"FIRMS" Inquiry Regarding Hotspot Data Characteristics and Missing Values

[mariomanroe]

I am a student in the Data Science program at Institut Teknologi Sumatera. I am currently conducting research on hotspot patterns in Sumatra, Indonesia, using data obtained from the FIRMS NASA platform. My research focuses on forecasting daily hotspot occurrences, and I have encountered some challenges with missing values in the dataset.

Specifically, I noticed that there are certain dates in the time series data where no hotspots are recorded, resulting in missing values. Since my research involves daily forecasting, having a continuous and sequential date range is crucial. However, the absence of data on some dates creates gaps that need to be addressed for accurate modeling.

To ensure the accuracy and integrity of my research, I would like to better understand the characteristics of the hotspot data provided by FIRMS, especially in relation to these missing values. My questions are as follows:

1. Nature of Missing Values: Could you please clarify whether the missing values in the dataset indicate a lack of detected hotspots on those specific dates, or if they represent periods when data might not have been captured or recorded?

2. Handling Missing Values: Given the nature of my research, which involves daily forecasting of hotspots and requires a continuous date range, do you have any recommendations or best practices for handling these missing values within the dataset?

Understanding these details is crucial for ensuring that my analysis and forecasting models are based on accurate and representative data.

I greatly appreciate any insights or guidance you can provide on this matter. Thank you in advance for your assistance, and I look forward to your response.

I appologize if this has already been answered. I appreciate any guidance on this.

[LANCE & FIRMSx - ddavies]

Regarding question 1: Data losses due to outages in the MODIS and VIIRS record, where data was not collected, are available at https://modaps.modaps.eosdis.nasa.gov/services/production/outages.html. This resource can be used to compare the observed periods of missing data relative to these outage periods.

Regarding question 2: We provide the data as is and can provide information on the data itself but we can't advise on the use of the your hotspot modeling forecast. There are many factors that effect whether a hotspot will occur and you will need to look at the scientific literature to decide how to move forward.

All the best with your studies!