SPI for the period September OctoberNovember 2005

Fig. 17.11. Variation of the SPI for the months from September to November 2005, on a quarterly scale (SPI-3) for the State of Rio Grande do Sul, highlighting the location of the Auto-mated Meteorological Stations.

Baixa

wsiruuiçio Je o rota m 11 dad e pi) ce »correncia de fhlvîs ?rr relaïao a rrecia niitcrica

Baixa

Disttibuiçào de probabilidáde de occrrència de en uvas em rel-acoO 3 media rus tonca

Acima do média histórica

Baixa

Baixa

— Média-Alta wsiruuiçio Je o rota m 11 dad e pi) ce »correncia de

Disttibuiçào de probabilidáde de occrrència de fhlvîs ?rr relaïao a rrecia niitcrica en uvas em rel-acoO 3 media rus tonca

Acima do média histórica

Pioxirrca* s rr-c a

Aba xc da média hstorica

Abaixo da media histórica

Aï reai5S3 hachjrsdis mdicarn i corfi^biidace

A& reflioes hachuradas mdicam aconfistjihdade ca -¡tevasa ivae legfnta na fii|u*a da previsao (vide legenda na figura)

Fig. 17.12. Summary of the climatic prognosis corresponding to the December 2005 to February 2006, and January to March 2006 quarters, prepared by CPTEC/INPE and INMET.

(INPE) periodically carry out a climatic prognosis indicating if the meteorological conditions, especially precipitation, will be above or below the historic average (Figure 17.12). ttis trend is used by several institutions especially those dedicated to agricultural planning.

Figure 17.13 presents the prognosis for precipitation prepared for the State of Rio Grande do Sul by the 8th Meteorological District of INMET, together with the Meteorology Department of the Pelotas Federal University (CPPMet/UFPEL, 2005), for the months ofjanuary, February and March 2006.

For CIIAGRO and INFOSECA in the State of Sao Paulo, climatic prognosis is not carried out, but the INMET/INPE prognosis is used for agricultural purposes and planning. One of the uses is to establish a monthly prognosis of the SPI and its effect on agriculture. Since the climatic likelihood for the coming months was of precipitation below the average the SPI trend was projected as a function of the possibility of this event for the July-August quarter (Table 17.10). Since results indicate the persistency of the meteorological drought at least to the end of September in the regions comprising the states of Paraná, Sao Paulo and Minas Gerais, projecting unfavorable conditions for the sugarcane, coffee and citrus crop, and a delay for summer crops planting.

Fig. 17.13. Prognosis for precipitation in the State of Rio Grande do Sul in the months from January to March 2006. ^e areas represented in white indicate rainfall in the climatological average, yellow below average and blue above-average (source: www.inmet.gov.br/climatologia/cond-cli-ma/bol-dez2005.pdf

Fig. 17.13. Prognosis for precipitation in the State of Rio Grande do Sul in the months from January to March 2006. ^e areas represented in white indicate rainfall in the climatological average, yellow below average and blue above-average (source: www.inmet.gov.br/climatologia/cond-cli-ma/bol-dez2005.pdf

Agrometeorological Aspects of Drought

Several institutions in Brazil try to make quantification and the monitoring of drought from a meteorological and agronomic standpoint. Some examples are the Ceara Meteorological Foundation (FUNCEME), National Meteorology Institute (INMET) and the National Space Research Institute (INPE). Nevertheless, few of these institutions routinely consider the assessment and characteristics of this phenomenon towards agriculture and civil defense, embracing agronomy soil characteristics and crop behavior.

With regards to the SPI, several assessments have been made specifically for the southern region of Brazil, the INMET has tried to compare the behavior of the crops with the SPI values. Figure 17.14 shows the behavior of soybean yield and the SPI index for six months (SPI-6) calculation based on data from the Passo Fundo (soybean producing region) and Santa Maria, for the period ranging from October to March, tte estimation for the index for the October 2005 and March 2006 semester, and the forecast for the next crop, performed by CONAB, are indicat-

Table17.10. Estimated monthly values for the Standardized Precipitation Index (SPI-1) in relation to the prognosis of rainfall

Estimated Monthly Values for the SPI ( SPI-1)

