Multichannel Satellite Data Application for Global Earth Study

**119**

**Chapter 8**

Space

**Abstract**

*and Alexandra Hurduc*

a root mean square difference of 0.24 day.

Monchique fire (Portugal)

health and economy [19].

**1. Introduction**

Near- and Middle-Infrared

Monitoring of Burned Areas from

We describe a methodology to discriminate burned areas and date burning events that use a burn-sensitive (V, W) index system defined in near-/mid-infrared space. Discrimination of burned areas relies on a monthly composite of minimum of W and on the difference between this composite and that of the previous month. The rationale is to identify pixels with high confidence of having burned and aggregate new burned pixels on a contextual basis. Dating of burning events is based on the analysis of time series of *W,* and searching for the day before maximum temporal separability is achieved. The procedure is applied to the fire of Monchique, a large event that took place in the southwest of Portugal in August 2018. When the obtained pattern of burned pixels is compared against a reference map, the overall accuracy is larger than 99%; the commission and omission errors are lower than 5 and 10%, respectively; and the bias and the Dice coefficient are above 0.95 and 0.9, respectively. Differences between estimated dates of burning and reference dates derived from remote-sensed observations of active fires show a bias of 0.03 day and

**Keywords:** burned area, dates of burning, (V, W) index system, VIIRS sensor,

Vegetation fires have significant direct and indirect impacts on all components of the Earth system, including the anthroposphere. They are a source of greenhouse gases, aerosols, and trace gases to the atmosphere [1–3]; they induce modifications in most radiative forcing terms [4, 5] and disturb the radiative budget and cloud microphysics [6, 7]; they lead to changes in soil properties [8] and in the hydrological cycle [9–11]; they play a key role in biodiversity reduction, loss of genetic diversity, forest ecosystem functioning [12, 13], and land use/cover dynamics [14–16]; and they cause damages to human health [17, 18] and have adverse effects on public

A thorough understanding of spatial and temporal patterns of burned area (BA) by wildfires is therefore of fundamental importance when assessing either climate or anthropogenic influences on the Earth system [20, 21]; when addressing a very wide range of subjects that include the fields of atmospheric physics and chemistry,

*Carlos C. DaCamara, Renata Libonati, Miguel M. Pinto*

## **Chapter 8**

## Near- and Middle-Infrared Monitoring of Burned Areas from Space

*Carlos C. DaCamara, Renata Libonati, Miguel M. Pinto and Alexandra Hurduc* 

## **Abstract**

We describe a methodology to discriminate burned areas and date burning events that use a burn-sensitive (V, W) index system defined in near-/mid-infrared space. Discrimination of burned areas relies on a monthly composite of minimum of W and on the difference between this composite and that of the previous month. The rationale is to identify pixels with high confidence of having burned and aggregate new burned pixels on a contextual basis. Dating of burning events is based on the analysis of time series of *W,* and searching for the day before maximum temporal separability is achieved. The procedure is applied to the fire of Monchique, a large event that took place in the southwest of Portugal in August 2018. When the obtained pattern of burned pixels is compared against a reference map, the overall accuracy is larger than 99%; the commission and omission errors are lower than 5 and 10%, respectively; and the bias and the Dice coefficient are above 0.95 and 0.9, respectively. Differences between estimated dates of burning and reference dates derived from remote-sensed observations of active fires show a bias of 0.03 day and a root mean square difference of 0.24 day.

**Keywords:** burned area, dates of burning, (V, W) index system, VIIRS sensor, Monchique fire (Portugal)

## **1. Introduction**

Vegetation fires have significant direct and indirect impacts on all components of the Earth system, including the anthroposphere. They are a source of greenhouse gases, aerosols, and trace gases to the atmosphere [1–3]; they induce modifications in most radiative forcing terms [4, 5] and disturb the radiative budget and cloud microphysics [6, 7]; they lead to changes in soil properties [8] and in the hydrological cycle [9–11]; they play a key role in biodiversity reduction, loss of genetic diversity, forest ecosystem functioning [12, 13], and land use/cover dynamics [14–16]; and they cause damages to human health [17, 18] and have adverse effects on public health and economy [19].

 A thorough understanding of spatial and temporal patterns of burned area (BA) by wildfires is therefore of fundamental importance when assessing either climate or anthropogenic influences on the Earth system [20, 21]; when addressing a very wide range of subjects that include the fields of atmospheric physics and chemistry,

ecology, agriculture and forestry, hydrology, biology, sociology, and economy; and when defining climate, environment, and health policies [22–26]. When specifically focusing on fire management that comprises fire prevention, fire presuppression, and fire suppression measures, reliable information about the extent, location, and time of occurrence of BA is of high added value [25]. Accurate BA information is also crucial to land and fire decision-makers, as well as to research groups and ecologists, government agencies, and NGOs when implementing environmental policies aiming to reduce socioeconomic impacts from vegetation fires on ecosystems and people [27].

 The use of remotely sensed information for BA detection is well established, and there is a consensus about its usefulness from global down to regional levels [28–32]. Spaceborne sensors are a cost-effective way to map vegetation fires and the unique source of information for large areas with limited access at regional and global scales and for continuous monitoring over time [33, 34]. Over the last decades, several initiatives have been carried out to generate global and regional long-term maps of BA using remote sensing. These include, among others, (1) the 1-km L3JRC product, covering the period from April 2000 to March 2007, produced from SPOT VEGETATION data [35]; (2) the 1-km GLOBCARBON BA product, spanning the period April 1998–December 2007, derived from SPOT VEGETATION, Along Track Scanning Radiometer (ATSR-2), and Advanced ATSR (AATSR) imagery using a combination of mapping algorithms [36]; (3) the MCD45 [37] and MCD64A1 [30] BA products derived by NASA using data collected by the Moderate Resolution Imaging Spectroradiometer (MODIS); (4) the Global Fire Emissions Database (GEFD) initiative that consists in monthly BA estimates aggregated at 0.5° spatial resolution, covering the period from July 1996 to mid-2009 using four satellite data sets [38]; (5) the AQM-MODIS product [39] that was derived for Brazil and consists in monthly maps of BA at 1 km spatial resolution from 2000 up to the present; (6) the global burned area algorithm based on Medium Resolution Imaging Spectrometer (MERIS) reflectance and MODIS hotspots from 2006 to 2008 [29]; and (7) the recent global burned area product based on MODIS bands with a spatial resolution of 250 m [40].

Remote-sensed detection of burned vegetation makes use of spectral bands that are sensitive to spectral changes induced by fire events [41], namely, those associated to the deposit of char and ash on the surface and the change or destruction of vegetation structure [33]. Spectral indices have revealed to be the most appropriate to uncover changes in the radiometric signals of surfaces in operational applications [42], and a large variety of spectral indices for burned area discrimination have been developed in the last decades using a variety of techniques and different spectral bands, such as the red (R, about 0.6–0.7 μm), the near infrared (NIR, about 0.7–1.3 μm), the shorter short-wave infrared (SSWIR, about 1.3–1.9 μm), and the longer short-wave infrared (LSWIR, about 1.9–2.5 μm). Developed approaches include, among others, the Burned Area Index (BAI) [43] based on R and NIR and its improved version BAIM [44] based on NIR and LSWIR, the NIR and LSWIRbased Normalized Burn Index (NBI) [45], the Normalized Burn Ratio (NBR) [46] and derived indices from the latter [47–51], and the Mid-Infrared Burned Index (MIRBI) [52] based on SSWIR and LSWIR.

A burn-sensitive vegetation index system, the so-called (V, W) system, has also been defined on the NIR/MIR space with the aim of optimally discriminating burned vegetation [53, 54]. Here we present and discuss the use of the (V, W) index system to design an automated algorithm aiming at both mapping burned area and dating the associated burning events. As an example of application, the procedure is applied to the fire of Monchique, a large event that took place in the southwest of Portugal in August 2018 (**Figure 1**).

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

#### **Figure 1.**

*Land cover map of the Iberian Peninsula showing the geographical location (shaded rectangular area) and a zoom (top right box) of the study area near the southern coast of Portugal (source of land cover data: Modis collection 6 global land cover, https://lpdaac.usgs.gov/sites/default/files/public/product\_documentation/ mcd12\_user\_guide\_v6.pdf).* 

 The fire of Monchique started on August 3 about noon and was not dominated until August 9. The fire resulted in about 27,000 hectares of burned area, 41 people injured and millions of euros in economic losses. By the second day of the event, about 700 firefighters and 11 aerial resources were fighting the fire, and this number kept increasing up to about 1400 firefighters and 14 aerial resources. The fire occurred within a context of very high temperatures and intense and highly variable winds in terrain with difficult access and high accumulation of biomass.

### **2. Data and pre-processing**

 Input data to the algorithms to compute (V, W) consist of top-of-the-atmosphere (TOA) values of middle-infrared (MIR) and thermal-infrared (TIR) radiances and of near-infrared (NIR) reflectance, as acquired by the Visible Infrared Imaging Radiometer Suite (VIIRS) instrument on board of the joint NASA/NOAA Suomi National Polar-Orbiting Partnership (Suomi-NPP) satellite [55]. VIIRS data were reprojected onto a geographical grid of 0.0045° in latitude by 0.0059° in longitude, corresponding to about 500 m in spatial resolution. Data over Portugal, covering the period of July and August 2018, were extracted from the VIIRS/NPP Level 1B 375 m product [56] and correspond to bands I2 (NIR, centered at 0.865 μm), I4 (MIR, centered at 3.74 μm), and I5 (TIR, centered at 11.45 μm).

Geolocation data, as well as land/sea mask and solar and view angle information for each VIIRS tile, were obtained from the VIIRS geolocation product (VIIRS/NPP Imagery Resolution Terrain-Corrected Geolocation). Values of MIR reflectance were then computed using VIIRS bands I4 (MIR) and I5 (TIR) radiances [57]. All images acquired at solar zenith angles (SZA) greater than 55° were rejected, and, when more than one image was available for the same day, the image selected was the one with the lowest solar zenith angle (SZA). Images used as input to the algorithm for burned area discrimination were further restricted to those with view zenith angles (VZA) not exceeding 45° in order to prevent large distortions in pixel size [53].

Information about active fire data was obtained from the VIIRS 375 m Active Fire product [58]. Finally, radiative power data were obtained from the fire radiative power (FRP) product developed by the Land Surface Analysis Satellite Application Facility (LSA SAF); this product is derived from data acquired by the Spinning Enhanced Visible and Infrared Imager (SEVIRI) onboard Meteosat Second Generation (MSG) series of EUMETSAT geostationary satellites [59].

 A reference map of burned area in the study region was derived from geospatial information provided by the Rapid Mapping products of the Copernicus Emergency Management Service (EMS) [60]. The Copernicus EMS service was activated by the Portugal National Authority for Civil Protection on August 5 at 16:11 UTC (reference code EMSR303). We used the Delineation Map provided as of August 10 that has an estimated geometric accuracy of 5 m or better, derived by visual interpretation from Sentinel-2 and SPOT satellite observations.

