Mutlu Ozdogan

Multiscale analysis and validation of the MODIS LAI product. I. Uncertainty assessment

The development of appropriate ground-based validation techniques is critical to assessing uncertainties associated with satellite data-based products. Here we present a method for validation of the Moderate Resolution Imaging Spectroradiometer (MODIS) Leaf Area Index (LAI) product with emphasis on the sampling strategy for field data collection. This paper, the first of two-part series, details the procedures used to assess uncertainty of the MODIS LAI product. LAI retrievals from 30 m ETM+ data were first compared to field measurements from the SAFARI 2000 wet season campaign. The ETM+ based LAI map was thus as a reference to specify uncertainties in the LAI fields produced from MODIS data (250-, 500-, and 1000-m resolutions) simulated from ETM+. Because of high variance of LAI measurements over short distances and difficulties of matching measurements and image data, a patch-by-patch comparison method, which is more realistically implemented on a routine basis for validation, is proposed. Consistency between LAI retrievals from 30 m ETM+ data and field measurements indicates satisfactory performance of the algorithm. Values of LAI estimated from a spatially heterogeneous scene depend strongly on the spatial resolution of the image scene. The results indicate that the MODIS algorithm will underestimate LAI values by about 5% over the Maun site if the scale of the algorithm is not matched to the resolution of the data.

Irrigation-induced changes in potential evapotranspiration in southeastern Turkey : test and application of Bouchet's complementary hypothesis

Received 31 October 2003; revised 30 January 2004; accepted 23 February 2004; published 21 April 2004. [1] Since the early 1980s the southeastern part of Turkey has experienced major land use changes because of increasing irrigation. To determine the influence of large-scale irrigation on surface hydrologic fluxes, we investigated feedback mechanisms between the land surface and atmosphere within the framework of Bouchet’s complementary relationship between potential and actual evapotranspiration. Using 23 years of meteorological observations of temperature, humidity, wind speed, and radiation within the advection-aridity model (which is built on the complementary theory), we found that potential evaporation has decreased over 50%, from about 14 mm d � 1 to about 7 mm d � 1 , paralleling the increase in irrigated acreage over the same period. Pan evaporation data show a similar trend. The observed decline in potential evaporation is a direct result of decreases in wind speed and, to a lesser degree, increases in humidity. A comparison between reference values of evapotranspiration obtained from irrigation water use and evapotranspiration estimates from the advection-aridity method show a reasonable correlation (root-mean-square error = 1.1 mm d � 1 ), especially for large values of both variables. These results suggest that the rapid expansion of irrigation in southeastern Turkey has strongly modified the lower atmosphere in a way that appears consistent with Bouchet’s complementary theory. INDEX TERMS: 1803 Hydrology: Anthropogenic effects; 1818 Hydrology: Evapotranspiration; 1833 Hydrology: Hydroclimatology; 1842 Hydrology: Irrigation;

Global Land Cover Mapping: A Review and Uncertainty Analysis

Given the advances in remotely sensed imagery and associated technologies, several global land cover maps have been produced in recent times including IGBP DISCover, UMD Land Cover, Global Land Cover 2000 and GlobCover 2009. However, the utility of these maps for specific applications has often been hampered due to considerable amounts of uncertainties and inconsistencies. A thorough review of these global land cover projects including evaluating the sources of error and uncertainty is prudent and enlightening. Therefore, this paper describes our work in which we compared, summarized and conducted an uncertainty analysis of the four global land cover mapping projects using an error budget approach. The results showed that the classification scheme and the validation methodology had the highest error contribution and implementation priority. A comparison of the classification schemes showed that there are many inconsistencies between the definitions of the map classes. This is especially true for the mixed type classes for which thresholds vary for the attributes/discriminators used in the classification process. Examination of these four global mapping projects provided quite a few important lessons for the future global mapping projects including the need for clear and uniform definitions of the classification scheme and an efficient, practical, and valid design of the accuracy assessment.