Locality

July

August

September

Ara^atuba -SP

-0.89

-0.72

-0.55

Catanduva -SP

-0.86

-0.69

-0.52

Jaú -SP

-0.84

-0.66

-0.49

Piracicaba -SP

-0.81

-0.63

-0.46

Ribeirao Preto -SP

-0.78

-0.60

-0.43

Sao losé do Rio Preto -SP

-0.75

-0.58

-0.40

Cambará-SP

-1.05

-0.93

-0.81

loaquim Távora -PR

-1.02

-0.9

-0.78

Maringá -PR

-0.99

-0.87

-0.75

Paranavai-PR

-0.96

-0.84

-0.72

Uberaba-MG

-0.69

-0.66

-0.63

ed with distinct colors and highlighted by the oval (source production - CONAB 2005). tte results highlight the importance of analyzing crop yield and rainfall patterns.

Concerning the State of Sao Paulo, the adverse effect of these precipitation anomalies has been assessed for some crops. For example, the assessment of sugarcane yield in the Ribeirao Preto region, demonstrated a good correlation between the SPI values in 9-month scales (Figure 17.15) and sugar yield. Normally the growing period for the crop is from September to May, in which accumulation and the increment of dry matter is directly influenced by the climate and in such a case, tte SPI for May with the 9-month recurrence (SPI-9) adequately reflects the water conditions in this soil for this crop. But for maize in the off-season cropping in the Assis region when planting is performed between January and March, it is observed that the averaged SPI on a monthly scale adequately reflects the water conditions for this crop, ttis relationship is presented on Figure 17.16, and a good relationship between the SPI and the productivity levels can be observed.

Another parameter that is adequately related to the agricultural production is the Palmer Drought Severity Index (PDSI). tte relationship between the average PDSI adap values and maize yield in the State of Sao Paulo is presented in Figure 17.17, indicating the potential of this easily used index, ttese results are quite will correlated to overall maize grain production in the State, and the same figures were observed in the 2005/2006 crop growing season.

Fig. 17.14. Comparison between soybean yield and the values for the SPI on a six-month scale (SPI-6) for the State of Rio Grande do Sul, considering the period from 0ctober/05 to March/06.

Drought Monitoring and Mitigation Center tte State of Sao Paulo, through the Agronomic Institute (IAC) in a partnership with the State Extension Service Agency CATI created the INFOSECA (Drought and Hydrometeorological Adversities Mitigation and Monitoring Center), an operational system that brings immediate reports of the actions and effects of meteorological adversities upon agriculture and proposes ways of monitoring and mitigating the negative impact of these adversities, most notably, drought.

tte work of INFOSECA allows systematically following up on the evolution of drought conditions in the State, proposing mitigating and relief measures, as well as physical and agronomic processes to bypass the problem, ttese processes may include future prognosis of the drought conditions, and is available at the site: www.infoseca.sp.gov.br.

crop yield

crop yield

crop calendar- year

Fig. 17.15. Relationship between the decreasing productivity for sugarcane and the SPI values < the nine-month scale (SPI-9) for the Ribeirao Preto - SP, region.

Crop Yield

1990 1992 1994 1996 1998 2000 2002 2004

Crop Calendar year

Fig. 17.16. Relationship between the yield for the off season maize and the average monthly values for the (SPI-1).

Crop Yield

1990 1992 1994 1996 1998 2000 2002 2004

Crop Calendar year

Fig. 17.16. Relationship between the yield for the off season maize and the average monthly values for the (SPI-1).

tte users have two basic lines of work, in other words, a user may analyze the effect of the drought from a fully meteorological as well as, an agrometeorologi-cal standpoint, tte INFOSECA system has the purpose of processing and making available the agrometeorological information related to drought indices, and communicates agrometeorological warning and outlook of these adversities to the agribusiness, ttis system is based on agrometeorological parameters and relies on a management model and the direct data input via web from the meteorological stations. Furthermore, it has a module to provide information and counseling and real-time consulting via Internet. Meteorological data are collected (mainly, precipitation, maximum and minimum air temperatures) from 130 locations in dif-

1000 3000 5000 7000 9000

1000 3000 5000 7000 9000

Maize Yield -Kg/Ha

Fig. 17.17. Relationship between the yield of summer maize in the State of Sao Paulo and the average values for the Palmer Drought Severity Index (PDSI) values during crop growing sea-son.

