MCP 服务器

DMAP-AI MCP Server

MCP server for DMAP-AI drought monitoring and SPI drought analysis using NASA POWER and ERA5-Land/CDS data.

README

DMAP-AI MCP Server

Drought Monitoring and Analysis Platform — Model Context Protocol Server

DMAP-AI is a remote MCP server by AgriMetSoft LLC that exposes drought monitoring and analysis tools to AI assistants — including Claude, ChatGPT, and any MCP-compatible client. It provides SPI tables, drought severity event detection, wavelet scalogram analysis, and drought periodicity diagnostics over real climate data (NASA POWER and ERA5).

Quick Connect

Claude (claude.ai)

Go to Settings → Connectors → Add custom connector
Name: DMAP-AI
URL: https://droughtanalysis.com/mcp
Click Add, then enable it in your chat via the + button

ChatGPT / Gemini

DMAP-AI may also be available through compatible AI apps or custom assistant workflows. MCP support depends on the client’s current connector features.

Any MCP client (mcp.json / config)

{
  "mcpServers": {
    "dmap-ai": {
      "type": "streamable-http",
      "url": "https://droughtanalysis.com/mcp"
    }
  }
}

Tools

All tools are read-only (readOnlyHint: true, destructiveHint: false). They fetch and compute drought diagnostics from real climate data. No user data is written or stored.

`debug_ping`

Confirms the server is live and returns the current tool version and available free tools.

Input:  none
Output: server status, tool version, available tools list

`run_spi_table`

Returns a Standardized Precipitation Index (SPI) table for a given location and time period.

Input:
  latitude        float    e.g. 42.03
  longitude       float    e.g. -93.62
  start_date      string   YYYY-MM-DD  (default: 1981-01-01)
  end_date        string   YYYY-MM-DD  (default: 2024-12-31)
  baseline_start  int      (default: 1981)
  baseline_end    int      (default: 2024)
  spi_scale       int      1–48 months  (default: 12)
  data_source     string   "nasa_power" | "era5"  (default: nasa_power)
  step            string   "yearly" | "monthly"   (default: yearly)
  yearly_method   string   "jan_dec_totals"       (default)
  category_scheme string   "classic_spi" | "usdm_style" | "percentile_band" | "kmeans"
  max_rows        int      (default: 1000)

Output: SPI table with date, precipitation, SPI value, and drought category per row

Example prompt:

"Show me the SPI-12 table for Ames, Iowa from 1981 to 2024."

`run_drought_severity_events`

Detects and returns drought events (start, end, duration, minimum SPI, magnitude) using a configurable severity threshold.

Input:
  latitude / longitude / start_date / end_date / baseline_start / baseline_end
  spi_scale       int      (default: 12)
  data_source     string   "nasa_power" | "era5"
  step / yearly_method / category_scheme
  severity_threshold  string  SPI threshold, e.g. "-0.99", "-1.5"  (default: -0.99)

Output: drought events table (start date, end date, duration, min SPI, magnitude)

Example prompt:

"Find all drought events in Ames, Iowa from 1981 to 2024 using SPI-12 and a threshold of -1.0."

`run_wavelet_scalogram`

Runs a continuous wavelet transform and returns a compact time–period power matrix (scalogram), highlighting high-power drought zones.

Input:
  latitude / longitude / start_date / end_date / baseline_start / baseline_end
  spi_scale       int      (default: 12)
  data_source     string   "nasa_power" | "era5"
  step / yearly_method
  max_periods     int      downsampled period axis limit  (default: 24)
  max_times       int      downsampled time axis limit    (default: 60)

Output: compact scalogram matrix, top high-power cells, period and time axes

Example prompt:

"Run a wavelet scalogram for SPI-12 in Ames, Iowa from 1981 to 2024 and show the high-power zones."

`run_drought_periodicity_analysis`

Computes the global wavelet spectrum and identifies dominant and secondary drought periods (cycles) for a location.

Input:
  latitude / longitude / start_date / end_date / baseline_start / baseline_end
  spi_scale       int      (default: 12)
  data_source     string   "nasa_power" | "era5"
  step / yearly_method

Output: dominant period, top ranked periods by global power, global wavelet spectrum

Example prompt:

"Use DMAP-AI to run global wavelet analysis for SPI-12 in Ames, Iowa from 1981 to 2024. Then explain the dominant period."

Data Sources

Source	Description	Coverage
`nasa_power`	NASA POWER (default)	Global, ~0.5° grid
`era5`	ERA5 / Copernicus reanalysis	Global, ~0.25° grid

Supported Category Schemes

Scheme	Description
`classic_spi`	WMO standard SPI thresholds (default)
`usdm_style`	US Drought Monitor–style D0–D4 categories
`percentile_band`	Empirical percentile-based bands
`kmeans`	AI/ML k-means cluster labels

Notes

Free tools: SPI table, drought severity events, wavelet scalogram, periodicity/global wavelet spectrum
Not included in the free MCP version: Copula analysis and other advanced licensed features
All tools are stateless and idempotent — the same inputs always return the same outputs
Data source is NASA POWER by default (no API key required)