# MPCID, A new high-resolution multi-precipitation concentration indicators dataset for mainland China

## TL;DR

This paper introduces a new dataset for mainland China that tracks precipitation patterns from 1961 to 2100, helping understand climate change impacts on water and ecosystems.

## Contribution

The novel contribution is the creation of a high-resolution, spatiotemporally continuous multi-precipitation concentration indicators dataset for mainland China.

## Key findings

- MPCID integrates historical and projected precipitation data from 1961 to 2100 with high spatial resolution.
- PCD is the most reliable indicator with minimal errors and strong correlation with station data.
- DPCI shows moderate error control but limited daily-scale correlation due to precipitation stochasticity.

## Abstract

Global climate change is intensifying the hydrological cycle, manifested through an increased frequency of extreme precipitation events that pose substantial threats to water security and ecosystem resilience. Precipitation concentration indicators are critical for diagnosing these changes, yet their application has been constrained by data limitations: a reliance on fragmented station observations and a critical disconnect between historical benchmarks and future projections. To bridge this gap, we present the multi-precipitation concentration indicators dataset (MPCID) for mainland China, a spatiotemporally continuous resource spanning 1961–2100. MPCID integrates historical in-situ and gridded observations (1961–2022) with high-resolution (0.25°), statistically downscaled CMIP6 projections across four SSP scenarios (2015–2100). The dataset incorporates four key indicators: the precipitation concentration degree (PCD), precipitation concentration period (PCP), daily precipitation concentration index (DPCI), and monthly precipitation concentration index (MPCI). Rigorous validation against station data established PCD as the most reliable indicator, characterized by minimal errors, near-optimal correlation, and negligible bias across both historical and future scenarios. While DPCI exhibited moderate error control, its limited daily-scale correlation points to inherent stochasticity in short-term precipitation. MPCI demonstrated reduced sensitivity to extreme precipitation events, whereas PCP showed systematic limitations in temporal phase alignment despite retaining pattern recognition capability. By integrating historical fidelity with future scenarios, MPCID overcomes prior data fragmentation and establishes an indispensable foundation for investigating precipitation dynamics, assessing climate impacts on hydrology and agriculture, and informing adaptive management strategies.