ferent regions of the State of Sâo Paulo, that are recorded in to the CIIAGRO system. Data are consisted, assessed and transformed into agronomic parameters and displayed in the form of tables and maps of indices (SPI, Palmer, ETM/ETP and DI) and the agrometeorological indices (CMI, CWS, CWSI, IPER and Crop Development Index). A daily bulletin containing drought prognosis is supplied, tte system was developed using the Sis Plant technology and is based on the HTML, ASP, VbScript and SQL languages. Communication of Web data and database server is performed via ODBC, using the MySQL database. Information is provided at municipal level and consolidated by Administrative Region, Regional Development Offices - EDR/CATI, Water Resource Management Units - UGRH and Regional Research Centers.

tte study allows the analysis of the meteorological conditions and drought through the use of the universally adopted indices and introducing new analysis that take into account soil characteristics, crop évapotranspiration and the relationship between potential évapotranspiration and water availability in the soil and the development of the root system. In this aspect, results referring to the different depths of root systems are also presented, as for crop with superficial root systems and consequently more sensitive water storage such as rice, beans, onions and deeper root systems, such as citrus, coffee and fruit.

Table 17.11 presents the average water stress conditions for the off-season maize crop with a root system at 50cm in the period ranging from March 1st to April 30th, 2006, as well as for the sugarcane crops during the same period, however with a the root system of 1 m deep. It can be noted that for sugarcane, the agrometeorological conditions were not considered critical, due to larger soil volume exploration by the sugar cane rooting system, however, for the maize crop, the situation was highly prejudicial.

Table17.11. Average conditions ofwater stress for the off-season corn crop (Z=50cm) andfor the sugarcane crop (Z=100cm) in the period ranging from March 1, 2006 to April 30, 2006

Locality

Rooting Depth (cm)

ACWDI

Condition

Guariba

(Maize-early stage) 25

0.03

Extreme Severe

Guariba

(Maize- tasseling period stage ) 50

0.33

Not Favorable

Guariba

(Sugar cane-full development stage) 100

0.65

Good

laboticabal

(Maize -early stage) 25

-0.27

Extremely Severe

laboticabal

(Maize -tasseling period) 50

-0.31

Extremely Severe

laboticabal

(Sugar cane-full development stage )100

0.23

Climatic Risk Zoning

One of the most important aspects of agrometeorology is to define the timeframe and location with probability of occurrence of drought and other adverse phenomena for specific crop development stage, or the climatic risk assessment for agriculture exploitation. Specifically considering drought, this assessment of water shortage probability is made by comparing the crop water demand and the water availability in the ecosystem imposed by the rainfall precipitation regime. Crop Water Requirement Index (CWRI), can be defined as:

where:

ETR - actual crop évapotranspiration ; and

ETM - maximum crop évapotranspiration, as defined by

Kc - crop coefficient

ETo - reference crop évapotranspiration

tte studies that sought to quantify the climate-plant relationship and the risks of meteorological adversities are one of the basic tools used in the agricultural financing programs. As examples we can name the PROAGRO at Federal Government level and the FEAP at the State of Sao Paulo Government level.

Figure 17.18 presents the climatic risk zoning for the summer maize crop in the State of Sao Paulo (Brunini et al. 2001) used in the PROAGRO Agricultural Insurance Program.

Fig. 17.18: Probability of water supply during the tasseling period of the maize crop in the State of Sao Paulo. Between the 1st and 10th of October (source : Brunini etal 2001).

One further step was taken by the government of the State of Sao Paulo in this system for the risk characterization, with the introduction of the "Sistema de Aval-iagao de Riscos Climáticos e Monitoramento Agrometeorológico de Culturas" (Climatic Risk Assessment System and Agrometeorological Monitoring of Crops).

In this process, climatic risks related to drought are assessed as well as the probability of addressing the water demand for any crop, be it annual or perennial, tte likelihood of addressing the water requirement is made on the beta distribution (P), that the best represents the agro-system being analyzed, since the ETR/ETM ratio has values between 0 and 1. Furthermore, following up on the evolution of the agrometeorological parameters and behavior is allowed, tte study can be made for all critical phenological phases of the crop, and a subroutine allows that the soil volume for each the crop inferred by the root system is also inferred (climate Risk Evaluation and Crop Agrometeorology System).Information on the probability for meeting crop water requirements for each planting scheduling for each critical phase of the crop is automatically inserted into the CIIAGRO, enabling the online assessment of climatic risks.