Multiscale analysis and validation of the MODIS LAI product II. Sampling strategy

The development of appropriate ground-based validation techniques is critical to assessing uncertainties associated with satellite data-based products. In this paper, the second of a two-part series, we present a method for validation of the Moderate Resolution Imaging Spectroradiometer Leaf Area Index (MODIS LAI) product with emphasis on the sampling strategy for field data collection. Using a hierarchical scene model, we divided 30-m resolution LAI and NDVI images from Maun (Botswana), Harvard Forest (USA) and Ruokulahti Forest (Finland) into individual scale images of classes, region and pixel. Isolating the effects associated with different landscape scales through decomposition of semivariograms not only shows the relative contribution of different characteristic scales to the overall variation, but also displays the spatial structure of the different scales within a scene. We find that (1) patterns of variance at the class, region and pixel scale at these sites are different with respect to the dominance in order of the three levels of landscape organization within a scene; (2) the spatial structure of LAI shows similarity across the three sites, that is, ranges of semivariograms from scale of pixel, region and class are less than 1000 m. Knowledge gained from these analyses aids in formulation of sampling strategies for validation of biophysical products derived from moderate resolution sensors such as MODIS. For a homogeneous (within class) site, where the scales of class and region account for most of the spatial variation, a sampling strategy should focus more on using accurate land cover maps and selection of regions. However, for a heterogeneous (within class) site, accurate point measurements and GPS readings are needed.

Examination of the Bouchet–Morton Complementary Relationship Using a Mesoscale Climate Model and Observations under a Progressive Irrigation Scenario

The complementary relationship between actual and potential evaporation over southeastern Turkey was examined using a mesoscale climate model and field data. Model simulations of both actual and potential evaporation produce realistic temporal patterns in comparison to those estimated from field data; as evaporation from the surface increases with increasing irrigation, potential evaporation decreases. This is in accordance with the Bouchet–Morton complementary relationship and suggests that actual evapotranspiration can be readily computed from routine meteorological observations. The driving mechanisms behind irrigation-related changes in actual and potential evaporation include reduced wind velocities, increased atmospheric stability, and depressed humidity deficits. The relative role of each in preserving the complementary relation is assessed by fitting a potential evaporation model to pan evaporation data. The importance of reduced wind velocity in maintaining complementarity was unexpected, and thus examined further using a set of perturbation simulation experiments with changing roughness parameters (reflecting growing cotton crops), changing moisture conditions (reflecting irrigation), and both. Three potential causes of wind velocity reduction associated with irrigation may be increased surface roughness, decreased thermal convection that influences momentum transfer, and the development of anomalous high pressure that counteracts the background wind field. All three are evident in the mesoscale model results, but the primary cause is the pressure-induced local wind system. The apparent necessity of capturing mesoscale dynamical feedbacks in maintaining complementarity between potential and actual evaporation suggests that a theory more complicated than current descriptions (which are based on feedbacks between actual evaporation and temperature and/or humidity gradients) is required to explain the complementary relationship.

Comparison of prognostic and diagnostic surface flux modeling approaches over the Nile River basin

[1]xa0Regional evapotranspiration (ET) can be estimated using diagnostic remote sensing models, generally based on principles of energy balance closure, or with spatially distributed prognostic models that simultaneously balance both energy and water budgets over landscapes using predictive equations for land surface temperature and moisture states. Each modeling approach has complementary advantages and disadvantages, and in combination they can be used to obtain more accurate ET estimates over a variety of land and climate conditions, particularly for areas with limited ground truth data. In this study, energy and water flux estimates from diagnostic Atmosphere-Land Exchange (ALEXI) and prognostic Noah land surface models are compared over the Nile River basin between 2007 and 2011. A second remote sensing data set, generated with Penman-Monteith approach as implemented in the Moderate Resolution Imaging Spectroradiometer (MODIS) MOD16 ET product, is also included as a comparative technique. In general, spatial and temporal distributions of flux estimates from ALEXI and Noah are similar in regions where the climate is temperate and local rainfall is the primary source of water available for ET. However, the diagnostic ALEXI model is better able to retrieve ET signals not directly coupled with the local precipitation rates, for example, over irrigated agricultural areas or regions influenced by shallow water tables. These hydrologic features are not well represented by either Noah or MOD16. Evaluation of consistency between diagnostic and prognostic model estimates can provide useful information about relative product skill, particularly over regions where ground data are limited or nonexistent as in the Nile basin.

