Application of GIbased Procedures for Soil Moisture Mapping

Application of GI-based Procedures for Soil Moisture Mapping and Crop Vegetation Status Monitoring in Romania Dr. Adriana MARICA, Dr. Gheorghe STANCALIE, Daniel ALEXANDRU, Argentina NERTAN, Cristian FLUERARU National Meteorological Administration Bucharest, Romania WMO, FAO & COST 718 WORKSHOP ON CLIMATIC ANALYSIS AND MAPPING FOR AGRICULTURE Bologna, 14 -17 June 2005

Presentation outline: 1. Introduction: Recent developments in the use of GI-based procedures in Romania; 2. Application of GIS for mapping historical soil moisture data (methodology & results); 3. Examples of derived products obtained by using Arc View applications for improving operational agrometeorological activity; 4. Application of remote sensing data to monitor crop vegetation status; 5. Conclusions.

1. Introduction: The agrometeorological activity undergoing within National Meteorological Administration (NMA), integrates complex issues concerning the evolution of the water supply of soils and vegetation state of the crops with respect to the meteorological parameters evolution, being a particularly important activity whose final objective is to elaborate/edit the agrometeorological bulletins; In our institut there is an intense activity in the developing applications of new methods and technologies to provide additional improved agrometeorological information for decision-makers and other users; Our recent developments in the use of GI-based procedures summarized in this presentation include: statistical analysis procedure on historical series of soil moisture data to produce agroclimatic classification in order to promote sustainable land water management; NDVI image analysis procedure to monitor current crop vegetation conditions and its variation with time.

2. Application of GIS for mapping historical soil moisture data Objectives: § to analyze the spatial and temporal evolution of soil moisture over a 30 -year period in order to identify the intervals and zones with high risk at the occurrence of extreme climatic events; § to provide information on the long term average of soil moisture, that are very useful for placing current or expected conditions in a historical context. The application for spatial representation of soil moisture is performed in VBA (Visual Basic for Applications) for Arc View 8. 3 and is structured on two components: a) the data collected and the different information layers that are organized in a database and all the information is integrated in a GIS; b) spatial representation of soil moisture by using “Kriging” interpolation method.

Input data used: • Monthly data of measured soil moisture (m 3/ha) from 50 agrometeorological stations over the 1970 -2000 interval, for: two soil depths (0 -20 cm and 0 -100 cm); three calendar dates specific to the phenological phases of maize crop (April 30, July 31 and August 31); • Available water capacity -AWC (m 3/ha) for different soil types; • Calculated % of AWC.

Classification of the soil moisture levels: % of AWC 0 20 35 50 70 - 20% - 35% - 50% - 70% - 100% >100% HUMIDITY CLASSES Extremly drought Strongly drought Moderate drought Satisfactory supply Optimal supply Above normal moisture values To assess long term average of soil moisture: -the frequency (%) of humidity class was computed -the prevalent class of soil moisture in different climatic and soil type zones was established

Soil moisture database integration in a GIS

The map with the agrometeorological stations

The final map with the spatial representation of soil moisture Information concerning the values of soil misture content, % of AWC, soil moisture deficit and humidity class

Average Soil Moisture, 1970 -2000, at 30 April Maize available water in 0 -20 cm depth of soil Moisture data CURTEA DE ARGES Satisfactory supply (50 -70% of AWC) Near optimal supply (70 -85% of AWC) Optimal supply (85 -1 oo% of AWC) Moisture reserve value (m 3/ha): 420 % AWC: 86 Water deficit: 70 Significance: OS OK CANCEL

Average Soil Moisture, 1970 -2000, at 31 July Maize available water in 0 -100 cm depth of soil Moisture data FUNDULEA Moisture reserve value (m 3/ha): 825 % AWC: 48 Water deficit: 885 Significance: MD OK Stronghly drought (20 -35% of AWC) Moderate drought (35 -50% of. AWC) Satisfactory supply (50 -70% of AWC) Near optimal supply (70 -85% of AWC) CANCEL

Average Soil Moisture, 1970 -2000, at 31 August Maize available water in 0 -100 cm depth of soil Moisture data TECUCI Moisture reserve value (m 3/ha): 465 % AWC: 32 Water deficit: 985 Significance: SD OK Extremely drought (0 -20% of AWC) Stronghly drought (20 -35% of AWC) Moderate drought (35 -50% of AWC) Satisfactory supply (50 -70% of AWC) Near optimal supply (70 -85% of AWC) CANCEL

Example of yearly soil moisture variabiity, in the period 1970 -2000, at CALARASI station Drought years occured in 50 -60% of cases (15 -18 years)

Monthly precipitation (JPG file) Weekly Soil Moisture (JPG file) Weekly Soil Temperature (JPG file) Weekly precipitation (JPG file) Tmed>0°C, 1 Feb-10 April

4. Application of remote sensing data to monitor crop vegetation conditions Applications of NDVI values derived from SPOT-VEGETATION imagery are used to: follow dynamic of crop vegetation status; • infer crop progress (crop growth and development) and conditions; • analyze the temporal shape of the NDVI; • estimate beginning of growing season, leaf growth, etc. based on temporal profile analysis. Key issues for NDVI series data analysis: • overlay of a mask in order to isolate agricultural lands; • overlay of reference points selection of representative cropcultivated areas with at least 4 sq. km surfaces (equiv. of 4 pixels) using terrain information or high resolution satellitederived landcover maps; • accurate localization of the reference points using GPS; • extraction of NDVI values; • analysis & interpretation.

The monitoring of crop vegetation conditions and its variation with time is based on the analysis of the dekadal maps of NDVI; Begining from the last dekade of March 2005 these maps have been included analysed in the Agrometeorological Bulletin; Generally, the winter crops status was better as compared with previous year, especially for the crops sown in optimal time, but a delay in crop development with about 1 -2 weeks was observed.

Exemples of application remote sensing technology for monitoring crop vegetation status NDVI / Spot-Vegetation / 11 -20 March 2005

NDVI / Spot-Vegetation / 1 -10 April 2005

NDVI / Spot-Vegetation / 11 -20 April 2005

NDVI / Spot-Vegetation / 21 -30 April 2005

NDVI / Spot-Vegetation / 1 -10 May 2005

Conclusions: Ø The examples given in this presentation shown that the combination of ground-measurements, GIS and satellite data gives promising results for the mapping soil moisture to define agroclimatic potentialities and restrictions, and for monitoring crop vegetation status; Ø The capability of GIS technology is significant, enabling users to develop accurate and precise maps based on quantitative data to be analysed in combination with other sources of data; Ø The continues application of GIS and remote sensing technologies to evaluate soil moisture and vegetation state of the crops will contribute to improving the agrometeorological information in order to provide better decision support in agricultural management;

Conclusions Ø The assessment and near-real time monitoring of the risk agrometeorological factors and their zoning over the country’s agricultural territory, during vegetation season of crops will allow timely identification of the agricultural areas the most vulnerable and the dissemination of the information to the users for taking adequate measures (irrigation, fertilizing, agrotechnics to preserve the water in the soil, etc. );

Thank you