Table 17.12 indicates the probability of attending crop water requirements water for the off-season maize crop in the region of Palmital - SP, as well as the risk of occurrence of frost or agricultural drought in the tasseling period.

Even though tables and charts allow the indication or the results of the occurrence of adverse phenomena, and the response of a crop in a given region, they do

Table 17.12. Probability of attending crop water demands during specific phenological phases of the corn crop planted between 1-5 January, and the risk of high or low air temperature

Day/Month

Max-Days

1/1

Sowing

8.03

91.62

0

0

6

6/1

Sowing

0.03

99.96

0

0

6

v^

Fast growing

0.01

99.98

0

0

13

6/2

Fast growing

0.01

99.98

0

0

13

16/2

Tas selling

0.01

99.98

0

0

21

21/2

Tas selling

0.01

99.98

0

0

21

26/2

Tasselling

0.01

99.98

0

0

21

1/3

Tas selling

0.01

99.98

0

0

21

6/3

Tasselling

0.03

99.96

0

0

21

'/4

Ripening

0.01

99.98

0

0

21

6/4

Ripening

0.01

99.98

0

0

21

11/4

Ripening

0.01

99.98

0

0

21

Prob - Probabilityfunction

Prob - Probabilityfunction not provide the spatial visualization of these parameters or their degree of occurrence in different time frames.

In order to make this information more readily understood by the general users and by the decision makers, these data are transformed into agrometeorological maps. Two basic tools were used for this- SURFER and ARG-GIS.

Figure 17.19 shows the water stress conditions for maize crop using the Surfer methodology. Note the differences as a function of the spatial variability and the topography of the state when the different types of soil are included.

On the other hand, with the use of the ARC-GIS, this information is more detailed, enabling the overlapping of other variables. Figure 17.20 presents the same map with the water stress conditions for the maize in the ARC GIS system. In this case, minimum and maximum air temperatures lower than 16°C and higher than 32°C were superimposed on the map indicating restrictive areas due to thermal insufficiency or elevated temperatures, as well as the water supply.

CWSN

Fig. 17.19. Average condition of water stress on the maize crop in the State of Sâo Paulo during the month ofMarch 2006, by the Surfer system.

Fig. 17.19. Average condition of water stress on the maize crop in the State of Sâo Paulo during the month ofMarch 2006, by the Surfer system.

CWSN

AGROCL MAT C COND T ONS FOR MA ZE CROP DEVELOPMENT

CarxlTMAX

Adéquats ow.ooo

Fig. 17.20. Average conditions of water stress on the maize crop in the State of Sao Paulo with overlapping of the areas with minimum air temperature below 14°C, by the ARC-GIS system.

AGROCL MAT C COND T ONS FOR MA ZE CROP DEVELOPMENT

CarxlTMAX

Adéquats ow.ooo

Fig. 17.20. Average conditions of water stress on the maize crop in the State of Sao Paulo with overlapping of the areas with minimum air temperature below 14°C, by the ARC-GIS system.

17.4

Conclusions tte assessment of the aspects presented and discussed enabled the following premises:

Drought is a constant phenomenon in agriculture in Brazil, thus requiring continuous prediction and monitoring to provide valuable mitigating measures.

As a result of the territorial extension, the mitigating measures are not necessarily identical and must take into consideration the cultural aspects of the population the climate regime, and the agricultural exploitation.

tte various indices presented have proven to be adequate for monitoring and mitigating the effects of drought, nevertheless, adjustments are necessary for the use of these indices for each region and crop. For the PDSI, the parameters of the equations should be estimated for each region in Brazil.

Every state should create a Drought Monitoring and Mitigation Center, subordinated to the State's Agricultural Secretariat. It should be the responsibility of the INMET, in association with state agencies, to propose norms and to define standards and policy for the monitoring and mitigation of drought on a regional and nationwide scale.

It should be understood that the drought phenomenon cannot be assessed and interpreted by only one field of expertise, but rather by a set of specialists and institutions. Furthermore, it is extremely important that researchers and specialist in the areas of agronomy, agrometeorology, meteorology, civil defense, agro extension service, and others, should be involved in the study of the drought phenomenon.

References

Abramowitz M Stegun IA (1965). Handbook of mathematical function. Dover, p.1046.