## **3. Methods**

#### **3.1 Simplified (V, W)**

Specially designed to discriminate burned areas, the (V, W) burn-sensitive vegetation index system is defined in a transformed MIR/NIR space that allows enhancing the spectral information about burned vegetation [53]. The transformed space is framed by the following two coordinates: (1) the distance, η, of each point in MIR/NIR space to a predefined convergence point, representative of a given target (e.g., a totally burned surface) and (2) the difference, ξ, between the respective MIR and NIR reflectance of each point. The coordinates η and ξ are accordingly defined as

$$
\eta = \sqrt{\left(\rho\_{\text{MIIR}} - \rho^{\text{O}}\_{\text{MIIR}}\right)^2 + \left(\rho\_{\text{NIR}} - \rho^{\text{O}}\_{\text{NIR}}\right)^2} \tag{1}
$$

$$\mathfrak{F} = \rho\_{\text{MIR}} - \rho\_{\text{NIR}} \tag{2}$$

 where *ρMIR* and *ρNIR* represent values of reflectance in MIR and NIR and(*ρ*<sup>0</sup> *MIR*,ρ<sup>0</sup> *NIR*) are the coordinates of an ideally totally burned pixel.

 Values of *<sup>ρ</sup>MIR*  0 and *<sup>ρ</sup>* 0 *NIR* for a given sensor may be estimated by the upper (lower) bound of reflectivity in MIR (NIR) for a large sample of recently burned pixels. To estimate these values for the VIIRS sensor, we used a sample of burned regions for 0 several fires in central Portugal that occurred in 2017. Obtained estimates are *<sup>ρ</sup>MIR* = 0.29 0 and *<sup>ρ</sup>NIR* = 0.06.

 The coordinate system (V, W) is then defined in the MIR/NIR space such that the following properties are met: (1) the V coordinate has a very small dispersion for pixels associated to surfaces containing organic matter and (2) the W coordinate increases with increasing water content of vegetated surfaces. Burned vegetation is characterized by very low values of *W* and by a sharp decrease of *W* following a fire event [53], both characteristics being especially conspicuous in monthly minimum composites of W and of differences of W between a given month and the previous one (**Figure 2**). In turn, non-vegetated surfaces like clouds and water bodies are characterized by low values of V.

Unlike VI3 [57] and GEMI3 [33], the (V, W) index system has the advantage of not having been heuristically derived; however, unlike traditional indices that rely on simple algebraic expressions and are easy to implement by users, the *Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

**Figure 2.** 

 *Spatial distribution over the study area of the minimum composite of Wmin for August 2018 (left panel) and of the difference of minimum composites of Wmin between August and July 2018 (right panel).* 

 computation of (V, W) is laborious, involving iterative methods and numerical computation of line integrals [53]. This disadvantage is circumvented by using the following approximation that is valid in a subdomain of the MIR/NIR space where the majority of observed values are located [54]:

$$\mathbf{V} = \frac{\text{(0.16 - 0.71\xi)}}{\eta} \tag{3}$$

$$\mathbf{V} = \mathbf{1.1}\eta \tag{4}$$

#### **3.2 Discrimination of burned areas**

Discrimination of burned areas for a given month is based on a procedure that uses as inputs a monthly composite of minimum of W and the difference between this minimum composite and that of the previous month together with locations of all identified hotspots during the considered month [39].

The rationale is first to identify burned pixels with high confidence of being burned and then use these points as seeds in a growing algorithm that will identify other burned pixels on a contextual basis and aggregate them as new seed points. Several studies [39, 61] have pointed out that the vast majority of hotspots are located inside or in the neighborhood of a burned area and that the number of burned pixels that are not close to a hotspot is low.

As suggested by results shown in **Figure 2**, the first seed points are therefore pixels characterized by (1) a low value of the monthly minimum composite of index W and (2) a sharp decrease in that minimum compared to the previous month.

Burned pixels are also expected to be outliers in respect to the statistical distribution pixels where no hotspots were identified. Commonly used in classification problems, the Mahalanobis distance is a measure of the distance of a point to a given distribution in units of the standard deviation in the direction to the point to the mean [62]. The square of the Mahalanobis distance in a p-dimensional space has a chi-square distribution with p degrees of freedom, a result that may be used to find outliers in a dataset [63].

Identification of burned pixels is accordingly performed in the following three steps:

 • First step: Let *W min* and Δ*W min* be the values for a given pixel of the monthly composite of minimum W and of the difference between this

composite and that of the previous month; the pixel is considered as burned if all three following conditions are met:


 ○ Δ*W min* ∗ < 0


#### **3.3 Dating burned events**

For each burned pixel identified by the algorithm above-described, the date of burning is estimated by analyzing the time series of W for that pixel and searching for the day where maximum temporal separability is achieved [64]. For most cases, time series of W present daily fluctuations of rather small amplitude (**Figure 3**) which allows identifying the day when the burning event took place by the significant decrease in W following the event. The day of burning may accordingly be identified as the one that maximizes the following index of temporal separability [65]:

$$\mathbf{S} = \frac{2\left(\mu\_b - \mu\_a\right)}{\sigma\_a + \sigma\_b} \tag{5}$$

 where *μa*(σ*a*) are the values of the mean (standard deviation) of index *W* of that pixel for a pre-specified number k of images starting at a given instant in time and *μb*(σ*b*) are the respective values for the same k number of images before that instant in time. The time series of W is scanned by two juxtaposed windows of fixed length *k*, and index *S* is computed for every available day (**Figure 3**). The burning event is considered to have taken place in the day prior to the date when *S*  is maximized.

#### **3.4 Validation procedures**

#### *3.4.1 Discrimination of burned areas*

The Monchique BA was validated against the data obtained from the Copernicus Emergency Management Service (EMSR303) that is used as the reference map. The

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

#### **Figure 3.**

*Time series of indices W (blue dots, left vertical scale) and S (red dots, right vertical scale) for a pixel located inside the burned scar. The vertical black dashed line indicates the day of maximum S (Eq. (5)).* 


#### **Table 1.**

*Contingency table for pixels classified as burned versus unburned.* 


#### **Table 2.**

 *Accuracy (OA), omission error (OE), commission error (CE), bias (B), and dice coefficient (DC), with a, b, c, and d as defined in* **Table 1***.* 

 quality of the classification map was assessed based on five verification measures derived from contingency tables [39]: overall accuracy (OA), omission error (OE), commission error (CE), bias (B), and Dice coefficient (DC). These verification measures are defined in **Table 2**. The agreement between the BA scar and the reference map is measured by the OA, a high value of OA reflecting a high accuracy in the classification. OE and CE are used to assess the discriminative power of the classifier. The bias should be close to one when burning events are not overestimated/ underestimated. Finally, DC measures the similarity between the reference and the classification maps by overlapping the classified burned pixels to the "truly burned" pixels in the reference map.

Since the reference map has a higher resolution than the classification map, the former was projected onto the 500 m resolution grid of the latter by computing the burned fraction inside each coarser pixel. The pixel was then considered as burned if the fraction of burned area was greater than 0.5.

#### *3.4.2 Dating burned events*

Validation of estimated dates of burning was made against data of radiative power from the FRP product developed by the LSA SAF [59]. This product, together with three other active fire products derived from SEVIRI imagery, was compared

 against active fire data collected by the MODIS sensor, and results obtained showed a higher detection rate of active fire pixels than the other products [66]. Albeit presenting a coarser resolution of about 4 km in the study region, the repeat cycle of 15 min by the SEVIRI instrument allows for a much better temporal resolution than when comparing against VIIRS or MODIS active fires that have only two samples per day. Furthermore, the VIIRS active fires at 375 m resolution were already used in the algorithm to discriminate burned areas and therefore should not be used for validation purposes. The estimated date of each pixel classified as burned was compared to the date of observation of the nearest SEVIRI pixel where a hotspot was identified. Obtained differences between the dates of the burning of the classified burned pixels and the dates of hotspots identified by the SEVIRI instrument were then used to assess the performance of the dating methodology.

### **4. Example of application**

The above-described procedure was applied to the study region in the southwest of Portugal in order to discriminate burned pixels during the Monchique fire episode and then estimate the respective date of burning.

 As described in Sections 3.1 and 3.2, the identification of burned areas in the study region relies on monthly minimum composites of W for August (**Figure 2**, left panel) and of differences between the minimum composite of August and that of July (**Figure 2**, right panel), hereby referred to as *Wmin* and *ΔWmin*, respectively. Both composites were obtained from daily values of *W* as derived from reflectance values of MIR and NIR from all available VIIRS images with SZA not exceeding 55° and VZA not exceeding 45°.

 When values of *Wmin* and *ΔWmin* for all pixels over the study region are represented in a scatter plot (**Figure 4**), two clusters may be identified: (1) one that is formed by a dense cloud with a large number of points that mostly spread over the subarea of the plot that is lower bounded by percentile 10 of the distribution of *Wmin* (identified in the plot by the orange-dashed horizontal line) and left bounded by percentile 10 of the distribution of Δ*Wmin* (identified by the orange-dashed vertical line) and (2) a second cluster that is composed of a less dense cloud with a lower number of points that occupy the subarea that is upper bounded by percentile 10 of the distribution of *Wmin* and right bounded by percentile 10 of the distribution of Δ*Wmin*.

 The second cluster, formed by points with low values of both *Wmin* and Δ*Wmin*, is therefore likely to be associated to burned pixels. Moreover, also as to be expected in case of burned surfaces, the second cluster contains a very large fraction of pixels where hotspots were identified from the VIIRS Active Fire product (plotted as red dots). However, there are points (plotted as green dots) in the second cluster that are not associated to any hotspot, and there are also points in the first cluster that are associated to a hotspot, despite the fact that the large values of both *Wmin* and Δ*Wmin* are not consistent with the characteristic signature of a burned pixel. Both situations are to be expected, since (1) a pixel may burn with no active fire having been spotted by VIIRS (e.g., because of cloud or smoke screening, or because the burning took place between passages of the satellite) and (2) an identified active fire may have originated a burned area that represents a small fraction of the area of the pixel, and therefore the radiometric signature is not strong enough to be detected. Both difficulties may be circumvented in part by selecting a set of pixels with high confidence of being burned as seed points to feed into a growing algorithm.

 As discussed in Section 3.2 (first step of the algorithm), seed points are defined as pixels belonging to a region of the space (*Wmin*,Δ*Wmin*) where there is a high

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

#### **Figure 4.**

 *W min (August 2018) versus Δ W min (difference between August and July 2018). Red (green) dots indicate pixels with (without) hotspots associated. The orange ellipse represents percentile 95 of the Mahalanobis distance, and the horizontal (vertical) orange-dashed line represents percentile 10 of the distribution of W min (ΔW min).* 

confidence that points are associated to burned pixels. Taking into account the above-discussed features presented by the distribution of points in the scatter plot (**Figure 4**), seed points were defined according to the following criteria:


Once seed points were identified, new burned pixels were then iteratively aggregated following the procedure described in Section 3.2 (second and third steps of the algorithm).

Results obtained are shown in **Figure 5** that also provides a comparison with the reference map that was obtained from information derived from the Copernicus EMS (EMSR303). There is an overall agreement between the downscaled higher-resolution reference map and the map generated by the proposed algorithm. Deviations from the reference map, either in the form of commission or omission errors, are located along the borders of the scar and are likely to be due to small errors in geolocation or of partially burned pixels that were differently classified (as burned or unburned) by the proposed algorithm and the downscaled reference map.

 The overall quality of the proposed algorithm in discriminating the burned pixels associated to the Monchique fire episode reflects on values of the contingency table that compares results from the proposed algorithm with those from the reference map from Copernicus EMS (**Table 3**) as well as on the five verification measures derived from the obtained contingency table (**Table 4**). The number of commission errors (45) and the number of omission errors (94) are one order of magnitude lower than the number of match ups (979). In turn, the overall accuracy is larger than 99%, the commission error is lower than 5%, and the omission error is lower than 10%; the bias is above 0.95, and the Dice coefficient is above 0.9.