Trends in Land Cover Mapping and Monitoring

It is an interesting time for mapping and monitoring of land cover using remote sensing. There are exciting new kinds of maps derived from remote sensing at a variety of spatial scales. For example, there are a number of new kinds of maps of land cover becoming available globally. Following in the footsteps of the global land cover products derived from the 1992 1km AVHRR time series (the IGBP Discover Map (Loveland et al., 2000) and the University of Maryland Land Cover Map (Hansen et al., 2000)), two new products have been developed. One is the GLC2000 Land Cover Map (Bartalev et al., 2003) made using data from the SPOT4-VEGETATION sensor. Another is the MODIS land cover map, made using a time series of data from the MODIS sensor (Friedl et al., 2002). All of these maps are categorical in nature and include classes roughly at the level of biomes. There are differences between the legends of these maps, but their basic nature is similar. A somewhat different set of products are based on the idea of continuous fields, with a good example being percent forest cover (Hansen et al., 2002). These maps are also global and provide a different perspective on land cover.

Parcel-Level Identifi cation of Crop Types Using Different Classifi cation Algorithms and Multi-Resolution Imagery in Southeastern Turkey

This research investigates the accuracy of pixel- and object-based classifi cation techniques across varying spatial resolutions to identify crop types at parcel level and estimate the area at six test sites to fithe optimum data source for the identifi cation of crop parcels. Multi-sensor data with spatial resolutions of 2.5 m, 5 m and 10 m from SPOT5 and 30 m from Landsat-5 TM were used. Maximum Likelihood (ML), Spectral Angle Mapper (SAM), and Support Vector Machines (SVM) were used as pixel-based methods in addition to object-based image classifi cation (OBC). Post-classifi cation methods were applied to the output of pixel-based classifi cation to minimize the noise effects and heterogeneity within the agricultural parcels. In addition, processing-time performance of the algorithms was evaluated for the test sites and district scale classifi cation. OBC results provided comparatively the best performance for both parcel identifi cation and area estimation at 10 m and fi ner spatial resolution levels. SVM followed OBC at 2.5 m and 5 m resolutions but accuracies decreased dramatically with coarser resolutions. ML and SAM results were worse up to 30 m resolution for both crop type identifi cation and area estimation. In general, parcel identifi cation effi ciency was strongly correlated with spatial resolution while the classifi cation algorithm was a more effective factor than spatial resolution for area estimation accuracy. Results also provided an opportunity to discuss the effects of image resolution and the classifi cation algorithm independent factors such as parcel size, spatial distribution of crop types and crop patterns.

Patterns in Forest Clearing Along the Appalachian Trail Corridor

Forest clearing in the vicinity of the Appalachian Trail National Park undermines the Trails value as a wilderness retreat for millions of annual hikers. We estimate that 75,000 hectares of forest were lost to clearing during the decade of the 1990s inside a 16 km-wide corridor centered on the Trail. This loss represents 2.45 percent of forests within 8 km of the 3,500 km-long trail. Managed forest harvests in northern New England accounted for 76.8 percent of forest clearing. The factor most closely related to forest clearing is land ownership: only 0.29 percent of protected forests were cleared, while unprotected and managed forests were cleared at rates of 2.05 percent and 4.03 percent, respectively. A combination of boosted decision tree classifiers, multitemporal Kauth-Thomas transforms and the GeoCover Landsat dataset enabled a single, un-funded analyst to rapidly map land-cover change at 28.5-meter resolution within a 3.8 million hectare study area that spanned 16 Landsat scenes.

Monitoring changes in irrigated lands in Southeastern Turkey with remote sensing

Satellite based remotely-sensed data were used to document changes in irrigated lands over a period of ten years (1993 – 2002) in Southeastern Turkey. While MODIS (Moderate Resolution Imaging Spectroradiometer) derived temporal information was used to identify the time of year in which irrigated lands could be most easily differentiated from natural vegetation and rain-fed agriculture, seven near-yearly Landsat data were used to map irrigated lands with necessary spatial detail. A classification methodology based on thresholding of Landsat NDVI data from the late summer period reveals that the total irrigated land area has expanded three-folds from 35,179 ha in 1993 to 101,524 in 2002 in the Harran Plain, locus of the first phase irrigated lands. These results are comparable in magnitude to the official state provided estimates for each irrigation district, but are slightly higher than official results in absolute area of irrigated lands. One possible reason for the difference is that state estimates are based on area to be irrigated in the following water year, rather than actual area irrigated, as estimated with remotely sensed data. The results also show that data from the MODIS sensor, designed primarily for large area mapping, has the potential for supporting local scale studies where temporal information is lacking. Keywords-MODIS, Landsat, irrigation, NDVI