Anunciado YMT, Fortes LTG (2006) Monitoramento da precipitado no rio grande do sul com vistas á previsao para o veráo 2005/2006. Brasilia: INMET, 2006. 25p. (INMET/cdp/rt-001-2006)

Assad E, Macedo MA, Zullo lunior J, Pinto HS, Brunini O (2003) Avalia^áo de métodos geoestatiticos na espacializa^áo de indices agrometeorológicos para definir riscos climáticos, pesquisaagropecuáriabrasileira, Brasilia 38:161-171.

Blain GC (2005) Avalia^áo e adapta^áo do índice de severidade de seca de palmer (pdsi) e do índice padronizado de precipitado (spi) as condi^óes climáticas do estado de sao paulo, campiñas: IAC, 2005. 120f. disserta^áo (mestrado em agricultura tropical e subtropical) - Instituto Agronómico, Campinas - sp, 2005.

Blain GC, Brunini O (2006) Quantifica^áo da seca agrícola pelo índice padronizado de evapotranspira^áo real (iper) no estado de Sao Paulo. Bragantia 65:517-525.

Blain GC, Brunini O (2005) Avalia^áo e adapta^áo do índice de severidade de seca de Palmer (PDSI) e do Índice padronizado de precipitado (spi) as condi^óes climáticas do estado de Sao Paulo. Bragantia 64:695-705

Brunini O (2005) Centro de monitoramento e mitigado de seca e adversidades hidrometeorológicas. disponível em: www.infoseca.sp.gov.br. acesso em julho 2005.

Brunini O (2005). Infoseca. Campinas: Instituto Agronómico (fólder)

BruniniO (1983). Tese de modelo agrometeorológico para estimativa da produtividade de cultura de soja ciclo precoce. In: Congresso Brasileiro de Agrometeorologia, 3, 1983, Campinas, SP. Anais, Campinas: SBA, 1983. pp 245-253.

Brunini O, Abramides PLG, Brunini APC, Blain GC, Pedro lunior MJ, Camargo MBP, Tremocoldi WA, Pezzopane Ir, Rolim GS (2005) O infoseca no monitoramento e mitiga^áo de secae adversidades hidrometeorológicas. In: Congresso Brasileiro de Agroinformática, 5, Londrina. Anais, Londrina-Paraná: Sociedade Brasileirade Agroinformática, 1:1-15.

Brunini O, Abramides PLG, Rolim GS, Brunini APC, Ernandes ES, Blain GC (2006) O infoseca no alerta e monitoramento de risco de incendio. In: Congresso Brasileiro de Biometeorologia, 4, Ribeiráo Preto, SP. Anais, Ribeiráo Preto: Instituto de Zootecnia/Sociedade Brasileira de Biometeorologia, 1:98-104. CD-ROM.

Brunini O, Blain GC, Brunini APC, Santos RL, Brigante RS, Almeida EL (2002) Monitoramento das condi^óes de seca no estado de sao paulo utilizando o índice padronizado de precipitado, in: congressobrasileiro de meteorología, Foz do Igua^ú. Proceedings, Sociedade Brasileira de Meteorología, 1:1135-1139. CD-ROM.

Brunini O, Grohmann F, Benincasa M (1977). Varia^áo estacional do armazenamento e das perdas de água em unidades de solo do estado de sao paulo. Científica, laboticabal, V. Especial 5:119-123

Brunini O, Grohmann F, Santos IM (1981) Balando hídrico em condi^óes de campo para duas cultivares de arroz sob duas densidades de plantío. Revista Brasileira de Ciencia do Solo, 5:

Brunini O, Pinto HS, Zullo lunior J, Barbano MT, Camargo MBP (2000) Drought and desertification and preparedness in Brazil: ^e example of Sao Paulo State. In: Wilhite D, Sivakumar MVK (eds.) Early Warning Systems for Drought Preparadness and Drought Monitoring, Genebra, Sui^a: Organizado MeteorológicaMundial, pp. 89-103.

Brunini O, Sawazaki E, Duarte AP, Tsunechiro A, Kanthack RAD, Blain GC, Brunini APC, Zullo I (2003) Milho safrinha: zoneamento e riscos climáticos no estado de Sao Paulo. In: Congresso Brasileiro de Agrometeorologia, 13, Santa Maria-RS. Anais, Sociedade Brasileira de Agrometeorologia, 1:629-630.