#### **Figure 5.**

*Burned pixels (left panel) from the proposed algorithm and reference map (right panel) from Copernicus EMS (EMSR303). True positives, commission errors, and omission errors are colored in green, red, and blue, respectively.* 


#### **Table 3.**

*As in* **Table 1** *but with values obtained for the scar that resulted from the Monchique fire event of August 2018.* 


#### **Table 4.**

*As in* **Table 2** *but with the metrics derived from* **Table 3***.* 

Following the procedure described in Section 3.3, estimates were obtained of the date of burning for all pixels that were classified as burned within the study region. Results obtained (**Figure 6**, left panel) show a propagation from NW to SE, forming a pattern that is very similar to the one derived from the dates of detection of hotspots by the SEVIRI instrument (**Figure 6**, right panel). The agreement between the latter dates and the estimates by the proposed dating algorithm reflects on the obtained histogram of differences that has the null value of differences as the modal frequency, closely followed by a delay of 1 day in the estimates, such that about 70% of the pixels classified as burned have differences in the dates of less than ±1 day. When considering the distribution of differences as a whole, there is a bias of −0.03 day and a root mean square difference of 0.24 day, both values pointing out the very good overall agreement between estimates from the proposed algorithm and the reference dates derived from SEVIRI (**Figure 7**).

Results obtained using a similar procedure over the whole territory of Portugal for August and September 2005, one of the worst severe years in terms of burned area, [64] present an overall accuracy of 95.6% and commission and omission errors of 66.5 and 37.1%, respectively. However, the study encompasses a period of *Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

#### **Figure 6.**

*Dates of burning as obtained from the proposed dating algorithm (left panel) and as derived from dates of observation of hotspots by the SEVIRI instrument (right panel).* 

**Figure 7.** 

*Histogram of difference between dates assigned by the proposed methodology and dates derived from hotspots identified by the SEVIRI instrument.* 

2 months and a much wider area, covering a very large number of scars, and not a single one as in the present study. Regarding the estimated days of burning, 75% of estimated dates in the same study [64] presented deviations less than ±5 days from dates derived from hotspots identified by MODIS.

#### **5. Conclusions**

 Using TOA values of MIR and TIR radiances and NIR reflectance from VIIRS 375m imagery, a set of optimal indices, V and W, were used to discriminate burned areas and to assign dates to every burned pixel. The ability of *V* to discriminate between vegetated and non-vegetated surfaces may be used to build up composites of *W* free from contamination by clouds, whereas the low values of W associated to burned surfaces suggest generating composites of minimum values of W to discriminate burned areas. Adopting this rationale, and in line with previous work [39, 64], discrimination of burned areas was performed using values (*Wmin*) of a monthly composite of minimum of W and values (Δ*Wmin*) of

 differences between that composite and the one of the previous month. First seed points are identified as the pixels that (1) are outliers in respect to pixel where no hotspots were identified, (2) present low values of *Wmin* characteristic of burning event, and (3) are associated to negative values of Δ*Wmin*, indicating a decrease of *Wmin* that is expected to occur after a burning event. New burned pixels are then successively aggregated using a seeded region-growing algorithm that starts with the previously identified seed points.

 The algorithm was applied to the Monchique fire episode, a large event that occurred in southwestern Portugal during August 2018. The discriminative power of the algorithm was validated against the scar identified by Copernicus EMS303. Results obtained show that the (V, W) algorithm is suited to discriminate burned area over a mainland Portugal, supported by the good agreement, with a Dice coefficient of 0.933, between the burned area scar and the reference map. The commission and omission errors have values of 9 and 5%, respectively. Estimated dates of burning, obtained through analysis of time series of values of W, were compared against times of observation of hotspots obtained from the SEVIRI FRP product. About 70% of the estimated dates presented deviations of 1 day or less.

 The development of reliable algorithms to discriminate and date burned areas is crucial for a better understanding of the biosphere-atmosphere interactions, for estimating burning emissions, for future projections of fire regime, and for mitigation and adaptation actions in Portugal, which is recurrently affected by severe fire events. In particular, accurate estimates of the date of burning are crucial when considering fire regime modeling, due to the constraint imposed by biomass availability into the spread of fire, and are also important for reducing uncertainties in biomass burning emissions [34]. The recent VIIRS sensor will allow the development of new burned area products at high spatial resolution, continuing and enhancing the imaging of the Earth initiated by the Advanced Very High-Resolution Radiometer (AVHRR) and the MODIS instruments. The present work represents a first attempt to assess the potential of using VIIRS imagery to identify burnt scars in Portugal. Results obtained in this work and in related previous ones pave the way to the generation of a long-term series of burned area maps containing accurate information about the extent, location, and time of occurrence of vegetation fires.

## **Acknowledgements**

This research is supported by FAPESP/FCT Project Brazilian Fire-Land-Atmosphere System (BrFLAS) (FCT 2015/01389-4 and FAPESP/1389/2014) and by the EUMETSAT Land Surface Analysis Satellite Application Facility (LSA SAF). Research by Renata Libonati was funded by Serrapilheira Institute (grant number Serra-1708-15159) and supported by Centro de Estudos Florestais (CEF) of the University of Lisbon, a research unit funded by FCT (UID/AGR/00239/2013). Research by Miguel M. Pinto was supported by FCT through PhD grant PD/ BD/142779/2018. Research by Alexandra Hurduc was supported by a grant in the framework of Project "Reabilitação das Áreas Queimadas na Freguesia de Alvares," financed by donation of Observador on time SA.

## **Conflict of interest**

The authors declare they have no conflicts of interest.

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

## **Author details**

Carlos C. DaCamara1 \*, Renata Libonati2,3, Miguel M. Pinto1 and Alexandra Hurduc1

1 Faculdade de Ciências, Instituto Dom Luiz, Universidade de Lisboa, Lisboa, Portugal

2 Departmento de Meteorologia, Instituto de Geociências, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil

3 Centro de Estudos Florestais, Universidade de Lisboa, Lisboa, Portugal

\*Address all correspondence to: cdcamara@fc.ul.pt

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

## **References**

[1] Dentener F, Kinne S, Bond T, Boucher O, Cofala J, Generoso S, et al. Emissions of primary aerosol and precursor gases in the years 2000 and 1750 prescribed data-sets for AeroCom. Atmospheric Chemistry and Physics. 2006;**6**:4321-4344. DOI: 10.5194/ acp-64321-2006

[2] Zhang J, Webster J, Powers RF, Mills J. Reforestation after the fountain fire in northern California: An untold success story. Journal of Forestry. 2008;**106**: 425-430. DOI: 10.1093/jof/106.8.425

 [3] van Leeuwen W, Casady G, Neary D, Bautista S, Alloza J, Carmel J, et al. Monitoring post-wildfire vegetation response with remotely sensed time series data in Spain, USA and Israel. International Journal of Wildland Fire. 2010;**19**:75-93. DOI: 10.1071/WF08078

[4] Bowman DMJS, Balch JK, Artaxo P, Bond WJ, Carlson JM, Cochrane MA, et al. Fire in the Earth system. Science. 2009;**324**(5926):481-484. DOI: 10.1126/ science.1163886

 [5] Storelvmo T, Leirvik T, Lohmann U, Phillips PC, Wild M. Disentangling greenhouse warming and aerosol cooling to reveal Earth's climate sensitivity. Nature Geoscience. 2016;**9**(4):286. DOI: 10.1038/ngeo2670

[6] Andreae MO, Rosenfeld D, Artaxo P, Costa AA, Frank GP, Longo KM, et al. Smoking rain clouds over the Amazon. Science. 2004;**303**:1337-1342. DOI: 10.1126/science.1092779. DOI: 10.1126/ science.1092779

[7] Lohmann U, Rotstayn L, Storelvmo T, Jones A, Menon S, Quaas J, et al. Total aerosol effect: Radiative forcing or radiative flux perturbation? Atmospheric Chemistry and Physics. 2010;**10**(7):3235-3246. DOI: 10.5194/ acp-10-3235-2010

[8] Certini G. Effects of fire on properties of forest soils: A review. Oecologia. 2005;**143**(1):1-10. DOI: 10.1007/s00442-004-1788-8

[9] Rosenfeld D. TRMM observed first direct evidence of smoke from forest fires inhibiting rainfall. Geophysical Research Letters. 1999;**26**(20):3105-3108. DOI: 10.1029/1999GL006066

[10] Menon S, Hansen J, Nazarenko L, Luo Y. Climate effects of black carbon aerosols in China and India. Science. 2002;**297**:2250-2253. DOI: 10.1126/ science.1075159

[11] Koren I, Kaufman YJ, Remer LA, Martins JV. Measurement of the effect of Amazon smoke on inhibition of cloud formation. Science. 2004;**303**:1342-1345. DOI: 10.1126/science.1089424

[12] Fisher B, Turner RK, Morling P. Defining and classifying ecosystem services for decision making. Ecological Economics. 2009;**68**(3):643-653. DOI: 10.1016/j.ecolecon.2008.09.014

[13] Driscoll D, Lindenmayer D, Bennett AF, Bode M, Bradstoc RA, Cary G, et al. Fire management for biodiversity conservation: Key research questions and our capacity to answer them. Biological Conservation. 2010;**143**(9):1928-1939. DOI: 10.1016/j. biocon.2010.05.026

[14] Sellers PJ, Bounoua L, Collatz GJ, Randall DA, Dazlich DA, Los SO, et al. Comparison of radiative and physiological effects of doubled atmospheric CO2 on climate. Science. 1996;**271**:1402-1406. DOI: 10.1126/ science.271.5254.1402

[15] Jin Y, Roy DP. Fire-induced albedo change and its radiative forcing at the surface in northern Australia. Geophysical Research

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

Letters. 2005;**32**:L13401. DOI: 10.1029/2005GL022822

[16] Lambin EF, Geist HJ. Land-Use and Land-Cover Change. Local Processes and Global Impacts. Berlin: Springer Science & Business Media; 2008. p. WP6319

[17] Bowman DMJS, Johnston FH. Wildfire smoke, fire management, and human health. EcoHealth. 2005;**2**(1):76-80. DOI: 10.1007/ s10393-004-0149-8

[18] Fowler CT. Human health impacts of forest fires in the southern United States: A literature review. Journal of Ecological Anthropology. 2003;**7**(1):39- 63. DOI: 10.5038/2162-4593.7.1.3

[19] Patz JA, Engelberg D, Last J. The effects of changing weather on public health. Annual Review of Public Health. 2000;**21**(1):271-307. DOI: 10.1146/ annurev.publhealth.21.1.271

[20] Bowman D. Wildfire science is at a loss for comprehensive data. Nature. 2018;**560**:7. DOI: 10.1038/ d41586-018-05840-4

 [21] Andela N, Morton DC, Giglio L, Chen Y, Van Der Werf GR, Kasibhatla PS, et al. A human-driven decline in global burned area. Science. 2017;**356**:1356-1362. DOI: 10.1126/ science.aal4108

[22] Flannigan MD, Krawchuk MA, de Groot WJ, Wotton BM, Gowman LM. Implications of changing climate for global wildland fire. International Journal of Wildland Fire. 2009;**18**(5):483- 507. DOI: 10.1071/WF08187

 [23] Langmann B, Duncan B, Textor C, Trentmann J, van der Werf GR. Vegetation fire emissions and their impact on air pollution and climate. Atmospheric Environment. 2009;**43**(1):107-116. DOI: 10.1016/j. atmosenv.2008.09.047

