Introducing Two New Gauge Enhanced Satellite Precipitation Datasets - CHIMES and CHIRPS-v3

The human and economic costs of hydro-climatic shocks are massive and increasing. Between 2015 and 2019, the impacts and losses associated with droughts, floods, and hurricanes surged through the interaction of increasing exposure and more extreme weather and climate (Funk 2021 Drought Flood Fire). In 2023, the Aon-Benfield Reinsurance Company reported that cyclones, floods, and droughts resulted in $229 billion (USD) in 2022 losses. Hurricane Ian, Cyclone Freddy, floods in Australia and Pakistan, and crippling drought in the US, Europe, China and East Africa harmed millions of peoples lives and livelihoods. Many of these crises and impacts occur in the Global South, which suffers from low and decreasing densities of in situ observations. As the risks associated with hydrologic hazards increase, we also face severe challenges associated with declines in the available rain gauge observations.

Here, we describe recent improvements in the Climate Hazards Center (CHC) archive and two new satellite-gauge precipitation datasets: the Climate Hazards IMErg with Stations (CHIMES) and version 3 of the Climate Hazards Infrared Precipitation with Stations (CHIRPS-v3).

The 1981-to-present CHIRPS-v3 product is based on geostationary TIR cold cloud duration observations. The 2000-to-present CHIMES is based on the NASA IMERG Late product. Both products are built around the same high-resolution (0.05°) climatology. Both CHIRPS and CHIMES are enhanced with a large quality controlled set of global gauge observations.

Over the past several years, considerable efforts have been made to identify updatable sources of in situ rain gauge data, access this data, and set up processes for its routine quality control.

In Ethiopia, Somalia, El Salvador, Ecuador, Bolivia, Mozambique, Panama, Southern Africa, Honduras, Mexico, Guatemala, Brazil, Chile, Costa Rica, and Colombia thousands of additional stations were identified by (primarily) the CHC regional scientists. These observations are now being routinely ingested and quality controlled, some at the sub-monthly time scale. The total number of different station sources has increased from 24 in version 2 to more than 80 in version 3. Figure 1 contrasts monthly counts of CHIRPS station observations. CHIRPS-v3 adds ~10,000 observations per month over CHIRPS-v2.

The latest effort to reverse the reporting crisis and support capacity building in Africa comes in the form of 3D-Printed Automatic Weather Stations (3D-PAWS). A low-cost weather-observing network. One of the innovations is the use of dual tipping buckets for measuring rainfall. If the value of the two buckets start to drift, it becomes clear that the instrument needs some attention. Also there is lots of data logging of support info about battery life, network connectivity along with many other meteorological sensors. We expect this network to grow significantly over the next few years.