Brunini O, Zullo lunior J, Pinto HS, Assad E, Sawasaki E, Paterniani MEZ (2001) Riscos climáticos para a cultura do milho no estado de sao paulo. Revista Brasileira de Agrometeorologia, Santa Maria, 9:519-526.

Camargo MBP, Brunini O, Miranda MAC (1986) Modelo agrometeorológico para estimativa da produtividade para cultura de soja no estado de sao paulo. Bragantia 45 :231-286.

Camargo MBP, Hubbard KG (1999). Drought sensitivity indices for a sorghum crop. I Prod Agrie 12:312-316

Camargo MBP, Hubbard KG, Flores-Mendoza F (1994) Test of a soil water assessment model for a sorghum crop under different irrigation treatments. Bragantia 53:95-105

Companhia nacional de abastecimentos (conab). produtividade de soja, safras de 1990/1991 a 2004/2005: séries históricas, disponível em: http://www.conab.gov.br/ download/safra/ sojaseriehist.xls). acesso em nov. de 2005.

Fortes LTG, Anunciado YMT, Lucio PS (2006) Comparado entre métodos alternativos para informar anomalías de precipitado. Brasilia: INMET, Trabalho submetido ao XIV Congresso Brasileiro de Meteorología.

Infoseca (2005) Disponível em http://www.infoseca.sp.gov.br. acesso em dez. de 2005.

Instituto Nacional de Meteorología (2005) Oitavo distrito meteorológicxo (INMET/8 DISME). Universidade Federal de Pelotas. Centro de Pesquisas e Previsóes Meteorológicas. (UFPEL. CPPMET). Prognóstico climático para o estado do rio grande do sul. Disponível em: http://www.inmet.gov.br/climatologia/cond_clima/bol_dez2005.pdf. acesso em dez. de 2005.

Instituto Nacional de Pesquisas Espaciais. Centro de Previsáo de Tempo e Estudos Climáticos (INPE, CPTEC). Instituto Nacional de Meteorología (INMET) (2005). Prognóstico climático de consenso (CPTEC/INPE e INMET para o trimestre janeiro, fevereiro e mar^o de 2006).

Disponível em: http://www.inmet.gov.br prev_clima_tempo/prognostico_climatico_ trimestral/pc/pc0512_a.pdf.acesso em dez. de 2005.

Internacional Research Institute for Climate and Society (IRI) (2005) Global spi analyses, disponivel em: http://www.ingrid. ldeo.columbia.edu/maproom/.global/.precipitation/spi. html ). acesso em dez. de 2005.

Karl T (1986) ^e sensitivity of the palmer drought severity index and palmer's z-index to their calibration coefficients including potencial évapotranspiration. I Clim Appl Meteorol 25: 77-86.

Mckee TB, Doesken NJ, Kleist I (1993) ^e relationship of drought frequency and duration to time scale. In: Proceedings of the Conference on Applied Climatology, Boston: AMS, pp 179-184.

Normáis Climatológicas 1961-1990 (1992). Brasilia: Departamento nacional de meteorología, 84 pP.

Palmer WC (1968) Keeping track of crop moisture conditions, nationwide: the new crop moisture index. Weatherwise 21:156-161

Palmer WC (1965) Meteorological drought. US Weather Bureau Res. No. 45, 58 pp

Santos FAS, Anjos RI (2001) Utilizaçâo do indice de precipitaçâo padronizada (spi) no monitoramento da seca no estado de pernambuco. In: Congresso Brasileiro de Agrometeorologia, Fortaleza, Anais, SBA/FUNCEME, pp 121-122.

Secretaria de Agricultura e Abastecimento do Estado de Sâo Paulo (2006) Sistema de avaliaçâo e monitoramento de riscos climáticos e agrometeorológicos. Sâo Paulo, CD-ROM.

^ornthwaite CW, Mather IR (1955) ^e Water Balance. Publications in Climatology, Vol. 3, No.,1, 104 pp.

CHAPTER 18

Disaster Survival Guides Collection

Disaster Survival Guides Collection

This is a set of 3 guides all about surviving disasters. Within this set you will learn the following subjects: Earthquakes, Evacuations, Survivor Family and Tsunami.

Get My Free Ebook


Post a comment