[24] Kochi I, Donovan GH, Champ PA, Loomis JB. The economic cost of adverse health effects from wildfire-smoke exposure: A review. International Journal of Wildland Fire. 2010;**19**(7):803-817. DOI: 10.1071/WF09077

[25] Bowman DM, Balch J, Artaxo P, Bond WJ, Cochrane MA, D'Antonio CM, et al. The human dimension of fire regimes on Earth. Journal of Biogeography. 2011;**38**(12):2223-2236. DOI: 10.1111/j.1365-2699.2011.02595.x

[26] Johnston FH, Henderson SB, Chen Y, Randerson JT, Marlier M, DeFries RS, et al. Estimated global mortality attributable to smoke from landscape fires. Environmental Health Perspectives. 2012;**120**(5):695. DOI: 10.1289/ehp.1104422

 [27] Pacheco AP, Claro J, Fernandes PM, de Neufville R, Oliveira TM, Borges JG, et al. Cohesive fire management within an uncertain environment: A review of risk handling and decision support systems. Forest Ecology and Management. 2015;**347**:1-17. DOI: 10.1016/j.foreco.2015.02.033

[28] Nogueira J, Ruffault J, Chuvieco E, Mouillot F. Can we go beyond burned area in the assessment of global remote sensing products with fire patch metrics? Remote Sensing. 2016;**9**(1):7. DOI: 10.3390/rs9010007

[29] Alonso-Canas I, Chuvieco E. Global burned area mapping from ENVISAT-MERIS and MODIS active fire data. Remote Sensing of Environment. 2015;**163**:140-152. DOI: 10.1016/j. rse.2015.03.011

[30] Giglio L, Boschetti L, Roy DP, Humber ML, Justice CO. The collection 6 MODIS burned area mapping algorithm and product. Remote Sensing of Environment. 2018;**217**:72-85. DOI: 10.1016/j.rse.2018.08.005

 [31] Pereira AA, Pereira J, Libonati R, Oom D, Setzer AW, Morelli F, et al. Burned area mapping in the Brazilian savanna using a one-class support vector machine trained by active fires. Remote Sensing. 2017;**9**(11):1161. DOI: 10.3390/ rs9111161

[32] Kasischke ES, French NHF, Harrell P, Christensen NL Jr, Ustin SL, Barry D. Monitoring of wildfires in boreal forests using large area AVHRR NDVI composite image data. Remote Sensing of Environment. 1993;**45**:61-71

[33] Pereira JMC. A comparative evaluation of NOAA/AVHRR vegetation indices for burned surface detection and mapping. IEEE Transactions on Geoscience and Remote Sensing. 1999;**37**(1):217-226. DOI: 10.1109/36.739156

[34] Mouillot F, Schultz MG, Yue C, Cadule P, Tansey K, Ciais P, et al. Ten years of global burned area products from spaceborne remote sensing—A review: Analysis of user needs and recommendations for future developments. International Journal of Applied Earth Observation and Geoinformation. 2014;**26**:64-79. DOI: 10.1016/j.jag.2013.05.014

[35] Tansey K, Gregoire JM, Defourny P, Leigh R, Pekel JFO, van Bogaert E, et al. A new, global, multi-annual (2000-2007) burnt area product at 1 km resolution. Geophysical Research Letters. 2018;**35**:1. DOI: 10.1029/2007GL031567

[36] Plummer S, Arino O, Simon M, Steffen W. Establishing an earth observation product service for the terrestrial carbon community: The GLOBCARBON initiative. Mitigation and Adaptation Strategies for Global Change. 2006;**11**:97-111. DOI: 10.1007/ s11027-006-1012-8

[37] Roy DP, Boschetti L, Justice CO, Ju J. The collection 5 MODIS burned area product: Global evaluation by comparison with the MODIS active fire product. Remote Sensing of Environment. 2008;**112**:3690-3707. DOI: 10.1016/j.rse.2008.05.013

[38] Giglio L, Randerson JT, van der Werf GR, Kasibhatla PS, Collatz GJ, Morton DC, et al. Assessing variability and long-term trends in burned area by merging multiple satellite fire products. Biogeosciences. 2010;**7**:1171-1186. DOI: 10.5194/bg-7-1171-2010

[39] Libonati R, DaCamara CC, Setzer AW, Morelli F, Melchiori AE. An algorithm for burned area detection in the Brazil Cerrado using 4 μm MODIS imagery. Remote Sensing. 2015;**7**:15782- 15803. DOI: 10.3390/rs71115782

[40] Chuvieco E, Lizundia-Loiola J, Pettinari ML, Ramo R, Padilla M, Mouillot F, et al. Generation and analysis of a new global burned area product based on MODIS 250 m reflectance bands and thermal anomalies. Earth System Science Data Discussions. 2018;**512**:1-24. DOI: 10.5194/essd-2018-46

[41] Trigg S, Flasse S. Characterising the spectral-temporal response of burned savannah using in situ spectroradiometry and infrared thermometry. International Journal of Remote Sensing. 2000;**21**:3161-3168. DOI: 10.1080/01431160050145045

[42] Verstraete MM, Pinty M. Designing optimal spectral indexes for remote sensing applications. IEEE Transactions on Geoscience and Remote Sensing. 1996;(5):1254-1265. DOI: 10.1109/36.536541

[43] Isabel MDPM. Cartografía e inventario de incendios forestales en la Península Ibérica a partir de imágenes NOAA-AVHRR. Universidad de Alcalá de Henares; 1999

[44] Martín MP, Gómez I, Chuvieco E. Performance of a burned-area index

*Near- and Middle-Infrared Monitoring of Burned Areas from Space DOI: http://dx.doi.org/10.5772/intechopen.82444* 

(BAIM) for mapping Mediterranean burned scars from MODIS data. In: Proceedings of the 5th International Workshop on Remote Sensing and GIS Applications to Forest Fire Management: Fire Effects Assessment. Paris: Universidad de Zaragoza, GOFC GOLD, EARSeL; 2005. pp. 193-198

[45] Alleaume S, Hely C, Le Roux J, Korontzi S, Swap RJ, Shugart HH, et al. Using MODIS to evaluate heterogeneity of biomass burning in southern African savannahs: A case study. International Journal of Remote Sensing. 2005;**26**:4219-4237. DOI: 10.1080/01431160500113492

[46] USGS. Product Guide: Landsat surface reflectance-derived spectral indices [Internet]. 2007. Available from: https://landsat.usgs.gov/sites/ default/files/documents/si\_product\_ guide.pdf [Accessed: October 28, 2018]

[47] Loboda T, O'Neal KJ, Csiszar I. Regionally adaptable dNBR-based algorithm for burned area mapping from MODIS data. Remote Sensing of Environment. 2007;**109**:429-442. DOI: 10.1016/j.rse.2007.01.017

[48] Miller JD, Yool SR. Mapping forest post-fire canopy consumption in several overstory types using multitemporal Landsat TM and ETM data. Remote Sensing of Environment. 2002;**82**(2-3):481-496. DOI: 10.1016/ S0034-4257(02)00071-8

 [49] Boer MM, Macfarlane C, Norris J, Sadler RJ, Wallace J, Grierson PF. Mapping burned areas and burn severity patterns in SW Australian eucalypt forest using remotely-sensed changes in leaf area index. Remote Sensing of Environment. 2008;**112**:4358-4396. DOI: 10.1015/j.rse.2008.08.005

[50] Miller JD, Thode AE. Quantifying burn severity in a heterogeneous landscape with a relative version of the delta normalized burn ratio (dNBR). Remote Sensing of Environment. 2007;**109**:66-80. DOI: 10.1016/j. rse.2006.12.006

 [51] Bastarrika A, Chuvieco E, Martín MP. Mapping burned areas from Landsat TM/ETM+ data with a two phase algorithm: Balancing omission and commission errors. Remote Sensing of Environment. 2011;**105**:1003-1012. DOI: 10.1016/j.rse.2010.12.005

[52] Trigg S, Flasse S. An evaluation of different bi-spectral spaces for discriminating burned shrub-savannah. International Journal of Remote Sensing. 2001;**22**(13):2641-2647. DOI: 10.1080/01431160110053185

[53] Libonati R, DaCamara CC, Pereira JMC, Peres LF. On a new coordinate system for improved discrimination of vegetation and burned areas using MIR/ NIR information. Remote Sensing of Environment. 2011;**114**:831-843. DOI: 10.1016/j.rse.2011.02.006

[54] DaCamara CC, Libonati R, Ermida SL, Calado TJ. A user-oriented simplification of the (V, W) burnsensitive vegetation index system. IEEE Geoscience and Remote Sensing Letters. 2016;**13**(12):1822-1826. DOI: 10.1109/ LGRS.2016.2614319

[55] NASA. Data Product User Guide for Suomi-National Polar-Orbiting Partnership (S-NPP) Sounder Science Investigator-led Processing System (SIPS) Advanced Technology Microwave Sounder (ATMS) Level 1B Products [Internet]. 2017. Available from: https://docserver. gesdisc.eosdis.nasa.gov/repository/ Mission/SNPP\_Sounder/3.3\_ ScienceDataProductDocumentation/ 3.3.4\_ProductGenerationAlgorithms/ ATMS\_Readme\_Vers\_1\_20170508.pdf [Accessed: October 28, 2018]

[56] NASA. Visible Infrared Imaging Radiometer Suite (VIIRS) 375 m &

750 m Active Fire Detection Data Sets Based on NASA VIIRS Land Science Investigator Processing System (SIPS) Reprocessed Data—Version 1 [Internet]. 2017. Available from: https://lpdaac.usgs.gov/sites/default/ files/public/product\_documentation/ vnp14\_user\_guide\_v1.3.pdf [Accessed: October 28, 2018]

[57] Kaufman YJ, Remer LA. Detection of forests using MID-IR reflectance: An application for aerosol studies. IEEE Transactions on Geoscience and Remote Sensing. 1994;**32**:672-683. DOI: 10.1109/36.297984

[58] Schroeder W, Oliva P, Giglio L, Csiszar IA. The new VIIRS 375 m active fire detection data product: Algorithm description and initial assessment. Remote Sensing of Environment. 2014;**143**:85-96. DOI: 10.1016/j. rse.2013.12.008

[59] Wooster MJ, Roberts G, Freeborn PH, Xu W, Govaerts Y, Beeby R, et al. LSA SAF Meteosat FRP products—Part 1: Algorithms, product contents, and analysis. Atmospheric Chemistry and Physics. 2015;**15**(22):13217-13239. DOI: 10.5194/acp-15-13217-2015

[60] Copernicus. Emergency Management Service [Internet]. 2018. Available from: http://emergency. copernicus.eu/ [Accessed: October 28, 2018]

[61] Hantson S, Padilla M, Cort D, Chuvieco E. Strengths and weaknesses of MODIS hotspots to characterize global fire occurrence. Remote Sensing of Environment. 2013;**131**:152-159. DOI: 10.1016/j.rse.2012.12.004

[62] Mahalanobis PC. On the generalised distance in statistics. Proceedings of the National Institute of Sciences of India. 1936;**2**(1):49-55

[63] Wilks DS. Statistical Methods in the Atmospheric Sciences. Vol. 100. Oxford: Academic Press; 2006. 676 pp

