Daily Casualties - Gaza
The numbers we publish here are not fully representative of the human toll of Israel's actions in Gaza. Read more about other factors you should consider.
This dataset provides daily values for those killed and injured in the Gaza Strip since October 7th, 2023.
Usage
The dataset is available as a single JSON array with an object for each report date. You can also download in CSV format below.
Minified
https://data.techforpalestine.org/api/v2/casualties_daily.min.json
Unminified
https://data.techforpalestine.org/api/v2/casualties_daily.json
The file is updated in the morning (eastern time).
Report Fields
Each daily report in this JSON array will have the following fields for Gaza.
(optional fields will be omitted from the JSON if there is no value reported)
| field name | value | optional |
|---|---|---|
| report_date | date in YYYY-MM-DD | no |
| report_source | one of mohtel, gmotel, unocha or missing (see below) | no |
| report_period | one of 24, 48, or 0, representing the hours length of the reporting period | no |
| massacres_cum | cumulative number of multiple fatality events | yes |
| ext_massacres_cum | same as massacres_cum but extrapolated (see below) | no |
| killed | killed persons on the given report date | yes |
| killed_cum | cumulative number of killed persons to the report date | yes |
| killed_children_cum | cumulative number of children killed to the report date | yes |
| killed_women_cum | cumulative number of women killed to the report date | yes |
| ext_killed | same as killed but extrapolated (see below) | no |
| ext_killed_cum | same as killed_cum but extrapolated (see below) | no |
| ext_killed_children_cum | same as killed_children_cum but extrapolated (see below) | no |
| ext_killed_women_cum | same as killed_women_cum but extrapolated (see below) | no |
| injured | injured persons on the given report date | yes |
| injured_cum | cumulative number of injured persons to the report date | yes |
| ext_injured | injured persons on the given report date | no |
| ext_injured_cum | cumulative number of injured persons to the report date | no |
| civdef_killed_cum | cumulative number of emergency services killed to date | yes |
| ext_civdef_killed_cum | same as civdef_killed_cum but extrapolated (see below) | no |
| med_killed_cum | cumulative number of medical personnel killed to the report date | yes |
| ext_med_killed_cum | same as med_martyed_cum but extrapolated (see below) | no |
| press_killed_cum | cumulative number of journalists killed to the report date | yes |
| ext_press_killed_cum | same as press_martyed_cum but extrapolated (see below) | no |
Update Interval
We update this dataset everyday, but some values are updated less frequently (typically at least once a week):
killed_children_cumkilled_women_cumcivdef_killed_cummed_killed_cumpress_killed_cum
If there's no updated value for the given report date, the last reported value is used (since all of these fields are cumulative). This necessarily also applies to their corresponding ext_-prefixed fields.
Daily Sources
There are four source values for daily casualty reports we use to build the time series:
mohtel- Gaza's Ministry of Health Telegram channelgmotel- Gaza's Government Media Office Telegram channelmissing- No official report was available for the given date (some reasons below)
The primary source is the Ministry of Health. The numbers they report only include those they can connect directly to an act of war. For example:
[The] Ministry of Health has a policy of recording only direct casualties of war, such as those caused by missile strikes or war-related injuries. There are other deaths indirectly caused by the war, but we do not add them to the list of martyrs. For example, children who die due to malnutrition, lack of care, or because their mothers gave birth under difficult conditions and carried them in poor health, resulting in the birth of an underdeveloped infant who dies after a few days or lacked proper feeding, these cases are documented but not recorded as martyrs.
For more detail on how the ministry gathers their numbers, read this Dropsite interview with Dr. Zaher al-Wahaidi from which the above excerpt is taken.
It's important to underscore that these sources do not account for the knock-on affects of these attacks. You can read more about all the other considerations that go into estimating the true toll in Gaza.
Our methodology is to prioritize official sources for this dataset in order to best align the reporting time periods (mohtel falling back to gmotel). When communications were cut to Gaza, gmotel would provide reporting but in some cases neither would. Reports were also stopped during the humanitarian pause.
Extrapolated (ext_ fields)
Since official numbers weren't always available, and it can be useful to have a timeseries where all dates have values, we're providing the same official fields in extrapolated form using the following methodology:
- if the missing field was a cumulative one and we had an official report for a daily killed or injury count, we calculate the cumulative using the daily increment
- if the missing field was a daily increment and we had cumulative counts, we subtracted the reported cumulative count from the prior period for the missing daily count
- if we were missing both sets of numbers for a given reporting period we average the difference between surrounding periods