapaty, A. C. (2010). Application of Pattern Recognition and Adaptive DSP Methods for Spatio-temporal Analysis of Satellite Based Hydrological Datasets. Boca Roton: Universal-Publishers.
Zani, S., Cerioli, A., Riani, M., & Vichi, M. (2007). Data Analysis, Classification and the Forward Search: Proceedings of the Meeting of the Classification and Data Analysis Group (CLADAG) of the Italian Statistical Society, University of Parma, June 6-8, 2005. Springer Science & Business Media.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