[64] Panisset J, DaCamara CC, Libonati R, Peres LF, Calado TJ, Barros A. Assigning dates and identifying areas affected by fires in Portugal based on MODIS data. Anais da Academia Brasileira de Ciências. 2017;**89**(3):1487-1501. DOI: 10.1590/0001-3765201720160707

 [65] Giglio L, Loboda T, Roy DP, Quayle B, Justice CO. An active-fire based burned area mapping algorithm for the MODIS sensor. Remote Sensing of Environment. 2009;**113**:408-420. DOI: 10.1016/j.rse.2008.10.006

[66] Roberts G, Wooster MJ, Xu W, Freeborn PH, Morcrette JJ, Jones L, et al. LSA SAF Meteosat FRP products— Part 2: Evaluation and demonstration for use in the Copernicus Atmosphere Monitoring Service (CAMS). Atmospheric Chemistry and Physics. 2015;**15**:13241-13267. DOI: 10.5194/ acp-15-13241-2015

## Chapter 9

## The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping the Water Balance over Puerto Rico for Water Resource Management

John R. Mecikalski and Eric W. Harmsen

## Abstract

A solar insolation satellite remote sensing product for Puerto Rico, the US Virgin Islands (USVI), Dominican Republic, Haiti, Jamaica, and Cuba became available in 2009 through a collaboration between the University of Puerto Rico-Mayagüez Campus and the University of Alabama in Huntsville. Solar insolation data are available at 1 km resolution for Puerto Rico and the USVI and 2 km resolution for the other islands, as derived from 500 m resolution GOES-16 visible imagery. The insolation data demonstrate the powerful utility of satellite-derived fields for water resource applications, specifically the routine production of potential and reference evapotranspiration. This chapter describes the theoretical background and technical approach for estimating components of the daily water and energy balance in Puerto Rico. Useful information can be obtained from the model, which benefits disaster and emergency management, agriculture, human health, ecology, coastal water management, and renewable energy development at the island scale.

Keywords: incoming solar radiation, insolation, GOES, Puerto Rico, Caribbean, evapotranspiration, remote sensing, water resource management, reference evapotranspiration, potential evapotranspiration

## 1. Introduction

Estimates of incoming solar radiation (also known as "insolation") have been made from geostationary satellite data for many years, since the early to mid-1970s [1]. Related to the present effort, Geostationary Operational Environmental Satellite (GOES) visible channel (˜0.64 μm) data have been processed within a scalable and flexible insolation model, which is well documented and described in detail below. For ongoing water management support over Puerto Rico and the broader Caribbean, the Diak-Gautier insolation model [2] has been specifically structured to provide daily integrated, gridded solar insolation at 1–2 km spatial resolution. The insolation model has been rigorously tested and validated and operates on GOES imagery from GOES–4/–5 through the present day GOES–16/–17. Geostationary satellites are optimal for providing spatially and temporally continuous fields across all regions in their ˜55° latitude field of view, which as noted is a significant advantage over the use of only ground-based instrumentation. The use of a satellitebased insolation algorithm also ensures that a consistent algorithm is applied across an entire region, one which relies on data from only one instrument, specifically, the GOES Imager.

Over Puerto Rico (PR) and the Caribbean, as well as in other subtropical and tropical regions, evapotranspiration (ET) is a critical variable for water management, both in hydrologic flow simulations involving potential ET (PET) and water allocation and agricultural water use involving reference ET (RET or ETo). Importantly, solar insolation is a large, yet often unknown, determinant for temporal variation in PET and RET. Solar insolation is a primary determinant of spatial variation, particularly in areas with heterogeneous cloud cover, as common to subtropical and tropical regions where small cumulus clouds dominate the regional cloud climatology.

For an ET product to be desirable, it must be spatially continuous, rather than consisting of only point values derived from local weather station networks. Thus, mapping of ET is greatly facilitated by satellite-derived estimates that contain the actual spatial variability and distribution of solar insolation. Prior to 2009, regions across Puerto Rico and the Caribbean did not had access to a consistent, spatially continuous method of computing RET and PET. The original motivation for development of the Geostationary Operational Environmental Satellite-Water and Energy Balance (GOESWEB) model was to develop a robust insolation calibration framework coupled to a satellite-based insolation model, to provide a key radiative dataset that can grow over time toward 10-year and longer timeframes, thus forming an ET climatology that can be extended indefinitely.

The GOES-based insolation datasets are used in conjunction with other information, including net radiation (Rn), air temperature, relative humidity, wind speed, and land cover information, in the formulation of daily, 1- and 2-km estimates of RET across the Caribbean. RET is valuable for farm- and city-based water management, as well as irrigation scheduling; PET can be used as input into surface and groundwater hydrological models, whereas the solar insolation data themselves may be used as data input in certain ecosystem models.

#### 2. GOES solar insolation data

The use of geostationary satellite visible data has been used for estimating solar insolation for over 30 years. The main methods used for such estimation range from statistical-empirical relationships, such as [3], to varying complex physical models [2, 4–12]. Studies such as [13, 14] proved the utility and feasibility of satelliteestimated solar insolation methods, demonstrating that fairly accurate results can be produced from such models; hourly insolation estimates obtained from the most current models are within 5–10% of ground-based pyranometer data, during clearsky conditions (15–30% for all sky conditions), while daily estimates are found to be within 10–15% [15]. Studies by [16, 17] have further highlighted the overall utility of these methods.

The main advantages of using satellite-estimated insolation, over those collected by pyranometer networks, include wide-area spatial coverage, high spatial resolution (1–2 km), and the ability to produce useful data in remote, inaccessible, or in potentially hazardous areas, over large water bodies and oceans (e.g., [18]), and in locations where the installation of a ground-based pyranometer network is prohibitive. As an alternative to the methods used in this study, [19, 20] describe the use of the Global Energy and Water Cycle Experiment (GEWEX) Surface Radiation Budget (SRB) downward solar flux [11], as used within the North American Land Data

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

Assimilation project. Error statistics for the SRB product are comparable to those shown in [21], as used in this study, yet SRB resolutions are at best 0.5° and 3 hourly [22].

Related to the PRWEB applications to be developed here, the solar insolation product is derived from National Oceanic and Atmospheric Administration (NOAA) GOES-East satellite visible (0.64 μm) imagery. These data were processed using [4] methods to produce daily integrated solar insolation throughout Puerto Rico at 1-km horizontal spatial resolution. This 1-km resolution is chosen as it provides solar insolation observations between cumulus clouds, which comprise a significant component of the cloud climatology in subtropical regions.

#### 2.1 Details of the GOES solar insolation model

The GOES solar insolation model is developed by [4], which was later modified by [2] and updated by [23], and most recently by [24], which is the 2017 version of the solar insolation model employed in this study. This model will be referred to as the "GD" model from this point forward and employs a simple physical model that represents cloud and atmosphere radiative processes. The GD model was shown to perform even better than more complex solar insolation methods over a variety of land-surface and climatic conditions [5, 17, 18, 23, 26]. When comparing with pyranometer data, these prior studies list root mean square errors in hourly and daily insolation estimates as a percentage of the mean pyranometer observed value, which range from 17–28% to 9–10%, respectively. In [24, 25] the higher magnitudes of these errors were reported (˜28 and ˜10%, respectively) in a study over northern central Florida using GOES–12 data. However, the GD model has been proven to be valuable in operational use of near-real-time, regional-, and continental-scale insolation estimates for several main applications, including land-surface carbon and water flux assessments [27–29], the generation of agricultural forecasting products [30, 31], and subsurface hydrologic modeling.

The GD model is based on conservation of radiant energy in the Earthatmosphere column, with two modes for estimating solar insolation received at Earth's surface: (1) clear and (2) cloudy conditions. These modes are determined based on satellite-derived, visible channel surface albedo data. A reference albedo grid representative of clear-sky conditions per satellite pixel is developed within the GD algorithm, which captures the temporal changes in land-surface characteristics over time and season. This running 2-week minimum of this albedo data, reassessed at solar noon daily, is stored for each GOES satellite visible data pixel. This approach is considered representative of the true land-surface albedo, which is more accurate than using the daily estimated value as the latter may be corrupted by high albedo values when even low-cloud amounts are present during a given day. Note that this minimum albedo is wavelength-specific, is unique to the GOES Imager visible sensor (which includes some near-infrared contribution), and is not a true surface albedo.

As the GD algorithm runs across a series of GOES images per day, the digital brightness at each image pixel is compared to that of the stored clear-sky reference 2-week minimum albedo for that pixel. If the brightness exceeds that threshold, the pixel is deemed partly or completely cloudy. Based on this determination per GOES pixel, either the clear or cloudy model of atmospheric radiation processes (within the GD model) is used to calculate surface solar insolation received. Both clear and cloudy models incorporate parameterizations for Raleigh scattering, ozone absorption, and water vapor absorption within the atmospheric column, using simple bulk relationships, such as fixed ozone and aerosol contents. This rough parameterization works because these produce secondary sources of error to the instantaneous

surface solar insolation. The cloudy GD component estimates a cloud-top albedo and separately accounts for atmospheric effects above and below the cloud.

Forthe water vapor absorption parameterization, a fixed, approximate annual median value of precipitable water (PW) of 3.0 cm was used, which is considered appropriate for Puerto Rico. This annual median value helpsto estimate atmospheric column-integrated PW during the initial processing. [PW is defined asthe amount of waterthat would precipitate out of a vertical column of the atmosphere if all the water vapor were condensed into liquid]. PW data are used to calculate the slantwise path and subsequently the absorption coefficients [4]. Real-time PW data from numerical forecast model output may also be used in the GD model, versussetting a constant value.

#### 2.2 GOES data processing and quality control

The GOES-East series of satellites (the most recent additions being GOES–13 and –16) are in geostationary orbit above the Earth's equator at ˜75° W, which provides continuous, 5–15-minute resolution observations in visible and infrared radiation channels at high spatial (500 m to 1 km). GOES data are thus ideal for high-resolution estimates of solar insolation as used in GOESWEB, to be described below. Although the GOES visible sensors have a nadir (the point directly below the satellite) spatial resolution of 1 km (GOES–13 and prior) or 500 m (GOES–16), this resolution decreases the further from nadir the instrument scans: for Puerto Rico, the highest resolution attainable is about 1.25 km and 525 m, respectively, for GOES–13 and –16. All solar insolation data used for this study were provided at 1 km resolution. A simple method for computing sunrise and sunset times per pixel across the domain was used, as a means of determining daytime conditions.

Potential significant GOES data issues that may impact the error in the solar insolation product include (1) sensor degradation with time and (2) sun glint effects. The effects of the latter are small. In general, GOES satellite data are available on a continual basis with high reliability (>99%). As an example, Figure 1 shows the daily integrated solar radiation for October 16, 2018, for Puerto Rico, the USVI, Hispaniola, Jamaica, and Cuba.

#### 3. The GOESWEB modeling framework

GOESWEB performs daily water and energy balance calculations for the island of Puerto Rico. Twenty-seven hydro-agro-climate variables are available to the public for download (Table 1). Downloadable formats are available as images (jpg) or in comma-separated values (csv) and Matlab® formats. The variables in Table 1 are also available as monthly and annual averages or totals. Simplified versions of the algorithm have been developed for estimating reference ET on the islands of the USVI, Hispaniola, Jamaica, and Cuba.

ETo is estimated by three methods: Penman-Monteith [32, 33], Priestley-Taylor [34], and Hargreaves-Samani [35]. In [36], they described the methodology used to estimate ETo in the earliest version of the algorithm. Tavg, Tmin, and Tmax values were estimated from a lapse rate method developed by [37]. Td was assumed to be equal to the minimum daily Tmin [38]. Wind speed was assumed to be the worldwide average 2-m wind speed of 2 m(s)˜<sup>1</sup> [32]. The algorithms for Hispaniola, Jamaica, Cuba, and the USVI continue to use these simplified methods for estimating daily values of ETo.

Water and energy balances were added to the algorithm for Puerto Rico. The daily meteorological data used are described below. Table 2 summarizes the GOESWEB input data sources.

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

	- i. Solar radiation (Rs) is derived from the GOES satellite using the methodology described above.
	- ii. The ground level, 1-km resolution Rs product became available in Puerto Rico in March of 2009 and has been validated at two locations in Puerto Rico by [39].
	- iii. Occasionally the satellite-derived solar radiation is not available, in which case the previous days' Rs values are used.
	- iv. Prior to GOES–16, 1 km GOES–12 and –13 visible channel 1 data were used over Puerto Rico and the USVI and 2 km data over the other islands.

#### Figure 1.

Daily solar insolation for (a) Puerto Rico, (b) Hispaniola, (c) Jamaica, (d) Cuba, and (e) St. Croix USVI, (f) St. Thomas, and (g) St. John (USVI) on October 16, 2018.


#### Table 1.

Hydro-agro-climate variables produced daily by GOESWEB.


#### Table 2.

Summary table of required GOESWEB input data sources.

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

	- i. Tavg, Tmin, Tmax, and Td data were obtained from the National Digital Forecast Database (NDFD) website [40] from January 1, 2009, to December 31, 2016.
	- ii. Temperatures were obtained from the CariCOOS operational gridded Weather Research and Forecasting (WRF) model starting on January 1, 2017.
	- iii. On occasion, weather parameters from the WRF model are not available. NDFD air temperatures and wind speed are used in those cases. As a final resort, the lapse rate method of [37] is used.
	- i. During the period January 1, 2009, through September 30, 2015, daily average wind speed was obtained from the average of eight NDFD 3 hour values [40].
	- ii. From October 1, 2015, to the present, daily average wind speed was obtained from the average of 24-hourly wind speed values obtained from the Caribbean Coastal Ocean Observing System (CariCOOS) WRF.
	- iii. For the reference ET calculation, 10-m wind speeds are adjusted to 2 meters [32].
	- iv. If wind speed is not available, then the previous day's data are used.

#### 3.1 Net radiation calculations

Net radiation (Rn) is estimated using the methodology described by [32] and used by [41].

$$R\_n = R\_{ns} + R\_{nl} \tag{1}$$

where Rn is net radiation, Rns is net shortwave radiation, and Rnl is net long wave radiation.

$$R\_{ns} = (\mathbf{1} - a)R\_s \tag{2}$$

where α is albedo and Rs solar radiation. α is defined as 0.23 for estimating RET, and values are derived from a lookup table associated with 32 land cover classes [42] for estimating actual ET (ETa). Rs is derived from the GOES satellite. The net long wave radiation is estimated from the equation

$$\mathbf{R\_{nl}} = \sigma \left[ \frac{\mathbf{T\_{max,K}}^4 - \mathbf{T\_{min,K}}^4}{2} \right] (\mathbf{0.34} - \mathbf{0.14}\sqrt{\mathbf{e\_4}}) \left( \mathbf{1.35}\frac{\mathbf{R\_s}}{\mathbf{R\_{so}}} - \mathbf{0.35} \right) \tag{3}$$

where σ is the Stefan-Boltzmann constant, Tmax is maximum absolute temperate during the 24-hour period, Tmin is minimum absolute temperature during the 24-hour period, ea is actual vapor pressure, Rs/Rso is relative shortwave radiation (limited to ≤1.0), and Rso calculated clear-sky radiation. Actual vapor pressure is estimated by

Satellite Information Classification and Interpretation

$$\mathbf{e}\_{\mathbf{a}} = \mathbf{0}.6\mathbf{1}08 \,\,\exp\frac{(\mathbf{1}7.27\mathbf{T}\_{\mathrm{d}})}{(\mathbf{T}\_{\mathrm{d}} + 237.3)}\tag{4}$$

where Td is dew point temperature. The calculated clear-sky radiation is estimated by

$$\mathbf{R\_{so}} = \left(\mathbf{0.75} + \mathbf{2 10^{-5} z}\right) \mathbf{R\_{a}} \tag{5}$$

where z is elevation above mean sea level and Ra is extraterrestrial radiation.

$$\mathbf{R\_{4}} = \frac{\mathbf{12(60)}}{\pi} \mathbf{G\_{sc}} \mathbf{d\_{r}} [ (\boldsymbol{\alpha\_{2}} - \boldsymbol{\alpha\_{1}}) \sin(\boldsymbol{\uprho}) \sin(\boldsymbol{\delta}) + \cos(\boldsymbol{\uprho}) \cos(\boldsymbol{\updelta}) (\sin(\boldsymbol{\upmu\_{2}}) - \sin(\boldsymbol{\upmu\_{1}}))] \tag{6}$$

where Gsc is the solar constant = 0.0820 and dr is the relative distance Earth-Sun, defined as

$$\mathbf{d}\_{\mathbf{f}} = \mathbf{1} + \mathbf{0}.\mathbf{3}\mathbf{3}\cos\left(\frac{2\pi}{\mathbf{3}65}\mathbf{J}\right) \tag{7}$$

where J is Julian day (e.g., January 1 is 1 and December 31 is 365). ω<sup>1</sup> in Eq. (6) is solar time angle at the beginning of the period and ω<sup>2</sup> solar time angle at end of period, generally expressed as

$$\rho\_t = \frac{\pi}{2} - \arctan\left[\frac{-\tan\left(\rho\right)\tan\left(\delta\right)}{X^{0.5}}\right] \tag{8}$$

where φ is latitude and δ solar declination expressed as

$$\delta = 0.409 \sin\left(\frac{2\pi}{365}l - 1.39\right) \tag{9}$$

and X is defined as

$$X = \mathbf{1} - \left[\tan\left(\rho\right)\right]^2 \left[\tan\left(\delta\right)\right]^2 \tag{10}$$

and X = 0.00001 if X ≤ 0.

#### 3.2 Reference evapotranspiration estimates

The Penman-Monteith (PM) equation is given by Eq. 1 [32], which applies specifically to a hypothetical reference crop with an assumed crop height of 0.12 m, an albedo of 0.23, a fixed surface resistance of 70 sec m�<sup>1</sup> , and an aerodynamic resistance equal to 208/u2, where u2 is wind speed at 2 m height:

$$ET\_o = \frac{0.408\Delta (R\_n - G) + \chi \left(\frac{900}{T + 273}\right) u\_2 (\mathbf{e}\_s - \mathbf{e}\_a)}{\Delta + \chi (1 + 0.34u\_2)}\tag{11}$$

where Δ is the slope of the vapor pressure curve, G is soil heat flux, γ is the psychrometric constant, T is mean daily temperature at 2 m height, es is the saturation vapor pressure, and ea is the actual vapor pressure.

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

The second method used to estimate ETo is the Priestly-Taylor Equation [34], a simplification of the Penman Equation [43, 44]:

$$ET\_o = a \frac{\Delta (R\_n - G)}{\Delta + \chi} \tag{12}$$

where α is the Priestly-Taylor constant. Values in the literature for α range from 1.26 [34] to 1.32 [45]. In this study we use a value of α equal to 1.3.

The third method used to estimate ETo is the Hargreaves-Samani ETo Equation [35] given by

$$ET\_o = 0.408[0.0135 \text{R}](T + 17.8) \tag{13}$$

The value 0.0135 is a constant and 0.408 converts the result from MJ m�<sup>2</sup> day to mm (day)�<sup>1</sup> . In [38] they showed that this method produces comparable results with the PM method in PR.

The PM method is considered superior to the other two methods because it accounts for the major variables that control ET (Rn, T, VPD and u), and the PM method has been rigorously validated [33].

#### 3.3 Energy balance

In GOESWEB, an energy balance approach is used similar to [46]. The basic energy balance equation is given as

$$R\_n - LE - H - G = 0\tag{14}$$

Rn is obtained from the calculation procedure presented above. Albedo, which is used in the Rn calculation, is obtained from a lookup table [42], which assigns values of the parameters to 32 different land covers.

LE, H, and G are the latent, sensible, and soil heat fluxes, respectively. LE is estimated using the following Equation [47]:

$$LE = \frac{\rho \mathbf{C}\_p (e\_o(T\_s) - e(T\_d))}{\gamma (r\_a + r\_s)} \tag{15}$$

where ρ is mean air density, Cp is specific heat, ra is aerodynamic resistance, and rs is surface resistance. G is the soil heat flux, assumed to be zero for the daily analysis. H is estimated using the following equation:

$$H = \frac{\rho \mathbf{C}\_p (T\_s - T\_a)}{r\_a} \tag{16}$$

The effective surface temperature is difficult to obtain from remote sensing under cloudy conditions. Therefore, Ts is obtained by an implicit approach similar to that described by [48]. When Eq. (14) is expanded using Eqs. (1), (15), and (16), Ts is the only unknown variable, which is obtained using the recursive root function fzero in MatLab® (http://www.mathworks.com).

The aerodynamic resistance (ra) is calculated with the following Equation [46]:

$$r\_a = r\_{ao} \phi + r\_{bh} \tag{17}$$

where rao is the aerodynamic resistance under conditions of neutral atmospheric stability and rbh is the excessive resistance. rao is expressed as

$$r\_{ao} = \frac{\ln\left[\frac{x - x\_{dip}}{x\_o}\right] \ln\left[\frac{x - x\_{dip}}{(0.1)x\_o}\right]}{k^2 u} \tag{18}$$

where z is the virtual height at which meteorological measurements are taken. In this study z is assumed to be within the inertial sublayer and equal to 1.5(zo/0.13) [47], which is equivalent to the canopy height (h). The NDFD or WRF modelderived wind speeds at 10 m height are adjusted to the "virtual instrument height," depending on the height of the vegetation. Roughness length (zo) and the zeroplane displacement (zdisp) are derived from a lookup table for various land use/ vegetation categories [42]. k is Von Karman's constant (k = 0.41). u is the wind velocity at height z.

From [46], the atmospheric stability coefficient is

$$\phi = \left[ \mathbf{1} - \frac{\left[ \eta \left( \mathbf{z} - \mathbf{z}\_{dip} \right) \mathbf{g} \left( T\_s - T\_d \right) \right]}{T\_o \mu^2} \right] \tag{19}$$

where g is the gravitational constant and the coefficient η is taken as 5 [46]. The temperature, To, is the average of the values of Ts and Ta. Other variables and parameters were previously defined.

The excess resistance in Eq. (17) is given by the equation

$$r\_{bh} = \frac{4}{\left(\frac{ku}{\ln\left[\left(x - x\_{\text{dip}}\right)/x\_{\text{o}}\right]}\right]}\tag{20}$$

Bulk surface resistance (rs) is estimated using the equation of [49]:

$$r\_s = \frac{\rho \mathbf{C}\_p \mathbf{VPD}}{\Delta (R\_n - \mathbf{G}) \mathbf{C}\_f} \left(\frac{\theta - \theta\_{\rm WP}}{\theta\_{\rm FC} - \theta\_{\rm WP}}\right)^{-1} \tag{21}$$

where VPD is the vapor pressure deficit, Cf is a calibration coefficient equal to 1 for root depth <1 m and 5 for root depth >1 m, and θFC and θWP are the volumetric soil moisture content (θ) at field capacity and wilting point, respectively. Field capacity and wilting point were obtained from regression equations of [50] based on percent sand, silt, and clay. Soil properties for sand, silt, and clay for Puerto Rico were obtained from the USDA Natural Resource Conservation Service (NRCS) and Soil Survey Geographic (SSURGO) database.

#### 3.4 Water balance

The water balance is estimated from the equation

$$\text{SMD2} = \text{Precip} - \text{ET}\_a - \text{RO} - \text{DP} + \text{SMD1} \tag{22}$$

where SMD1 and SMD2 are the depths of soil moisture in the root zone (Rdepth) at times 1 and 2, respectively. In GOESWEB the time step is 1 day. Precip is rainfall, RO is surface runoff, and DP is deep percolation below the root zone. The daily ETa is obtained by converting LE to an equivalent depth of water by dividing by the latent heat of vaporization (2.45 MJ kg�<sup>1</sup> ). Root depths for various land use/vegetation categories are obtained from [42] lookup table. Twenty-four-hour rainfall is obtained from NOAA's Advanced Hydrologic Prediction Service (AHPS). In PR, AHPS rainfall is bias-corrected radar rainfall using rain gauge data.

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

Surface runoff is estimated using the curve number (CN) method of the NRCS [51]:

$$RO = \frac{(Precip - 0.2S)^2}{(Precip + 0.8S)} \tag{23}$$

$$\mathcal{S} = \left[ \left( \frac{25400}{\text{CN}} \right) - 254 \right] \tag{24}$$

where S is the maximum potential difference between rainfall and runoff at the moment of rainfall initiation and CN is a proportion of rainfall converted to runoff, adjusted for antecedent rainfall conditions. CN values were derived for Puerto Rico using the method described by [51], based on land use, hydrologic soil group, and antecedent rainfall conditions.

To estimate DP, the following procedure is followed: SMD2i = Precip – ETa – RO + SMD1. If the value of SMD2i is larger than the depth of water in the soil profile at field capacity (FCD), then DP = SMD2i – FCD, and the value of SMD2 is equal to FCD. If SMD2i < FCD, then DP = 0 and SMD2 = SMD2i.

#### 3.5 GOESWEB model accuracy and validation

In this section accuracy and validation data are presented for remotely sensed solar radiation, RET, soil moisture, and stream flow. Solar radiation is a critically important variable in the estimation of ET. Figures 2 and 3 show comparisons of the daily integrated solar radiation at the University of Puerto Rico (UPR) Fortuna Agricultural Experimental near Juana Diaz, PR, and the UPR-Mayaguez Campus (UPRM) in Mayaguez, PR, respectively [39]. The figures show a high degree of correlation between the remote sensing solar radiation and the measured solar radiation. The coefficients of determination (r<sup>2</sup> ) for the UPRM and experimental station data were 0.88 and 0.83, respectively. (From [39]).

Figure 4 shows a comparison of the ETo computed by the GOESWEB algorithm and from weather station data from the UPR Fortuna Agricultural Experiment Station, near Juana Diaz, PR. The ETo data covers the period from December 12, 2013 to April 20, 2016 (858 days). Although the vast majority of data pairs fall close to the 1:1 line, indicating close agreement between the two methods, a smaller number of data pairs fall relatively far from the 1:1 line, producing the scatter in the

#### Figure 2.

Comparison of remote sensing and pyranometer-measured daily integrated solar radiation at the UPR Fortuna Agricultural Experiment Station, near Juana Diaz, PR (From [39]). The r <sup>2</sup> value for this comparison is 0.88.

#### Figure 3.

Comparison of remote sensing and pyranometer-measured daily integrated solar radiation at UPRM (From [39]). The r <sup>2</sup> value for this comparison is 0.83.

#### Figure 4.

Comparison of observed and simulated ETo at the UPR Fortuna Agricultural Experiment Station, near Juana Diaz, PR. The data cover the period December 12, 2013 through April 20, 2016. The r <sup>2</sup> value for this comparison is 0.31.

data set. For this comparison, the coefficient of determination (r<sup>2</sup> ) was 0.31. The average GOESWEB and weather station ETo were 4.6 mm and 4.14 mm, respectively, and the average calculated error was 11.2%. It should be noted that the weather station at this location does not comply with the required "reference conditions" for computing ETo. Reference conditions refer to a grass-type vegetation with an approximate height of 0.12 m, an albedo of 0.23, and a fixed surface resistance of 70 sec m˜<sup>1</sup> , receiving adequate water. The climate of southern PR is semiarid, and there are frequent times when there was no vegetation at all on the ground surrounding the weather station.

Figure 5 shows a time series comparison of soil moisture from GOESWEB and soil moisture from a weather station located at the UPR Fortuna Agricultural Experiment Station. The weather station soil moisture is an average of five sensors positioned at depths of 0.0508 m (2 in.), 0.1016 m (4 in.), 0.2032 m (8 in.), 0.508 m (20 in.), and 1.016 m (40 in.). Immediately after rainfalls the weather station soil moisture tended to rise to higher soil moisture values than the soil moisture from the model. It is important to know that maximum soil moisture values in GOESWEB are limited to the field capacity, as excess water is routed below the root zone as deep percolation. Furthermore, the sensor soil moisture represent a single point (approximately 1 m<sup>2</sup> ), whereas the model represents an

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

#### Figure 5.

Comparison of the GOESWEB and weather station soil moisture at the UPR Fortuna Agricultural Experiment Station, near Juana Diaz, PR. The data covers the period January 1, 2014–December 31, 2016. The r <sup>2</sup> value for this comparison is 0.73.

Figure 6.

Comparison of observed and simulated stream flow for two watersheds in southwest Puerto Rico. The data cover the 36-month period during 2010–2012. The r <sup>2</sup> value for this comparison is 0.72 for all data.

area of 1 km<sup>2</sup> (1,000, 000 m<sup>2</sup> ), and therefore, complete agreement between the two methods would not be expected.

Figure 6 compares the monthly stream flow values for two watersheds in southwest Puerto Rico. Observed stream flow values were obtained from the US Geological Survey (USGS). The results are presented as a depth of water in millimeters (i.e., monthly stream volume/watershed area). The total stream flow for the model was assumed to be the surface runoff plus the deep percolation (or aquifer recharge). The latter term represents the stream base flow. To obtain the monthly value of stream flow in the model, the surface runoff and deep percolation were averaged for every 1 km2 pixel within the watershed. The model does a reasonably good job of simulating monthly stream flow.

### 4. High-resolution products for Puerto Rico

Figure 7 shows an example of selected water and energy balance components for Puerto Rico on October 16, 2018. Rainfall, surface runoff, percolation below the

Figure 7.

Example of water and energy balance components from the GOESWEB algorithm for November 24, 2015.

Figure 8.

NOAA's Drought Monitor for Puerto Rico, October 18, 2018. The municipalities of Aibonito, Cayey, and Cidra are experiencing abnormally dry conditions.

root zone, soil moisture content, actual ET, Rn, LE, and H are included. Approximately 60 mm of rain fell along the northern coast of the island. High values of surface runoff occurred in the rainy area where soil textures have high clay content. High values of percolation below the root zone occurred in small areas where the soil sand content approaches 90%. The soil moisture map indicates a lobe of dry area in Salinas, Cayey, Aibonito, and Cidra. Figure 8 shows the NOAA Drought Monitor for Puerto Rico for October 18, 2018, indicating abnormally dry conditions for Cayey, Aibonito, Cidra, and a portion of Barranquitas. The figure also shows LE The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

Figure 9. Rainfall over Puerto Rico during the week of Hurricane Maria, during September 2017.

and H fluxes, which sum to the Rn (i.e., Eq. (14)). ETa is the LE flux divided the latent heat of vaporization constant equal to 2.45 MJ kg˜<sup>1</sup> .

Figure 9 shows the rainfall during the week of September 17, 2017, the same week Hurricane Maria occurred. The maximum rainfall for the week was nearly 1300 mm (51 in.) in southeast Puerto Rico. The rainfall data were derived from rain gauge data, since the Doppler radar in Cayey, PR, was severely damaged during the hurricane. The National Weather Service (NWS) combined the gauge rainfall for September 20 and 21. The maximum rainfall during the 2-day period was 950 mm

Figure 10.

Rainfall over Puerto Rico on September 20, the day that Hurricane Maria made landfall on Puerto Rico. The gauge rainfall reported by the NWS was for the 20th and 21st; therefore the rainfall for September 20 was assumed to be half the amount.

#### Figure 11.

Estimated surface runoff over Puerto Rico on September 20, the day that Hurricane Maria made landfall on Puerto Rico.

Figure 12.

Root zone soil moisture saturation for September 19, 2018, 1 day before Hurricane Maria made landfall on Puerto Rico.

#### Figure 13.

ETo for October 16, 2018, for (a) Puerto Rico, (b) Hispaniola, (c) Jamaica, (d) Cuba, and (e) St. Croix, (f) St. Thomas, and (g) St. John (USVI). 152

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

(37.5 in.) in southeast Puerto Rico. To simulate the daily hydrology, rainfall was evenly divided between the 2 days. Figures 9 and 10 show the rainfall and surface runoff for September 20, respectively. Note that the surface runoff is almost identical to the rainfall, as seen in Figure 11. Nearly 100% of the rainfall was converted to surface runoff because the soils were already saturated the day before Hurricane Maria arrived (September 19), as shown in Figure 12.

## 5. High-resolution ETo products across the Caribbean

GOESWEB provides daily values of ETo for Puerto Rico, the USVI, Hispaniola, Jamaica, and Cuba. As an example, the ETo for each of the islands for October 16, 2018, are presented in Figure 13. In the study by [52], they describe a web-based method for determining irrigation requirements using the GOESWEB ETo maps.

#### 6. Conclusions

The above study demonstrates the operational utility of incorporating spatially continuous, high spatial resolution (1 km) GOES–16-derived solar insolation, using the model described by [24], into the water balance model GOESWEB, to then estimate the complete water budget. In this demonstration, applications of water balance were performed over the US territory of Puerto Rico, a subtropical location that is very sensitive to high rates of ET, relative to various crop types and vegetation characteristics, and that also receives high amounts of rainfall. High rainfall causes significant runoff, for which the GOESWEB water balance model can help identify related to actual rainfall events. Expanding GOESWEB to other island regions would be a future avenue for the research and algorithm development activities described here.

#### Acknowledgements

This work was supported by the NOAA-CREST (grant NA06OAR4810162), National Science Foundation (grant 0313747 and 1832576), and the U.S. Department of Agriculture (Hatch Project H402). Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Oceanic and Atmospheric Administration, NSF, or the USDA. Reference to a commercial product in no way constitutes an endorsement of the product by the authors.

## Conflict of interest

None.

#### Notes/thanks/other declarations

Thanks to Luis Aponte-Bermudez for providing the Matlab® computer code and for reading the daily WRF wind speed data sets and CariCOOS for providing the WRF wind speed, air temperature, and relative humidity data. There are numerous other people that directly or indirectly assisted in this research and to those people we would like to express our appreciation.

## Nomenclature


The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460


## Author details

John R. Mecikalski<sup>1</sup> \* and Eric W. Harmsen<sup>2</sup>

1 University of Alabama in Huntsville, Huntsville, Alabama, United States of America

2 University of Puerto Rico–Mayaguez Campus, Mayaguez, Puerto Rico, United States of America

\*Address all correspondence to: johnm@nsstc.uah.edu

© 2019 The Author(s). Licensee IntechOpen. This chapteris distributed underthe terms oftheCreative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

## References

[1] Tarpley JD, Schneider SR, Bragg JE, Waters MP III. Satellite data set for solar incoming radiation studies. NOAA Technical Memorandum. 1978;96. 36 pp

[2] Diak GR, Gautier C. Improvements to a simple physical model for estimating insolation from GOES data. Journal of Climate and Applied Meteorology. 1983;22:505-508

[3] Tarpley JD. Estimating incident solar radiation at the surface from geostationary satellite data. Journal of Applied Meteorology. 1978;18:1172-1181

[4] Gautier C, Diak GR, Masse S. A simple physical model to estimate incident solar radiation at the surface from GOES satellite data. Journal of Applied Meteorology. 1980;19: 1007-1012

[5] Gautier C, Diak GR, Masse S. An investigation of the effects of spatially averaging satellite brightness measurements on the calculation of insolation. Journal of Climate and Applied Meteorology. 1984;23: 1380-1386

[6] Möser W, Raschke E. Incident solar radiation over Europe from METEOSAT data. Journal of Climate and Applied Meteorology. 1984;23:166-170

[7] Pinker RT, Ewing JA. Modeling surface solar radiation: Model formulation and validation. Journal of Climate and Applied Meteorology. 1985; 24:389-401

[8] Dedieu G, Deschamps PY, Kerr YH. Satellite estimates of solar irradiance at the surface of the earth and of surface albedo using a physical model applied to meteosat data. Journal of Climate and Applied Meteorology. 1987;26:79-87

[9] Darnell WL, Staylor WF, Gupta SK, Denn FM. Estimation of surface

insolation using sun-synchronous satellite data. Journal of Climate. 1988;1: 820-835

[10] Frouin R, Chertock B. A technique for global monitoring of net solar irradiance at the ocean surface. Part I: Model. Journal of Applied Meteorology. 1992;31:1056-1066

[11] Pinker RT, Laszlo I. Modeling surface solar irradiance for satellite applications on global scale. Journal of Applied Meteorology. 1992;31:194-211

[12] Weymouth G, LeMarshall L. An operational system to estimate global solar exposure over the Australian region from satellite observations. Australian Meteorological Magazine. 1999;48:181-195

[13] Schmetz J. Towards a surface radiation climatology. Retrieval of downward irradiance from satellites. Atmospheric Research. 1989;23:287-321

[14] Pinker RT, Frouin R, Li Z. A review of satellite methods to derive surface shortwave irradiance. Remote Sensing of Environment. 1995;51:105-124

[15] Mecikalski JR, Shoemaker WB, Wu Q, Holmes MA, Paech SJ, Sumner DM. A 20-Year high-resolution GOES insolation–evapotranspiration dataset for water resource management over the State of Florida. Journal of Irrigation and Drainage Engineering. 2018;144(9): 04018025

[16] Stewart JB, Watts CJ, Rodriguez JC, De Bruin HAR, van den Berg AR, Garatuza-Payan J. Use of satellite data to estimate radiation and evaporation for northwest Mexico. Agricultural Water Management. 1999;38:181-193

[17] Otkin J, Anderson MC, Mecikalski JR, Diak GR. Validation of GOES-Based The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

insolation estimates using data from the United States Climate Reference Network. Journal of Hydrometeorology. 2005;6:460-475. https://doi.org/10.1175/ JHM440.1

[18] Frouin R, Gautier C, Katsaros KB, Lind J. A comparison of satellite and empirical formula techniques for estimating insolation over the oceans. Journal of Applied Meteorology. 1988; 27:1016-1023

[19] Cosgrove BA, Lohmann D, Mitchell KE, Houser PR, Wood EF, Schaake J, et al. Realtime and retrospective forcing in the North American Land Data Assimilation Systems (NLDAS) project. Journal of Geophysical Research. 2003; 108(D22):8842. DOI: 10.1029/ 2002JD003118

[20] Cosgrove BA, Lohmann D, Mitchell KE, Houser PR, Wood EF, Schaake JC, et al. Land surface model spinup behavior in the North American Land Data Assimilation System (NLDAS). Journal of Geophysical Research. 2003; 108(D22):8845. DOI: 10.1029/ 2002JD003119

[21] Paech SJ, Mecikalski JR, Sumner DM, Pathak CS, Wu Q, Islam S, et al. Satellitebased solarradiation in support of potential and reference evapotranspiration estimates over Florida: A 10-year climatology. Water Resources Research. 2009;45(6):1328-1342

[22] Meng C, Pinker RT, Tarpley DJ, Laszlo I. A satellite approach for estimating regional land surface energy budget for GCIP/GAPP. Journal of Geophysical Research. 2003;108(D22):8861. DOI: 10.1029/ 2002JD003088

[23] Diak GR, Bland WL, Mecikalski JR. A note on first estimates of surface insolation from GOES-8 visible satellite data. Agricultural and Forest Meteorology. 1996;82:219-226

[24] Diak GR. Investigations of improvements to an operational GOES-satellite-data-based insolation system using pyranometer data from the U.S. Climate Reference Network (USCRN). Remote Sensing of Environment. 2017;195:79-95. DOI: 10.1016/j.rse.2017.04.002

[25] Jacobs JM, Myers DA, Anderson MC, Diak GR. GOES surface insolation to estimate wetlands evapotranspiration. Journal of Hydrology. 2002;56:53-65

[26] Jacobs JM, Anderson MC, Friess LC, Diak GR. Solar radiation, longwave radiation and emergent wetland evapotranspiration estimates from satellite data in Florida, USA. Hydrological Sciences Journal. 2004;49: 461-476

[27] Mecikalski JM, Diak GR, Anderson MC, Norman JM. Estimating fluxes on continental scales using remotely sensed data in an atmosphere–land exchange model. Journal of Applied Meteorology. 1999;38:1352-1369

[28] Anderson MC, Kustas WP, Norman JM. Upscaling and downscaling—A regional view of the soil-plantatmosphere continuum. Agronomy Journal. 2003;95:1408-1432

[29] Anderson MC, Norman JM, Mecikalski JR, Torn RD, Kustas WP, Basara JB. A multiscale remote sensing model for disaggregating regional fluxes to micrometeorological scales. Journal of Hydrometeorology. 2004;5:343-363

[30] Diak GR, Anderson MC, Bland WL, Norman JM, Mecikalski JR, Aune RM. Agricultural management decisions aids driven by real-time satellite data. Bulletin of the American Meteorological Society. 1998;79:1345-1355

[31] Anderson MC, Bland WL, Norman JM, Diak GR. Canopy wetness and humidity prediction using satellite and synoptic-scale meteorological

observations. Plant Disease. 2001;85: 1018-1026

[32] Allen RG, Periera LS, Raes D, Smith M. Crop evapotranspiration: Guidelines for computing crop requirements. Irrigation and Drainage. 1998. Paper No. 56. Rome, Italy: FAO

[33] Allen RG, Walter IA, Elliott RL, Howell TA, Itenfisu D, Jensen ME, et al. The ASCE Standardized Reference Evapotranspiration Equation. Reston, Virginia: American Society of Civil Engineers; 2005

[34] Priestley CHB, Taylor RJ. On the assessment of surface heat flux and evaporation using large-scale parameters. Monthly Weather Review. 1972;100:81-92

[35] Hargreaves GH, Samani ZA. Reference crop evapotranspiration from temperature. Applied Engineering in Agriculture - ASAE. 1985;1(2):96-99

[36] Harmsen EW, Mecikalski JR, Cardona-Soto MJ, Gonzalez AR, Vasquez R. Estimating daily evapotranspiration in Puerto Rico using satellite remote sensing. WSEAS Transactions on Environment and Development. 2009;6(5):456-465

[37] Goyal MR, González EA, Chao de Báez C. Temperature versus elevation relationships for Puerto Rico. The Journal of Agriculture of the University of Puerto Rico. 1988;72(3):449-467

[38] Harmsen EW, Goyal MR, Justiniano ST. Estimating evapotranspiration in Puerto Rico. Journal of Agriculture of the University of Puerto Rico. 2002;86 (1–2):35-54

[39] Harmsen EW, Tosado P, Mecikalski JR. Calibration of selected pyranometers and satellite derived solar radiation in Puerto Rico. International Journal of Renewable Energy Technology Research. 2014;5(1):43-54

[40] NDFD. National Weather Service National Digital Forecast Database. 2018. Available from: http://www.wea ther.gov/forecasts/graphical/sectors/ puertorico.php

[41] ASCE. The ASCE Standardized Reference Evapotranspiration Equation. American Society of Civil Engineers; 2005

[42] ATMET. ATMET Technical Note, Number 1, Modifications for the Transition From LEAF-2 to LEAF-3. 2005. Available from: http://www.atme t.com/html/docs/rams/RT1-leaf2-3.pdf

[43] Penman HL. Natural evaporation from open water, bare soil and grass. Proceedings. Royal Society of London. 1948;A(194):120-145

[44] Penman HL. Vegetation and Hydrology. Technical Communication. 1963;53. Commonwealth Bureau of soils, Harpenden, England

[45] Morton FI. Operational estimates of areal evapotranspiration and their significance to the science and practice of hydrology. Journal of Hydrology. 1983;66:1-76

[46] Yunhao C, Xiaobing L, Peijun S. Estimation of regional evapotranspiration over Northwest China by using remotely sensed data. Journal of Geographical Sciences. 2001; 11(2):1997-1999

[47] Monteith JL, Unsworth MH. Principles of Environmental Physics. 3rd ed. Boston, MA: Academic Press; 2007. 440p

[48] Lascano RJ, van Bavel CHM. Explicit and recursive calculation of potential and actual evapotranspiration. Agronomy Journal. 2007;99:585-590

[49] Ortega-Farias S, Poblete C, Zuñiga M. Evaluation of a two-layer model to estimate vine transpiration and soil

The Use of Visible Geostationary Operational Meteorological Satellite Imagery in Mapping… DOI: http://dx.doi.org/10.5772/intechopen.82460

evaporation for vineyards. In: Proceedings of the ASCE World Environmental and Water Resources Congress, 2008. Ahupua'a

[50] Cemek B, Meral R, Apan M, Merdun H. Pedotransfer functions for the estimation of field capacity and permanent wilting point. Pakistan Journal of Biological Sciences. 2004; 7(4):535-541

[51] Huffman RL, Fangmeier DD, Elliot W, Workman SR. Soil and Water Conservation Engineering. 7th ed. Published by ASABE, St. Joseph, MI; 2013

[52] Harmsen EW. TECHNICAL NOTE: A Simple Web-Based Method for Scheduling Irrigation in Puerto Rico. The Journal of Agriculture of the University of Puerto Rico. 2012;96:3-4

## *Edited by Rustam B. Rustamov*

Without a doubt, understanding what we must do to save our home, our planet, and how we are to do it is of the gravest importance for the present generation and the next. Clearly, advances won through space technology and applications of the same to the study of Earth play an excellent and vital role in classifcation and interpretation of the processes taking place on the Earth and in space. Today, space technology helps us understand Earth and how we can support and manage its state, to keep it in working condition under the current circumstances.How can we do this? Obviously, we must use appropriate methods and instruments to collect the information we need. In the meantime, it is necessary to develop systems to analyze and process the data collected.

Published in London, UK © 2019 IntechOpen © Christopher Burns / unsplash